Merge remote-tracking branch 'origin/master' into norm_norm_norm

README.md

@@ -23,6 +23,18 @@ I'm fortunate to be able to dedicate significant time and money of my own suppor
 
 ## What's New
 
+### Jan 14, 2022
+* Version 0.5.4 w/ release to be pushed to pypi. It's been a while since last pypi update and riskier changes will be merged to main branch soon....
+* Add ConvNeXT models /w weights from official impl (https://github.com/facebookresearch/ConvNeXt), a few perf tweaks, compatible with timm features
+* Tried training a few small (~1.8-3M param) / mobile optimized models, a few are good so far, more on the way...
+  * `mnasnet_small` - 65.6 top-1
+  * `mobilenetv2_050` - 65.9
+  * `lcnet_100/075/050` - 72.1 / 68.8 / 63.1
+  * `semnasnet_075` - 73
+  * `fbnetv3_b/d/g` - 79.1 / 79.7 / 82.0
+* TinyNet models added by [rsomani95](https://github.com/rsomani95)
+* LCNet added via MobileNetV3 architecture
+
 ### Nov 22, 2021
 * A number of updated weights anew new model defs
   * `eca_halonext26ts` - 79.5 @ 256
@@ -255,10 +267,12 @@ All model architecture families include variants with pretrained weights. There
 A full version of the list below with source links can be found in the [documentation](https://rwightman.github.io/pytorch-image-models/models/).
 
 * Aggregating Nested Transformers - https://arxiv.org/abs/2105.12723
+* BEiT - https://arxiv.org/abs/2106.08254
 * Big Transfer ResNetV2 (BiT) - https://arxiv.org/abs/1912.11370
 * Bottleneck Transformers - https://arxiv.org/abs/2101.11605
 * CaiT (Class-Attention in Image Transformers) - https://arxiv.org/abs/2103.17239
 * CoaT (Co-Scale Conv-Attentional Image Transformers) - https://arxiv.org/abs/2104.06399
+* ConvNeXt - https://arxiv.org/abs/2201.03545
 * ConViT (Soft Convolutional Inductive Biases Vision Transformers)- https://arxiv.org/abs/2103.10697
 * CspNet (Cross-Stage Partial Networks) - https://arxiv.org/abs/1911.11929
 * DeiT (Vision Transformer) - https://arxiv.org/abs/2012.12877
@@ -276,11 +290,11 @@ A full version of the list below with source links can be found in the [document
     * MNASNet B1, A1 (Squeeze-Excite), and Small - https://arxiv.org/abs/1807.11626
     * MobileNet-V2 - https://arxiv.org/abs/1801.04381
     * Single-Path NAS - https://arxiv.org/abs/1904.02877
+    * TinyNet - https://arxiv.org/abs/2010.14819
 * GhostNet - https://arxiv.org/abs/1911.11907
 * gMLP - https://arxiv.org/abs/2105.08050
 * GPU-Efficient Networks - https://arxiv.org/abs/2006.14090
 * Halo Nets - https://arxiv.org/abs/2103.12731
-* HardCoRe-NAS - https://arxiv.org/abs/2102.11646
 * HRNet - https://arxiv.org/abs/1908.07919
 * Inception-V3 - https://arxiv.org/abs/1512.00567
 * Inception-ResNet-V2 and Inception-V4 - https://arxiv.org/abs/1602.07261
@@ -288,7 +302,11 @@ A full version of the list below with source links can be found in the [document
 * LeViT (Vision Transformer in ConvNet's Clothing) - https://arxiv.org/abs/2104.01136
 * MLP-Mixer - https://arxiv.org/abs/2105.01601
 * MobileNet-V3 (MBConvNet w/ Efficient Head) - https://arxiv.org/abs/1905.02244
+  * FBNet-V3 - https://arxiv.org/abs/2006.02049
+  * HardCoRe-NAS - https://arxiv.org/abs/2102.11646
+  * LCNet - https://arxiv.org/abs/2109.15099
 * NASNet-A - https://arxiv.org/abs/1707.07012
+* NesT - https://arxiv.org/abs/2105.12723
 * NFNet-F - https://arxiv.org/abs/2102.06171
 * NF-RegNet / NF-ResNet - https://arxiv.org/abs/2101.08692
 * PNasNet - https://arxiv.org/abs/1712.00559
@@ -314,6 +332,7 @@ A full version of the list below with source links can be found in the [document
 * Transformer-iN-Transformer (TNT) - https://arxiv.org/abs/2103.00112
 * TResNet - https://arxiv.org/abs/2003.13630
 * Twins (Spatial Attention in Vision Transformers) - https://arxiv.org/pdf/2104.13840.pdf
+* Visformer - https://arxiv.org/abs/2104.12533
 * Vision Transformer - https://arxiv.org/abs/2010.11929
 * VovNet V2 and V1 - https://arxiv.org/abs/1911.06667
 * Xception - https://arxiv.org/abs/1610.02357

benchmark.py

@@ -21,7 +21,7 @@ from functools import partial
 from timm.models import create_model, is_model, list_models
 from timm.optim import create_optimizer_v2
 from timm.data import resolve_data_config
-from timm.utils import AverageMeter, setup_default_logging
+from timm.utils import setup_default_logging, set_jit_fuser
 
 
 has_apex = False
@@ -95,7 +95,8 @@ parser.add_argument('--precision', default='float32', type=str,
                     help='Numeric precision. One of (amp, float32, float16, bfloat16, tf32)')
 parser.add_argument('--torchscript', dest='torchscript', action='store_true',
                     help='convert model torchscript for inference')
-
+parser.add_argument('--fuser', default='', type=str,
+                    help="Select jit fuser. One of ('', 'te', 'old', 'nvfuser')")
 
 
 # train optimizer parameters
@@ -186,7 +187,7 @@ def profile_fvcore(model, input_size=(3, 224, 224), batch_size=1, detailed=False
 class BenchmarkRunner:
     def __init__(
             self, model_name, detail=False, device='cuda', torchscript=False, precision='float32',
-            num_warm_iter=10, num_bench_iter=50, use_train_size=False, **kwargs):
+            fuser='', num_warm_iter=10, num_bench_iter=50, use_train_size=False, **kwargs):
         self.model_name = model_name
         self.detail = detail
         self.device = device
@@ -194,6 +195,8 @@ class BenchmarkRunner:
         self.channels_last = kwargs.pop('channels_last', False)
         self.amp_autocast = torch.cuda.amp.autocast if self.use_amp else suppress
 
+        if fuser:
+            set_jit_fuser(fuser)
         self.model = create_model(
             model_name,
             num_classes=kwargs.pop('num_classes', None),
@@ -311,10 +314,7 @@ class TrainBenchmarkRunner(BenchmarkRunner):
         super().__init__(model_name=model_name, device=device, torchscript=torchscript, **kwargs)
         self.model.train()
 
-        if kwargs.pop('smoothing', 0) > 0:
-            self.loss = nn.CrossEntropyLoss().to(self.device)
-        else:
-            self.loss = nn.CrossEntropyLoss().to(self.device)
+        self.loss = nn.CrossEntropyLoss().to(self.device)
         self.target_shape = tuple()
 
         self.optimizer = create_optimizer_v2(
@@ -477,6 +477,7 @@ def decay_batch_exp(batch_size, factor=0.5, divisor=16):
 def _try_run(model_name, bench_fn, initial_batch_size, bench_kwargs):
     batch_size = initial_batch_size
     results = dict()
+    error_str = 'Unknown'
     while batch_size >= 1:
         torch.cuda.empty_cache()
         try:
@@ -484,13 +485,13 @@ def _try_run(model_name, bench_fn, initial_batch_size, bench_kwargs):
             results = bench.run()
             return results
         except RuntimeError as e:
-            e_str = str(e)
-            print(e_str)
-            if 'channels_last' in e_str:
-                print(f'Error: {model_name} not supported in channels_last, skipping.')
+            error_str = str(e)
+            if 'channels_last' in error_str:
+                _logger.error(f'{model_name} not supported in channels_last, skipping.')
                 break
-            print(f'Error: "{e_str}" while running benchmark. Reducing batch size to {batch_size} for retry.')
+            _logger.warning(f'"{error_str}" while running benchmark. Reducing batch size to {batch_size} for retry.')
         batch_size = decay_batch_exp(batch_size)
+    results['error'] = error_str
     return results
 
 
@@ -532,13 +533,14 @@ def benchmark(args):
     model_results = OrderedDict(model=model)
     for prefix, bench_fn in zip(prefixes, bench_fns):
         run_results = _try_run(model, bench_fn, initial_batch_size=batch_size, bench_kwargs=bench_kwargs)
-        if prefix:
+        if prefix and 'error' not in run_results:
             run_results = {'_'.join([prefix, k]): v for k, v in run_results.items()}
         model_results.update(run_results)
-    param_count = model_results.pop('infer_param_count', model_results.pop('train_param_count', 0))
-    model_results.setdefault('param_count', param_count)
-    model_results.pop('train_param_count', 0)
-    return model_results if model_results['param_count'] else dict()
+    if 'error' not in model_results:
+        param_count = model_results.pop('infer_param_count', model_results.pop('train_param_count', 0))
+        model_results.setdefault('param_count', param_count)
+        model_results.pop('train_param_count', 0)
+    return model_results
 
 
 def main():
@@ -582,13 +584,15 @@ def main():
             sort_key = 'train_samples_per_sec'
         elif 'profile' in args.bench:
             sort_key = 'infer_gmacs'
+        results = filter(lambda x: sort_key in x, results)
         results = sorted(results, key=lambda x: x[sort_key], reverse=True)
         if len(results):
             write_results(results_file, results)
     else:
         results = benchmark(args)
-    json_str = json.dumps(results, indent=4)
-    print(json_str)
+
+    # output results in JSON to stdout w/ delimiter for runner script
+    print(f'--result\n{json.dumps(results, indent=4)}')
 
 
 def write_results(results_file, results):

docs/training_hparam_examples.md

@@ -3,12 +3,12 @@
 ## EfficientNet-B2 with RandAugment - 80.4 top-1, 95.1 top-5
 These params are for dual Titan RTX cards with NVIDIA Apex installed:
 
-`./distributed_train.sh 2 /imagenet/ --model efficientnet_b2 -b 128 --sched step --epochs 450 --decay-epochs 2.4 --decay-rate .97 --opt rmsproptf --opt-eps .001 -j 8 --warmup-lr 1e-6 --weight-decay 1e-5 --drop 0.3 --drop-connect 0.2 --model-ema --model-ema-decay 0.9999 --aa rand-m9-mstd0.5 --remode pixel --reprob 0.2 --amp --lr .016`
+`./distributed_train.sh 2 /imagenet/ --model efficientnet_b2 -b 128 --sched step --epochs 450 --decay-epochs 2.4 --decay-rate .97 --opt rmsproptf --opt-eps .001 -j 8 --warmup-lr 1e-6 --weight-decay 1e-5 --drop 0.3 --drop-path 0.2 --model-ema --model-ema-decay 0.9999 --aa rand-m9-mstd0.5 --remode pixel --reprob 0.2 --amp --lr .016`
 
 ## MixNet-XL with RandAugment - 80.5 top-1, 94.9 top-5
 This params are for dual Titan RTX cards with NVIDIA Apex installed:
 
-`./distributed_train.sh 2 /imagenet/ --model mixnet_xl -b 128 --sched step --epochs 450 --decay-epochs 2.4 --decay-rate .969 --opt rmsproptf --opt-eps .001 -j 8 --warmup-lr 1e-6 --weight-decay 1e-5 --drop 0.3 --drop-connect 0.2 --model-ema --model-ema-decay 0.9999 --aa rand-m9-mstd0.5 --remode pixel --reprob 0.3 --amp --lr .016 --dist-bn reduce`
+`./distributed_train.sh 2 /imagenet/ --model mixnet_xl -b 128 --sched step --epochs 450 --decay-epochs 2.4 --decay-rate .969 --opt rmsproptf --opt-eps .001 -j 8 --warmup-lr 1e-6 --weight-decay 1e-5 --drop 0.3 --drop-path 0.2 --model-ema --model-ema-decay 0.9999 --aa rand-m9-mstd0.5 --remode pixel --reprob 0.3 --amp --lr .016 --dist-bn reduce`
 
 ## SE-ResNeXt-26-D and SE-ResNeXt-26-T
 These hparams (or similar) work well for a wide range of ResNet architecture, generally a good idea to increase the epoch # as the model size increases... ie approx 180-200 for ResNe(X)t50, and 220+ for larger. Increase batch size and LR proportionally for better GPUs or with AMP enabled. These params were for 2 1080Ti cards:
@@ -21,7 +21,7 @@ The training of this model started with the same command line as EfficientNet-B2
 ## EfficientNet-B0 with RandAugment - 77.7 top-1, 95.3 top-5
 [Michael Klachko](https://github.com/michaelklachko) achieved these results with the command line for B2 adapted for larger batch size, with the recommended B0 dropout rate of 0.2.
 
-`./distributed_train.sh 2 /imagenet/ --model efficientnet_b0 -b 384 --sched step --epochs 450 --decay-epochs 2.4 --decay-rate .97 --opt rmsproptf --opt-eps .001 -j 8 --warmup-lr 1e-6 --weight-decay 1e-5 --drop 0.2 --drop-connect 0.2 --model-ema --model-ema-decay 0.9999 --aa rand-m9-mstd0.5 --remode pixel --reprob 0.2 --amp --lr .048`
+`./distributed_train.sh 2 /imagenet/ --model efficientnet_b0 -b 384 --sched step --epochs 450 --decay-epochs 2.4 --decay-rate .97 --opt rmsproptf --opt-eps .001 -j 8 --warmup-lr 1e-6 --weight-decay 1e-5 --drop 0.2 --drop-path 0.2 --model-ema --model-ema-decay 0.9999 --aa rand-m9-mstd0.5 --remode pixel --reprob 0.2 --amp --lr .048`
 
 ## ResNet50 with JSD loss and RandAugment (clean + 2x RA augs) - 79.04 top-1, 94.39 top-5
 
@@ -32,11 +32,11 @@ Trained on two older 1080Ti cards, this took a while. Only slightly, non statist
 ## EfficientNet-ES (EdgeTPU-Small) with RandAugment - 78.066 top-1, 93.926 top-5
 Trained by [Andrew Lavin](https://github.com/andravin) with 8 V100 cards. Model EMA was not used, final checkpoint is the average of 8 best checkpoints during training.
 
-`./distributed_train.sh 8 /imagenet --model efficientnet_es -b 128 --sched step --epochs 450 --decay-epochs 2.4 --decay-rate .97 --opt rmsproptf --opt-eps .001 -j 8 --warmup-lr 1e-6 --weight-decay 1e-5 --drop 0.2 --drop-connect 0.2  --aa rand-m9-mstd0.5 --remode pixel --reprob 0.2 --amp --lr .064`
+`./distributed_train.sh 8 /imagenet --model efficientnet_es -b 128 --sched step --epochs 450 --decay-epochs 2.4 --decay-rate .97 --opt rmsproptf --opt-eps .001 -j 8 --warmup-lr 1e-6 --weight-decay 1e-5 --drop 0.2 --drop-path 0.2  --aa rand-m9-mstd0.5 --remode pixel --reprob 0.2 --amp --lr .064`
 
 ## MobileNetV3-Large-100 - 75.766 top-1, 92,542 top-5
 
-`./distributed_train.sh 2 /imagenet/ --model mobilenetv3_large_100 -b 512 --sched step --epochs 600 --decay-epochs 2.4 --decay-rate .973 --opt rmsproptf --opt-eps .001 -j 7 --warmup-lr 1e-6 --weight-decay 1e-5 --drop 0.2 --drop-connect 0.2 --model-ema --model-ema-decay 0.9999 --aa rand-m9-mstd0.5 --remode pixel --reprob 0.2 --amp --lr .064 --lr-noise 0.42 0.9`
+`./distributed_train.sh 2 /imagenet/ --model mobilenetv3_large_100 -b 512 --sched step --epochs 600 --decay-epochs 2.4 --decay-rate .973 --opt rmsproptf --opt-eps .001 -j 7 --warmup-lr 1e-6 --weight-decay 1e-5 --drop 0.2 --drop-path 0.2 --model-ema --model-ema-decay 0.9999 --aa rand-m9-mstd0.5 --remode pixel --reprob 0.2 --amp --lr .064 --lr-noise 0.42 0.9`
 
 
 ## ResNeXt-50 32x4d w/ RandAugment - 79.762 top-1, 94.60 top-5

results/benchmark-infer-amp-nchw-pt110-cu113-rtx3090.csv

@@ -0,0 +1,704 @@
+model,infer_samples_per_sec,infer_step_time,infer_batch_size,infer_img_size,infer_gmacs,infer_macts,param_count
+tinynet_e,38522.1,6.633,256,106,0.03,0.69,2.04
+mobilenetv3_small_050,32214.84,7.935,256,224,0.03,0.92,1.59
+lcnet_035,31434.42,8.133,256,224,0.03,1.04,1.64
+lcnet_050,27990.74,9.135,256,224,0.05,1.26,1.88
+tf_mobilenetv3_small_minimal_100,24613.65,10.389,256,224,0.06,1.41,2.04
+mobilenetv3_small_075,23970.62,10.668,256,224,0.05,1.3,2.04
+tinynet_d,22226.54,11.506,256,152,0.05,1.42,2.34
+mobilenetv3_small_100,20961.66,12.202,256,224,0.06,1.42,2.54
+tf_mobilenetv3_small_075,20346.43,12.57,256,224,0.05,1.3,2.04
+tf_mobilenetv3_small_100,18514.72,13.814,256,224,0.06,1.42,2.54
+lcnet_075,18250.57,14.016,256,224,0.1,1.99,2.36
+levit_128s,17514.74,14.605,256,224,0.31,1.88,7.78
+mnasnet_small,16174.2,15.817,256,224,0.07,2.16,2.03
+mobilenetv2_035,16065.18,15.924,256,224,0.07,2.86,1.68
+regnetx_002,15056.58,16.991,256,224,0.2,2.16,2.68
+ghostnet_050,14915.33,17.152,256,224,0.05,1.77,2.59
+lcnet_100,14745.93,17.35,256,224,0.16,2.52,2.95
+regnety_002,13600.49,18.812,256,224,0.2,2.17,3.16
+mobilenetv2_050,12921.59,19.8,256,224,0.1,3.64,1.97
+mnasnet_050,12170.95,21.022,256,224,0.11,3.07,2.22
+levit_128,12112.09,21.125,256,224,0.41,2.71,9.21
+semnasnet_050,11653.92,21.956,256,224,0.11,3.44,2.08
+tinynet_c,11441.66,22.362,256,184,0.11,2.87,2.46
+levit_192,10512.15,24.341,256,224,0.66,3.2,10.95
+gernet_s,10252.23,24.96,256,224,0.75,2.65,8.17
+mixer_s32_224,9991.39,25.611,256,224,1.0,2.28,19.1
+mobilenetv3_large_075,9848.53,25.983,256,224,0.16,4.0,3.99
+ese_vovnet19b_slim_dw,9567.57,26.746,256,224,0.4,5.28,1.9
+lcnet_150,9561.66,26.763,256,224,0.34,3.79,4.5
+vit_small_patch32_224,9406.05,27.206,256,224,1.15,2.5,22.88
+regnetx_004,9291.08,27.542,256,224,0.4,3.14,5.16
+tf_mobilenetv3_large_minimal_100,9242.84,27.684,256,224,0.22,4.4,3.92
+tf_mobilenetv3_large_075,8812.22,29.037,256,224,0.16,4.0,3.99
+mobilenetv3_rw,8698.97,29.417,256,224,0.23,4.41,5.48
+vit_tiny_r_s16_p8_224,8636.97,29.629,256,224,0.44,2.06,6.34
+mobilenetv3_large_100,8617.52,29.696,256,224,0.23,4.41,5.48
+mobilenetv3_large_100_miil,8613.51,29.709,256,224,0.23,4.41,5.48
+ssl_resnet18,8520.11,30.036,256,224,1.82,2.48,11.69
+gluon_resnet18_v1b,8514.08,30.057,256,224,1.82,2.48,11.69
+resnet18,8481.72,30.171,256,224,1.82,2.48,11.69
+swsl_resnet18,8383.36,30.526,256,224,1.82,2.48,11.69
+ghostnet_100,8302.41,30.823,256,224,0.15,3.55,5.18
+mobilenetv2_075,8284.54,30.89,256,224,0.22,5.86,2.64
+mnasnet_075,8193.74,31.233,256,224,0.23,4.77,3.17
+seresnet18,7935.21,32.25,256,224,1.82,2.49,11.78
+levit_256,7830.25,32.682,256,224,1.13,4.23,18.89
+tf_mobilenetv3_large_100,7782.64,32.881,256,224,0.23,4.41,5.48
+legacy_seresnet18,7750.12,33.02,256,224,1.82,2.49,11.78
+regnetx_006,7679.18,33.326,256,224,0.61,3.98,6.2
+semnasnet_075,7673.81,33.35,256,224,0.23,5.54,2.91
+mobilenetv2_100,7580.19,33.761,256,224,0.31,6.68,3.5
+regnety_004,7520.0,34.031,256,224,0.41,3.89,4.34
+tinynet_b,7462.13,34.294,256,188,0.21,4.44,3.73
+hardcorenas_a,7310.63,35.006,256,224,0.23,4.38,5.26
+hardcorenas_b,7243.65,35.33,256,224,0.26,5.09,5.18
+mnasnet_100,7211.43,35.488,256,224,0.33,5.46,4.38
+mnasnet_b1,7197.57,35.556,256,224,0.33,5.46,4.38
+resnet18d,7139.78,35.845,256,224,2.06,3.29,11.71
+mnasnet_a1,6803.23,37.618,256,224,0.32,6.23,3.89
+ghostnet_130,6799.96,37.636,256,224,0.24,4.6,7.36
+semnasnet_100,6798.37,37.645,256,224,0.32,6.23,3.89
+hardcorenas_c,6791.21,37.684,256,224,0.28,5.01,5.52
+spnasnet_100,6742.99,37.955,256,224,0.35,6.03,4.42
+regnety_006,6741.6,37.962,256,224,0.61,4.33,6.06
+ese_vovnet19b_slim,6727.44,38.04,256,224,1.69,3.52,3.17
+hardcorenas_d,6616.74,38.678,256,224,0.3,4.93,7.5
+tf_efficientnetv2_b0,6404.53,39.959,256,224,0.73,4.77,7.14
+regnetx_008,6126.62,41.774,256,224,0.81,5.15,7.26
+efficientnet_lite0,5820.53,43.971,256,224,0.4,6.74,4.65
+dla46_c,5815.5,44.006,256,224,0.58,4.5,1.3
+resnetblur18,5731.2,44.657,256,224,2.34,3.39,11.69
+mobilenetv2_110d,5728.31,44.679,256,224,0.45,8.71,4.52
+tinynet_a,5715.04,44.781,256,192,0.35,5.41,6.19
+rexnet_100,5705.81,44.855,256,224,0.41,7.44,4.8
+rexnetr_100,5702.38,44.882,256,224,0.43,7.72,4.88
+hardcorenas_f,5633.4,45.432,256,224,0.35,5.57,8.2
+regnety_008,5562.41,46.011,256,224,0.81,5.25,6.26
+hardcorenas_e,5407.35,47.331,256,224,0.35,5.65,8.07
+fbnetc_100,5335.41,47.97,256,224,0.4,6.51,5.57
+skresnet18,5314.32,48.161,256,224,1.82,3.24,11.96
+tf_efficientnet_lite0,5277.93,48.492,256,224,0.4,6.74,4.65
+efficientnet_b0,5118.94,50.0,256,224,0.4,6.75,5.29
+mobilenetv2_140,5111.46,50.073,256,224,0.6,9.57,6.11
+ese_vovnet19b_dw,5070.62,50.475,256,224,1.34,8.25,6.54
+mnasnet_140,5006.18,51.125,256,224,0.6,7.71,7.12
+efficientnet_b1_pruned,5001.01,51.178,256,240,0.4,6.21,6.33
+resnet34,4971.16,51.486,256,224,3.67,3.74,21.8
+gluon_resnet34_v1b,4968.4,51.515,256,224,3.67,3.74,21.8
+tv_resnet34,4947.3,51.734,256,224,3.67,3.74,21.8
+visformer_tiny,4922.23,51.997,256,224,1.27,5.72,10.32
+hrnet_w18_small,4823.01,53.067,256,224,1.61,5.72,13.19
+semnasnet_140,4792.63,53.404,256,224,0.6,8.87,6.11
+levit_384,4732.88,54.078,256,224,2.36,6.26,39.13
+tf_efficientnet_b0_ns,4689.2,54.583,256,224,0.4,6.75,5.29
+tf_efficientnet_b0_ap,4677.78,54.715,256,224,0.4,6.75,5.29
+tf_efficientnet_b0,4677.33,54.72,256,224,0.4,6.75,5.29
+seresnet34,4638.93,55.174,256,224,3.67,3.74,21.96
+selecsls42b,4636.17,55.206,256,224,2.98,4.62,32.46
+dla46x_c,4631.52,55.26,256,224,0.54,5.66,1.07
+selecsls42,4604.3,55.589,256,224,2.94,4.62,30.35
+deit_tiny_patch16_224,4580.12,55.882,256,224,1.26,5.97,5.72
+vit_tiny_patch16_224,4575.68,55.937,256,224,1.26,5.97,5.72
+gernet_m,4537.32,56.41,256,224,3.02,5.24,21.14
+deit_tiny_distilled_patch16_224,4513.87,56.702,256,224,1.27,6.01,5.91
+nf_regnet_b0,4512.73,56.717,256,256,0.64,5.58,8.76
+legacy_seresnet34,4498.05,56.902,256,224,3.67,3.74,21.96
+resnet34d,4471.13,57.245,256,224,3.91,4.54,21.82
+rexnetr_130,4425.1,57.84,256,224,0.68,9.81,7.61
+pit_ti_distilled_224,4357.18,58.742,256,224,0.71,6.23,5.1
+pit_ti_224,4345.5,58.9,256,224,0.7,6.19,4.85
+mixnet_s,4322.59,59.212,256,224,0.25,6.25,4.13
+dla60x_c,4316.36,59.296,256,224,0.59,6.01,1.32
+tf_efficientnetv2_b1,4297.02,59.563,256,240,1.21,7.34,8.14
+rexnet_130,4232.23,60.476,256,224,0.68,9.71,7.56
+xcit_nano_12_p16_224_dist,4222.67,60.614,256,224,0.56,4.17,3.05
+xcit_nano_12_p16_224,4207.85,60.827,256,224,0.56,4.17,3.05
+resnet26,4180.52,61.224,256,224,2.36,7.35,16.0
+resmlp_12_distilled_224,4174.74,61.31,256,224,3.01,5.5,15.35
+resmlp_12_224,4169.39,61.389,256,224,3.01,5.5,15.35
+mobilenetv2_120d,4149.08,61.689,256,224,0.69,11.97,5.83
+vit_base_patch32_224_sam,4128.66,61.994,256,224,4.41,5.01,88.22
+vit_base_patch32_224,4107.27,62.317,256,224,4.41,5.01,88.22
+tf_mixnet_s,4070.89,62.873,256,224,0.25,6.25,4.13
+repvgg_b0,4034.35,63.443,256,224,3.41,6.15,15.82
+mixer_b32_224,3992.12,64.115,256,224,3.24,6.29,60.29
+selecsls60,3976.19,64.371,256,224,3.59,5.52,30.67
+selecsls60b,3972.94,64.425,256,224,3.63,5.52,32.77
+rexnetr_150,3836.3,66.719,256,224,0.89,11.13,9.78
+efficientnet_lite1,3836.28,66.719,256,240,0.62,10.14,5.42
+resnet26d,3798.91,67.377,256,224,2.6,8.15,16.01
+dla34,3769.64,67.899,256,224,3.07,5.02,15.74
+rexnet_150,3681.94,69.516,256,224,0.9,11.21,9.73
+ecaresnet50d_pruned,3648.06,70.162,256,224,2.53,6.43,19.94
+nf_resnet26,3644.75,70.227,256,224,2.41,7.35,16.0
+tf_efficientnet_lite1,3554.59,72.006,256,240,0.62,10.14,5.42
+pit_xs_224,3526.31,72.586,256,224,1.4,7.71,10.62
+regnetx_016,3513.85,72.844,256,224,1.62,7.93,9.19
+pit_xs_distilled_224,3500.42,73.123,256,224,1.41,7.76,11.0
+fbnetv3_d,3391.58,75.47,256,256,0.68,11.1,10.31
+efficientnet_es_pruned,3370.79,75.934,256,224,1.81,8.73,5.44
+fbnetv3_b,3370.57,75.939,256,256,0.55,9.1,8.6
+efficientnet_es,3367.33,76.013,256,224,1.81,8.73,5.44
+efficientnet_b2_pruned,3361.3,76.15,256,260,0.73,9.13,8.31
+efficientnet_cc_b0_8e,3277.21,78.103,256,224,0.42,9.42,24.01
+mixer_s16_224,3266.14,78.369,256,224,3.79,5.97,18.53
+efficientnet_cc_b0_4e,3265.3,78.389,256,224,0.41,9.42,13.31
+tf_efficientnet_es,3259.95,78.516,256,224,1.81,8.73,5.44
+tf_efficientnetv2_b2,3255.37,78.626,256,260,1.72,9.84,10.1
+resnest14d,3217.37,79.556,256,224,2.76,7.33,10.61
+nf_seresnet26,3202.97,79.914,256,224,2.41,7.36,17.4
+nf_ecaresnet26,3189.01,80.265,256,224,2.41,7.36,16.0
+gernet_l,3181.93,80.443,256,256,4.57,8.0,31.08
+regnety_016,3102.97,82.489,256,224,1.63,8.04,11.2
+tf_efficientnet_cc_b0_8e,3093.44,82.743,256,224,0.42,9.42,24.01
+tf_efficientnet_cc_b0_4e,3085.55,82.954,256,224,0.41,9.42,13.31
+mixnet_m,3056.44,83.746,256,224,0.36,8.19,5.01
+skresnet34,3030.56,84.459,256,224,3.67,5.13,22.28
+resnext26ts,3023.16,84.668,256,256,2.43,10.52,10.3
+repvgg_a2,3004.59,85.192,256,224,5.7,6.26,28.21
+vit_tiny_r_s16_p8_384,3002.92,85.238,256,384,1.34,6.49,6.36
+legacy_seresnext26_32x4d,3001.06,85.291,256,224,2.49,9.39,16.79
+vit_small_patch32_384,2997.9,85.382,256,384,3.45,8.25,22.92
+xcit_tiny_12_p16_224_dist,2978.68,85.929,256,224,1.24,6.29,6.72
+xcit_tiny_12_p16_224,2968.51,86.225,256,224,1.24,6.29,6.72
+resnet26t,2964.54,86.343,256,256,3.35,10.52,16.01
+gmixer_12_224,2952.25,86.702,256,224,2.67,7.26,12.7
+seresnext26ts,2943.83,86.951,256,256,2.43,10.52,10.39
+eca_resnext26ts,2943.35,86.964,256,256,2.43,10.52,10.3
+efficientnet_lite2,2941.13,87.03,256,260,0.89,12.9,6.09
+tf_efficientnet_b1_ap,2928.7,87.399,256,240,0.71,10.88,7.79
+tf_efficientnet_b1,2924.32,87.53,256,240,0.71,10.88,7.79
+tf_efficientnet_b1_ns,2924.08,87.537,256,240,0.71,10.88,7.79
+tf_mixnet_m,2922.35,87.587,256,224,0.36,8.19,5.01
+gcresnext26ts,2878.37,88.928,256,256,2.43,10.53,10.48
+efficientnet_b1,2842.35,90.055,256,256,0.77,12.22,7.79
+ecaresnet101d_pruned,2827.84,90.516,256,224,3.48,7.69,24.88
+seresnext26tn_32x4d,2809.03,91.123,256,224,2.7,10.09,16.81
+ecaresnext50t_32x4d,2808.18,91.151,256,224,2.7,10.09,15.41
+seresnext26t_32x4d,2807.76,91.164,256,224,2.7,10.09,16.81
+ecaresnext26t_32x4d,2799.26,91.441,256,224,2.7,10.09,15.41
+seresnext26d_32x4d,2790.46,91.729,256,224,2.73,10.19,16.81
+ecaresnetlight,2759.56,92.757,256,224,4.11,8.42,30.16
+rexnetr_200,2747.9,93.151,256,224,1.59,15.11,16.52
+nf_regnet_b2,2743.38,93.303,256,272,1.22,9.27,14.31
+tf_efficientnet_lite2,2742.42,93.334,256,260,0.89,12.9,6.09
+nf_regnet_b1,2740.87,93.389,256,288,1.02,9.2,10.22
+crossvit_tiny_240,2729.62,93.771,256,240,1.57,9.08,7.01
+resnetv2_50,2688.33,95.214,256,224,4.11,11.11,25.55
+rexnet_200,2686.43,95.282,256,224,1.56,14.91,16.37
+crossvit_9_240,2684.97,95.332,256,240,1.85,9.52,8.55
+eca_botnext26ts_256,2675.06,95.687,256,256,2.46,11.6,10.59
+vgg11,2663.05,96.12,256,224,7.61,7.44,132.86
+botnet26t_256,2660.83,96.197,256,256,3.32,11.98,12.49
+tresnet_m,2657.08,96.334,256,224,5.74,7.31,31.39
+eca_halonext26ts,2619.52,97.716,256,256,2.44,11.46,10.76
+halonet26t,2612.17,97.992,256,256,3.19,11.69,12.48
+crossvit_9_dagger_240,2602.86,98.339,256,240,1.99,9.97,8.78
+efficientnet_b3_pruned,2602.84,98.342,256,300,1.04,11.86,9.86
+resnet50,2602.03,98.373,256,224,4.11,11.11,25.56
+tv_resnet50,2597.64,98.539,256,224,4.11,11.11,25.56
+gluon_resnet50_v1b,2593.37,98.702,256,224,4.11,11.11,25.56
+ssl_resnet50,2588.68,98.881,256,224,4.11,11.11,25.56
+swsl_resnet50,2588.5,98.887,256,224,4.11,11.11,25.56
+convit_tiny,2557.96,100.066,256,224,1.26,7.94,5.71
+vovnet39a,2552.48,100.283,256,224,7.09,6.73,22.6
+resnetv2_50t,2536.73,100.905,256,224,4.32,11.82,25.57
+resnetv2_50d,2523.31,101.442,256,224,4.35,11.92,25.57
+resnet32ts,2520.72,101.547,256,256,4.63,11.58,17.96
+resnet33ts,2487.04,102.921,256,256,4.76,11.66,19.68
+bat_resnext26ts,2486.58,102.938,256,256,2.53,12.51,10.73
+vit_small_resnet26d_224,2484.66,103.019,256,224,5.07,11.12,63.61
+ese_vovnet39b,2483.84,103.055,256,224,7.09,6.74,24.57
+cspresnet50,2476.24,103.369,256,256,4.54,11.5,21.62
+eca_vovnet39b,2474.83,103.43,256,224,7.09,6.74,22.6
+hrnet_w18_small_v2,2473.49,103.485,256,224,2.62,9.65,15.6
+gluon_resnet50_v1c,2461.53,103.989,256,224,4.35,11.92,25.58
+cspresnext50,2459.33,104.08,256,224,3.1,12.14,20.57
+resnet50t,2452.68,104.364,256,224,4.32,11.82,25.57
+gluon_resnet50_v1d,2447.81,104.572,256,224,4.35,11.92,25.58
+resnet50d,2444.17,104.728,256,224,4.35,11.92,25.58
+dpn68b,2437.01,105.034,256,224,2.35,10.47,12.61
+legacy_seresnet50,2425.37,105.54,256,224,3.88,10.6,28.09
+gmlp_ti16_224,2422.83,105.65,256,224,1.34,7.55,5.87
+seresnet33ts,2418.32,105.847,256,256,4.76,11.66,19.78
+eca_resnet33ts,2414.21,106.028,256,256,4.76,11.66,19.68
+dpn68,2412.98,106.079,256,224,2.35,10.47,12.61
+selecsls84,2409.11,106.251,256,224,5.9,7.57,50.95
+vgg11_bn,2392.35,106.996,256,224,7.62,7.44,132.87
+mixnet_l,2360.8,108.426,256,224,0.58,10.84,7.33
+gcresnet33ts,2355.59,108.666,256,256,4.76,11.68,19.88
+lambda_resnet26t,2353.44,108.765,256,256,3.02,11.87,10.96
+pit_s_224,2338.74,109.449,256,224,2.88,11.56,23.46
+dla60,2331.19,109.802,256,224,4.26,10.16,22.04
+cspresnet50w,2330.53,109.833,256,256,5.04,12.19,28.12
+seresnet50,2324.39,110.125,256,224,4.11,11.13,28.09
+resnest26d,2321.38,110.267,256,224,3.64,9.97,17.07
+pit_s_distilled_224,2320.78,110.295,256,224,2.9,11.64,24.04
+cspresnet50d,2313.66,110.634,256,256,4.86,12.55,21.64
+deit_small_patch16_224,2312.17,110.706,256,224,4.61,11.95,22.05
+vit_small_patch16_224,2301.45,111.22,256,224,4.61,11.95,22.05
+deit_small_distilled_patch16_224,2273.38,112.595,256,224,4.63,12.02,22.44
+tf_efficientnet_b2_ap,2266.39,112.943,256,260,1.02,13.83,9.11
+tf_efficientnet_b2,2265.35,112.996,256,260,1.02,13.83,9.11
+tf_efficientnet_b2_ns,2264.55,113.035,256,260,1.02,13.83,9.11
+tf_mixnet_l,2263.46,113.087,256,224,0.58,10.84,7.33
+densenet121,2251.23,113.702,256,224,2.87,6.9,7.98
+res2net50_48w_2s,2244.59,114.041,256,224,4.18,11.72,25.29
+tv_densenet121,2243.68,114.087,256,224,2.87,6.9,7.98
+resnetaa50d,2224.41,115.074,256,224,5.39,12.44,25.58
+seresnet50t,2204.01,116.141,256,224,4.32,11.83,28.1
+resnetblur50,2197.45,116.486,256,224,5.16,12.02,25.56
+haloregnetz_b,2191.27,116.814,256,224,1.97,11.94,11.68
+ecaresnet50d,2187.04,117.042,256,224,4.35,11.93,25.58
+resnetrs50,2155.43,118.757,256,224,4.48,12.14,35.69
+densenet121d,2152.18,118.936,256,224,3.11,7.7,8.0
+gluon_resnet50_v1s,2129.0,120.233,256,224,5.47,13.52,25.68
+visformer_small,2128.36,120.269,256,224,4.88,11.43,40.22
+resmlp_24_224,2126.97,120.348,256,224,5.96,10.91,30.02
+efficientnet_b2a,2125.57,120.427,256,288,1.12,16.2,9.11
+resmlp_24_distilled_224,2124.32,120.497,256,224,5.96,10.91,30.02
+efficientnet_b2,2121.96,120.632,256,288,1.12,16.2,9.11
+regnetx_032,2121.9,120.635,256,224,3.2,11.37,15.3
+adv_inception_v3,2116.52,120.939,256,299,5.73,8.97,23.83
+gluon_inception_v3,2110.53,121.285,256,299,5.73,8.97,23.83
+inception_v3,2106.47,121.518,256,299,5.73,8.97,23.83
+tf_inception_v3,2102.15,121.765,256,299,5.73,8.97,23.83
+vovnet57a,2101.83,121.786,256,224,8.95,7.52,36.64
+resnetblur50d,2086.08,122.707,256,224,5.4,12.82,25.58
+efficientnet_em,2080.04,123.062,256,240,3.04,14.34,6.9
+efficientnet_cc_b1_8e,2060.35,124.238,256,240,0.75,15.44,39.72
+cspresnext50_iabn,2046.67,125.068,256,256,4.02,15.86,20.57
+densenetblur121d,2045.46,125.142,256,224,3.11,7.9,8.0
+tf_efficientnet_em,2031.54,125.998,256,240,3.04,14.34,6.9
+vit_base_resnet26d_224,2020.03,126.719,256,224,6.97,13.16,101.4
+ese_vovnet57b,2017.2,126.896,256,224,8.95,7.52,38.61
+ssl_resnext50_32x4d,2016.48,126.942,256,224,4.26,14.4,25.03
+swsl_resnext50_32x4d,2014.9,127.042,256,224,4.26,14.4,25.03
+seresnetaa50d,2014.43,127.071,256,224,5.4,12.46,28.11
+gluon_resnext50_32x4d,2013.28,127.144,256,224,4.26,14.4,25.03
+tv_resnext50_32x4d,2010.77,127.303,256,224,4.26,14.4,25.03
+resnext50_32x4d,2010.4,127.325,256,224,4.26,14.4,25.03
+tf_efficientnet_cc_b1_8e,1970.45,129.905,256,240,0.75,15.44,39.72
+cspdarknet53_iabn,1958.85,130.675,256,256,6.53,16.81,27.64
+dla60x,1947.14,131.46,256,224,3.54,13.8,17.35
+regnetx_040,1940.97,131.881,256,224,3.99,12.2,22.12
+skresnet50,1940.94,131.882,256,224,4.11,12.5,25.8
+nf_seresnet50,1937.18,132.139,256,224,4.21,11.13,28.09
+nf_ecaresnet50,1931.32,132.54,256,224,4.21,11.13,25.56
+res2net50_26w_4s,1927.67,132.79,256,224,4.28,12.61,25.7
+tf_efficientnetv2_b3,1921.68,133.203,256,300,3.04,15.74,14.36
+resnext50d_32x4d,1917.13,133.521,256,224,4.5,15.2,25.05
+regnety_040,1911.7,133.9,256,224,4.0,12.29,20.65
+gcresnet50t,1897.82,134.88,256,256,5.42,14.67,25.9
+efficientnetv2_rw_t,1893.97,135.153,256,288,3.19,16.42,13.65
+sehalonet33ts,1881.49,136.051,256,256,3.55,14.7,13.69
+lambda_resnet26rpt_256,1854.82,138.007,256,256,3.16,11.87,10.99
+skresnet50d,1849.16,138.429,256,224,4.36,13.31,25.82
+gcresnext50ts,1842.39,138.938,256,256,3.75,15.46,15.67
+seresnext50_32x4d,1842.11,138.958,256,224,4.26,14.42,27.56
+gluon_seresnext50_32x4d,1836.43,139.388,256,224,4.26,14.42,27.56
+dla60_res2net,1834.89,139.503,256,224,4.15,12.34,20.85
+legacy_seresnext50_32x4d,1833.54,139.608,256,224,4.26,14.42,27.56
+resnest50d_1s4x24d,1830.82,139.816,256,224,4.43,13.57,25.68
+repvgg_b1g4,1821.65,140.52,256,224,8.15,10.64,39.97
+darknet53,1820.66,140.594,256,256,9.31,12.39,41.61
+densenet169,1814.76,141.051,256,224,3.4,7.3,14.15
+gc_efficientnetv2_rw_t,1812.58,141.223,256,288,3.2,16.45,13.68
+res2net50_14w_8s,1809.09,141.495,256,224,4.21,13.28,25.06
+coat_lite_tiny,1803.47,141.935,256,224,1.6,11.65,5.72
+res2next50,1798.23,142.351,256,224,4.2,13.71,24.67
+cspdarknet53,1759.27,145.502,256,256,6.57,16.81,27.64
+ecaresnet26t,1753.56,145.977,256,320,5.24,16.44,16.01
+dla60_res2next,1751.16,146.174,256,224,3.49,13.17,17.03
+efficientnet_lite3,1747.68,146.468,256,300,1.65,21.85,8.2
+nf_regnet_b3,1737.85,147.296,256,320,2.05,14.61,18.59
+vgg13,1734.82,147.554,256,224,11.31,12.25,133.05
+regnetz_b16,1724.03,148.477,256,288,2.39,16.43,9.72
+coat_lite_mini,1717.28,149.059,256,224,2.0,12.25,11.01
+mixnet_xl,1685.34,151.886,256,224,0.93,14.57,11.9
+vit_small_r26_s32_224,1673.77,152.936,256,224,3.56,9.85,36.43
+sebotnet33ts_256,1634.05,156.655,256,256,3.89,17.46,13.7
+resnetv2_101,1630.31,157.012,256,224,7.83,16.23,44.54
+tf_efficientnet_lite3,1628.65,157.171,256,300,1.65,21.85,8.2
+convnext_tiny,1599.8,160.006,256,224,4.47,13.44,28.59
+tv_resnet101,1599.73,160.015,256,224,7.83,16.23,44.55
+gluon_resnet101_v1b,1599.19,160.07,256,224,7.83,16.23,44.55
+convnext_tiny_hnf,1598.61,160.125,256,224,4.47,13.44,28.59
+resnet101,1594.14,160.576,256,224,7.83,16.23,44.55
+xcit_tiny_24_p16_224_dist,1587.39,161.257,256,224,2.34,11.82,12.12
+xcit_tiny_24_p16_224,1582.27,161.779,256,224,2.34,11.82,12.12
+repvgg_b1,1574.77,162.553,256,224,13.16,10.64,57.42
+resnetv2_101d,1569.58,163.088,256,224,8.07,17.04,44.56
+xcit_small_12_p16_224_dist,1568.96,163.153,256,224,4.82,12.58,26.25
+xcit_small_12_p16_224,1566.49,163.411,256,224,4.82,12.58,26.25
+vit_base_resnet50d_224,1557.57,164.346,256,224,8.73,16.92,110.97
+resnest50d,1550.54,165.09,256,224,5.4,14.36,27.48
+gluon_resnet101_v1c,1550.28,165.12,256,224,8.08,17.04,44.57
+lambda_resnet50ts,1550.05,165.145,256,256,5.07,17.48,21.54
+vgg13_bn,1545.09,165.675,256,224,11.33,12.25,133.05
+gluon_resnet101_v1d,1542.32,165.972,256,224,8.08,17.04,44.57
+dla102,1542.08,165.995,256,224,7.19,14.18,33.27
+twins_svt_small,1526.51,167.691,256,224,2.94,13.75,24.06
+wide_resnet50_2,1515.12,168.952,256,224,11.43,14.4,68.88
+gmixer_24_224,1503.65,170.24,256,224,5.28,14.45,24.72
+resnetv2_50x1_bit_distilled,1496.53,171.051,256,224,4.23,11.11,25.55
+regnetx_080,1491.93,171.578,256,224,8.02,14.06,39.57
+xcit_nano_12_p16_384_dist,1489.5,171.858,256,384,1.64,12.15,3.05
+crossvit_small_240,1472.6,173.826,256,240,5.63,18.17,26.86
+res2net50_26w_6s,1468.77,174.282,256,224,6.33,15.28,37.05
+legacy_seresnet101,1466.72,174.527,256,224,7.61,15.74,49.33
+halonet50ts,1459.14,175.434,256,256,5.3,19.2,22.73
+resnetaa101d,1449.52,176.598,256,224,9.12,17.56,44.57
+fbnetv3_g,1435.53,178.32,256,288,1.77,21.09,16.62
+seresnet101,1425.31,179.597,256,224,7.84,16.27,49.33
+resmlp_36_224,1423.79,179.791,256,224,8.91,16.33,44.69
+resmlp_36_distilled_224,1422.98,179.892,256,224,8.91,16.33,44.69
+regnetx_064,1422.54,179.948,256,224,6.49,16.37,26.21
+resnetv2_50d_gn,1422.29,179.98,256,224,4.38,11.92,25.57
+densenet201,1416.58,180.704,256,224,4.34,7.85,20.01
+vit_large_patch32_224,1414.43,180.979,256,224,15.39,13.3,306.54
+gluon_resnet101_v1s,1407.4,181.883,256,224,9.19,18.64,44.67
+vgg16,1399.33,182.931,256,224,15.47,13.56,138.36
+lamhalobotnet50ts_256,1392.5,183.829,256,256,5.02,18.44,22.57
+resnetblur101d,1390.53,184.088,256,224,9.12,17.94,44.57
+nf_resnet50,1387.13,184.542,256,288,6.88,18.37,25.56
+nf_resnet101,1387.03,184.555,256,224,8.01,16.23,44.55
+ecaresnet101d,1375.85,186.054,256,224,8.08,17.07,44.57
+vit_base_r26_s32_224,1374.54,186.232,256,224,6.81,12.36,101.38
+regnety_032,1353.73,189.095,256,288,5.29,18.61,19.44
+crossvit_15_240,1351.08,189.465,256,240,5.81,19.77,27.53
+tresnet_l,1350.23,189.586,256,224,10.88,11.9,55.99
+gmlp_s16_224,1343.26,190.569,256,224,4.42,15.1,19.42
+tf_efficientnet_b3,1341.98,190.75,256,300,1.87,23.83,12.23
+tf_efficientnet_b3_ns,1340.42,190.973,256,300,1.87,23.83,12.23
+tf_efficientnet_b3_ap,1339.36,191.124,256,300,1.87,23.83,12.23
+resnet51q,1338.83,191.2,256,288,8.07,20.94,35.7
+hrnet_w18,1335.83,191.629,256,224,4.32,16.31,21.3
+vit_base_patch32_384,1328.33,192.713,256,384,13.06,16.5,88.3
+resnet50_gn,1327.88,192.777,256,224,4.14,11.11,25.56
+mixer_l32_224,1318.62,194.131,256,224,11.27,19.86,206.94
+xception,1316.52,194.441,256,299,8.4,35.83,22.86
+crossvit_15_dagger_240,1308.42,195.641,256,240,6.13,20.43,28.21
+dla102x,1302.91,196.468,256,224,5.89,19.42,26.31
+efficientnet_b3a,1302.53,196.528,256,320,2.01,26.52,12.23
+efficientnet_b3,1301.42,196.696,256,320,2.01,26.52,12.23
+botnet50ts_256,1288.18,198.716,256,256,5.54,22.23,22.74
+cait_xxs24_224,1287.06,198.889,256,224,2.53,20.29,11.96
+mixer_b16_224_miil,1283.42,199.455,256,224,12.62,14.53,59.88
+mixer_b16_224,1283.22,199.487,256,224,12.62,14.53,59.88
+skresnext50_32x4d,1276.26,200.574,256,224,4.5,17.18,27.48
+regnety_064,1272.95,201.095,256,224,6.39,16.41,30.58
+swsl_resnext101_32x4d,1266.33,202.147,256,224,8.01,21.23,44.18
+ssl_resnext101_32x4d,1265.67,202.252,256,224,8.01,21.23,44.18
+gluon_resnext101_32x4d,1265.25,202.319,256,224,8.01,21.23,44.18
+resnext101_32x4d,1264.74,202.401,256,224,8.01,21.23,44.18
+vgg16_bn,1261.15,202.977,256,224,15.5,13.56,138.37
+repvgg_b2g4,1251.52,204.539,256,224,12.63,12.9,61.76
+halo2botnet50ts_256,1248.54,205.028,256,256,5.02,21.78,22.64
+swin_tiny_patch4_window7_224,1248.01,205.115,256,224,4.51,17.06,28.29
+twins_pcpvt_small,1232.89,207.63,256,224,3.83,18.08,24.11
+regnety_080,1232.57,207.684,256,224,8.0,17.97,39.18
+resnest50d_4s2x40d,1216.47,210.433,256,224,4.4,17.94,30.42
+resnet61q,1209.86,211.583,256,288,9.87,21.52,36.85
+nf_seresnet101,1194.71,214.265,256,224,8.02,16.27,49.33
+ese_vovnet99b_iabn,1193.74,214.438,256,224,16.49,11.27,63.2
+nf_ecaresnet101,1193.13,214.55,256,224,8.01,16.27,44.55
+res2net50_26w_8s,1192.34,214.692,256,224,8.37,17.95,48.4
+eca_nfnet_l0,1189.26,215.246,256,288,7.12,17.29,24.14
+res2net101_26w_4s,1189.12,215.272,256,224,8.1,18.45,45.21
+nfnet_l0,1185.85,215.867,256,288,7.13,17.29,35.07
+dpn92,1179.85,216.965,256,224,6.54,18.21,37.67
+vit_tiny_patch16_384,1177.29,217.437,256,384,4.7,25.39,5.79
+convit_small,1173.71,218.096,256,224,5.76,17.87,27.78
+ese_vovnet99b,1158.79,220.908,256,224,16.51,11.27,63.2
+vgg19,1152.83,222.048,256,224,19.63,14.86,143.67
+seresnext101_32x4d,1152.16,222.18,256,224,8.02,21.26,48.96
+resnetv2_50d_evob,1150.91,222.42,256,224,4.33,11.92,25.59
+gluon_seresnext101_32x4d,1150.75,222.451,256,224,8.02,21.26,48.96
+legacy_seresnext101_32x4d,1150.68,222.464,256,224,8.02,21.26,48.96
+ese_vovnet39b_evos,1144.38,223.689,256,224,7.07,6.74,24.58
+hrnet_w32,1140.5,224.45,256,224,8.97,22.02,41.23
+resnetv2_152,1132.73,225.989,256,224,11.55,22.56,60.19
+xcit_nano_12_p8_224_dist,1129.67,226.602,256,224,2.16,15.71,3.05
+xcit_nano_12_p8_224,1124.18,227.71,256,224,2.16,15.71,3.05
+hrnet_w30,1121.05,228.345,256,224,8.15,21.21,37.71
+tv_resnet152,1117.27,229.117,256,224,11.56,22.56,60.19
+resnet152,1114.93,229.598,256,224,11.56,22.56,60.19
+gluon_resnet152_v1b,1112.41,230.118,256,224,11.56,22.56,60.19
+ecaresnet50t,1109.44,230.735,256,320,8.82,24.13,25.57
+regnetz_c16,1104.72,231.72,256,320,3.92,25.88,13.46
+resnetv2_152d,1101.83,232.328,256,224,11.8,23.36,60.2
+gluon_resnet152_v1c,1092.12,234.395,256,224,11.8,23.36,60.21
+repvgg_b2,1089.82,234.89,256,224,20.45,12.9,89.02
+gluon_resnet152_v1d,1082.98,236.373,256,224,11.8,23.36,60.21
+vgg19_bn,1079.7,237.091,256,224,19.66,14.86,143.68
+densenet161,1075.45,238.026,256,224,7.79,11.06,28.68
+xception41,1067.64,239.769,256,299,9.28,39.86,26.97
+mixnet_xxl,1054.35,242.792,256,224,2.04,23.43,23.96
+inception_v4,1054.27,242.81,256,299,12.28,15.09,42.68
+vit_small_resnet50d_s16_224,1053.44,243.001,256,224,13.48,24.82,57.53
+convmixer_1024_20_ks9_p14,1045.16,244.926,256,224,5.55,5.51,24.38
+dla169,1043.58,245.295,256,224,11.6,20.2,53.39
+nfnet_f0s,1043.56,245.302,256,256,12.62,18.05,71.49
+xcit_tiny_12_p16_384_dist,1043.38,245.342,256,384,3.64,18.26,6.72
+convnext_small,1038.48,246.501,256,224,8.7,21.56,50.22
+nfnet_f0,1038.01,246.613,256,256,12.62,18.05,71.49
+regnetx_120,1028.4,248.918,256,224,12.13,21.37,46.11
+nest_tiny,1021.17,250.679,256,224,5.83,25.48,17.06
+gluon_resnet152_v1s,1020.83,250.763,256,224,12.92,24.96,60.32
+coat_lite_small,1016.77,251.763,256,224,3.96,22.09,19.84
+legacy_seresnet152,1013.05,252.688,256,224,11.33,22.08,66.82
+repvgg_b3g4,1008.62,253.799,256,224,17.89,15.1,83.83
+jx_nest_tiny,1008.04,253.946,256,224,5.83,25.48,17.06
+vit_base_patch16_224_miil,1007.66,254.043,256,224,17.58,23.9,86.54
+crossvit_18_240,995.59,257.12,256,240,9.05,26.26,43.27
+seresnet152,990.73,258.382,256,224,11.57,22.61,66.82
+resnetv2_50d_evos,987.09,259.336,256,224,4.33,11.92,25.59
+vit_base_patch16_224_sam,982.02,260.673,256,224,17.58,23.9,86.57
+tresnet_xl,981.28,260.87,256,224,15.17,15.34,78.44
+regnety_120,981.1,260.919,256,224,12.14,21.38,51.82
+deit_base_patch16_224,980.91,260.97,256,224,17.58,23.9,86.57
+vit_base_patch16_224,978.17,261.701,256,224,17.58,23.9,86.57
+deit_base_distilled_patch16_224,975.08,262.528,256,224,17.68,24.05,87.34
+crossvit_18_dagger_240,969.31,264.09,256,240,9.5,27.03,44.27
+efficientnet_el,926.41,276.325,256,300,8.0,30.7,10.59
+efficientnet_el_pruned,925.7,276.536,256,300,8.0,30.7,10.59
+tf_efficientnet_el,907.79,281.986,256,300,8.0,30.7,10.59
+dm_nfnet_f0,904.97,282.867,256,256,12.62,18.05,71.49
+beit_base_patch16_224,904.15,283.123,256,224,17.58,23.9,86.53
+twins_pcpvt_base,902.77,283.561,256,224,6.68,25.25,43.83
+dla102x2,892.56,286.801,256,224,9.34,29.91,41.28
+efficientnetv2_s,884.22,289.51,256,384,8.44,35.77,21.46
+twins_svt_base,881.96,290.252,256,224,8.59,26.33,56.07
+wide_resnet101_2,875.33,292.449,256,224,22.8,21.23,126.89
+tf_efficientnetv2_s_in21ft1k,864.32,296.172,256,384,8.44,35.77,21.46
+cait_xxs36_224,863.8,296.351,256,224,3.77,30.34,17.3
+tf_efficientnetv2_s,863.76,296.365,256,384,8.44,35.77,21.46
+resnetrs101,852.6,300.247,256,288,13.56,28.53,63.62
+repvgg_b3,844.81,303.013,256,224,29.16,15.1,123.09
+efficientnetv2_rw_s,841.56,304.183,256,384,8.72,38.03,23.94
+dpn98,838.93,305.139,256,224,11.73,25.2,61.57
+pit_b_distilled_224,837.59,305.628,256,224,12.5,33.07,74.79
+pit_b_224,836.04,306.193,256,224,12.42,32.94,73.76
+regnetx_160,833.81,307.013,256,224,15.99,25.52,54.28
+xcit_small_24_p16_224_dist,833.22,307.225,256,224,9.1,23.64,47.67
+xcit_small_24_p16_224,831.56,307.841,256,224,9.1,23.64,47.67
+inception_resnet_v2,830.74,308.147,256,299,13.18,25.06,55.84
+ens_adv_inception_resnet_v2,830.17,308.358,256,299,13.18,25.06,55.84
+regnetz_d8,827.41,309.386,256,320,6.19,37.08,23.37
+nf_regnet_b4,822.49,311.237,256,384,4.7,28.61,30.21
+swin_small_patch4_window7_224,808.44,316.649,256,224,8.77,27.47,49.61
+efficientnet_lite4,804.5,318.199,256,380,4.04,45.66,13.01
+gluon_resnext101_64x4d,802.95,318.813,256,224,15.52,31.21,83.46
+resnext101_64x4d,801.6,319.349,256,224,15.52,31.21,83.46
+resnet200,795.01,321.996,256,224,15.07,32.19,64.67
+xcit_tiny_12_p8_224_dist,792.9,322.851,256,224,4.81,23.6,6.71
+xcit_tiny_12_p8_224,791.66,323.355,256,224,4.81,23.6,6.71
+gluon_xception65,782.06,327.33,256,299,13.96,52.48,39.92
+convnext_base_in22ft1k,778.5,328.824,256,224,15.38,28.75,88.59
+convnext_base,777.8,329.119,256,224,15.38,28.75,88.59
+xception65,776.49,329.678,256,299,13.96,52.48,39.92
+ssl_resnext101_32x8d,773.95,330.757,256,224,16.48,31.21,88.79
+swsl_resnext101_32x8d,773.24,331.062,256,224,16.48,31.21,88.79
+resnext101_32x8d,772.71,331.288,256,224,16.48,31.21,88.79
+ig_resnext101_32x8d,772.07,331.562,256,224,16.48,31.21,88.79
+resnet101d,769.32,332.747,256,320,16.48,34.77,44.57
+hrnet_w40,767.4,333.582,256,224,12.75,25.29,57.56
+tf_efficientnet_lite4,762.16,335.871,256,380,4.04,45.66,13.01
+resnest101e,757.06,338.138,256,256,13.38,28.66,48.28
+gluon_seresnext101_64x4d,754.82,339.139,256,224,15.53,31.25,88.23
+seresnext101_32x8d,729.3,351.006,256,224,16.48,31.25,93.57
+cait_s24_224,717.85,356.607,256,224,9.35,40.58,46.92
+hrnet_w48,714.5,358.278,256,224,17.34,28.56,77.47
+hrnet_w44,713.57,358.746,256,224,14.94,26.92,67.06
+tresnet_m_448,711.59,359.746,256,448,22.94,29.21,31.39
+regnetz_d32,706.63,362.269,256,320,9.33,37.08,27.58
+coat_tiny,706.1,362.539,256,224,4.35,27.2,5.5
+nest_small,681.35,375.712,256,224,10.35,40.04,38.35
+vit_large_r50_s32_224,679.9,376.512,256,224,19.58,24.41,328.99
+jx_nest_small,675.59,378.914,256,224,10.35,40.04,38.35
+twins_svt_large,672.39,380.718,256,224,15.15,35.1,99.27
+crossvit_base_240,669.47,382.378,256,240,21.22,36.33,105.03
+efficientnet_b4,667.37,383.583,256,384,4.51,50.04,19.34
+twins_pcpvt_large,655.12,390.752,256,224,9.84,35.82,60.99
+gmlp_b16_224,643.01,398.116,256,224,15.78,30.21,73.08
+densenet264d_iabn,638.28,401.062,256,224,13.47,14.0,72.74
+tf_efficientnet_b4,637.3,401.68,256,380,4.49,49.49,19.34
+tf_efficientnet_b4_ap,636.89,401.944,256,380,4.49,49.49,19.34
+tf_efficientnet_b4_ns,636.28,402.323,256,380,4.49,49.49,19.34
+convit_base,621.54,411.866,256,224,17.52,31.77,86.54
+densenet264,620.68,412.433,256,224,12.95,12.8,72.69
+dpn131,617.53,414.539,256,224,16.09,32.97,79.25
+swin_base_patch4_window7_224,616.27,415.39,256,224,15.47,36.63,87.77
+xcit_medium_24_p16_224_dist,612.37,418.035,256,224,16.13,31.71,84.4
+xcit_medium_24_p16_224,612.25,418.117,256,224,16.13,31.71,84.4
+vit_small_patch16_384,593.83,431.086,256,384,15.52,50.78,22.2
+coat_mini,591.16,433.03,256,224,6.82,33.68,10.34
+xception71,588.93,434.671,256,299,18.09,69.92,42.34
+vit_small_r26_s32_384,574.79,445.365,256,384,10.43,29.85,36.47
+hrnet_w64,567.27,451.268,256,224,28.97,35.09,128.06
+dpn107,564.61,453.398,256,224,18.38,33.46,86.92
+eca_nfnet_l1,558.87,458.051,256,320,14.92,34.42,41.41
+senet154,557.21,459.42,256,224,20.77,38.69,115.09
+vit_base_r50_s16_224,556.92,459.655,256,224,21.66,35.29,98.66
+gluon_senet154,556.63,459.899,256,224,20.77,38.69,115.09
+legacy_senet154,554.67,461.525,256,224,20.77,38.69,115.09
+xcit_tiny_24_p16_384_dist,554.63,461.554,256,384,6.87,34.29,12.12
+xcit_small_12_p16_384_dist,546.42,468.488,256,384,14.14,36.51,26.25
+resnet152d,540.83,473.331,256,320,24.08,47.67,60.21
+seresnet200d,537.6,476.175,256,256,20.01,43.15,71.86
+ecaresnet200d,537.5,476.265,256,256,20.0,43.15,64.69
+regnety_320,520.96,491.382,256,224,32.34,30.26,145.05
+nest_base,514.48,497.58,256,224,17.96,53.39,67.72
+regnety_160,513.33,498.688,256,288,26.37,38.07,83.59
+jx_nest_base,510.61,501.344,256,224,17.96,53.39,67.72
+tnt_s_patch16_224,491.14,521.218,256,224,5.24,24.37,23.76
+seresnet152d,484.32,528.565,256,320,24.09,47.72,66.84
+resnetrs152,479.63,533.736,256,320,24.34,48.14,86.62
+convnext_large,469.79,544.914,256,224,34.4,43.13,197.77
+convnext_large_in22ft1k,468.89,545.96,256,224,34.4,43.13,197.77
+regnetx_320,458.91,557.836,256,224,31.81,36.3,107.81
+halonet_h1,456.22,561.121,256,256,3.0,51.17,8.1
+vit_large_patch32_384,455.24,562.325,256,384,45.31,43.86,306.63
+efficientnetv2_m,448.05,571.353,256,416,18.6,67.5,54.14
+regnetz_e8,441.39,579.972,256,320,15.46,63.94,57.7
+mixer_l16_224,428.45,597.494,256,224,44.6,41.69,208.2
+seresnet269d,419.5,610.23,256,256,26.59,53.6,113.67
+nf_regnet_b5,418.97,611.004,256,456,11.7,61.95,49.74
+xcit_tiny_24_p8_224,417.22,613.573,256,224,9.21,45.39,12.11
+xcit_tiny_24_p8_224_dist,416.09,615.239,256,224,9.21,45.39,12.11
+xcit_small_12_p8_224,415.7,615.812,256,224,18.69,47.21,26.21
+xcit_small_12_p8_224_dist,414.47,617.648,256,224,18.69,47.21,26.21
+efficientnetv2_rw_m,401.01,638.372,256,416,21.49,79.62,53.24
+resnet200d,390.11,656.216,256,320,31.25,67.33,64.69
+resnetv2_50x1_bitm,388.79,658.439,256,448,16.62,44.46,25.55
+tnt_b_patch16_224,387.0,661.483,256,224,14.09,39.01,65.41
+xcit_nano_12_p8_384_dist,385.68,663.749,256,384,6.34,46.08,3.05
+swin_large_patch4_window7_224,383.12,668.188,256,224,34.53,54.94,196.53
+nfnet_f1s,376.13,680.6,256,320,35.97,46.77,132.63
+nfnet_f1,374.75,683.103,256,320,35.97,46.77,132.63
+xcit_large_24_p16_224,373.32,685.729,256,224,35.86,47.27,189.1
+xcit_large_24_p16_224_dist,372.9,686.493,256,224,35.86,47.27,189.1
+ssl_resnext101_32x16d,364.85,701.653,256,224,36.27,51.18,194.03
+ig_resnext101_32x16d,364.28,702.733,256,224,36.27,51.18,194.03
+swsl_resnext101_32x16d,364.28,702.739,256,224,36.27,51.18,194.03
+tresnet_l_448,351.48,728.329,256,448,43.5,47.56,55.99
+resnetrs200,347.71,736.231,256,320,31.51,67.81,93.21
+tf_efficientnetv2_m,339.44,754.17,256,480,24.76,89.84,54.14
+tf_efficientnetv2_m_in21ft1k,338.69,755.846,256,480,24.76,89.84,54.14
+regnetz_d8_evob,337.57,758.355,256,320,6.12,37.08,23.41
+efficientnet_b5,330.43,774.739,256,456,10.46,98.86,30.39
+vit_large_patch16_224,328.99,778.119,256,224,61.6,63.52,304.33
+dm_nfnet_f1,328.44,779.437,256,320,35.97,46.77,132.63
+convnext_xlarge_in22ft1k,319.16,802.079,256,224,60.97,57.5,350.2
+tf_efficientnet_b5,318.33,804.192,256,456,10.46,98.86,30.39
+tf_efficientnet_b5_ap,317.92,805.231,256,456,10.46,98.86,30.39
+tf_efficientnet_b5_ns,317.77,805.609,256,456,10.46,98.86,30.39
+crossvit_15_dagger_408,307.3,833.049,256,408,21.45,95.05,28.5
+beit_large_patch16_224,302.42,846.497,256,224,61.6,63.52,304.43
+xcit_small_24_p16_384_dist,290.22,882.071,256,384,26.72,68.58,47.67
+convmixer_768_32,283.01,904.552,256,224,19.55,25.95,21.11
+eca_nfnet_l2,276.43,926.087,256,384,30.05,68.28,56.72
+regnetz_d8_evos,272.53,939.333,256,320,7.03,38.92,23.46
+resnetv2_152x2_bit_teacher,270.79,945.382,256,224,46.95,45.11,236.34
+xcit_tiny_12_p8_384_dist,269.38,950.319,256,384,14.13,69.14,6.71
+convnext_base_384_in22ft1k,268.35,953.955,256,384,45.2,84.49,88.59
+tresnet_xl_448,264.06,969.471,256,448,60.65,61.31,78.44
+deit_base_patch16_384,258.82,989.071,256,384,55.54,101.56,86.86
+vit_base_patch16_384,258.72,989.47,256,384,55.54,101.56,86.86
+resnest200e,257.74,993.222,256,320,35.69,82.78,70.2
+deit_base_distilled_patch16_384,255.57,1001.681,256,384,55.65,101.82,87.63
+resnetv2_101x1_bitm,242.52,1055.587,256,448,31.65,64.93,44.54
+cait_xxs24_384,236.82,1080.978,256,384,9.63,122.66,12.03
+crossvit_18_dagger_408,228.68,559.729,128,408,32.47,124.87,44.61
+ecaresnet269d,228.16,1122.012,256,352,50.25,101.25,102.09
+vit_large_r50_s32_384,228.16,1121.984,256,384,57.43,76.52,329.09
+nasnetalarge,224.77,1138.915,256,331,23.89,90.56,88.75
+pnasnet5large,224.27,1141.456,256,331,25.04,92.89,86.06
+beit_base_patch16_384,222.85,1148.718,256,384,55.54,101.56,86.74
+resnetrs270,221.37,1156.398,256,352,51.13,105.48,129.86
+xcit_small_24_p8_224,217.0,1179.688,256,224,35.81,90.78,47.63
+xcit_small_24_p8_224_dist,216.72,1181.234,256,224,35.81,90.78,47.63
+nfnet_f2s,212.63,1203.934,256,352,63.22,79.06,193.78
+nfnet_f2,211.38,1211.098,256,352,63.22,79.06,193.78
+xcit_medium_24_p16_384_dist,210.86,1214.056,256,384,47.39,91.64,84.4
+resmlp_big_24_224,198.74,1288.096,256,224,100.23,87.31,129.14
+resmlp_big_24_224_in22ft1k,197.93,1293.386,256,224,100.23,87.31,129.14
+resmlp_big_24_distilled_224,197.54,1295.901,256,224,100.23,87.31,129.14
+efficientnetv2_l,196.77,1301.027,256,480,56.4,157.99,118.52
+tf_efficientnetv2_l,194.85,1313.812,256,480,56.4,157.99,118.52
+tf_efficientnetv2_l_in21ft1k,194.51,1316.142,256,480,56.4,157.99,118.52
+efficientnet_b6,189.94,673.894,128,528,19.4,167.39,43.04
+dm_nfnet_f2,186.3,1374.105,256,352,63.22,79.06,193.78
+tf_efficientnet_b6_ns,183.84,696.254,128,528,19.4,167.39,43.04
+tf_efficientnet_b6_ap,183.67,696.899,128,528,19.4,167.39,43.04
+tf_efficientnet_b6,183.44,697.757,128,528,19.4,167.39,43.04
+swin_base_patch4_window12_384,171.35,746.981,128,384,47.19,134.78,87.9
+vit_base_patch8_224,170.3,1503.253,256,224,78.22,161.69,86.58
+cait_xs24_384,168.32,1520.881,256,384,19.28,183.98,26.67
+vit_base_r50_s16_384,163.47,1566.011,256,384,67.43,135.03,98.95
+vit_base_resnet50_384,163.26,1568.003,256,384,67.43,135.03,98.95
+convmixer_1536_20,162.04,1579.822,256,224,48.68,33.03,51.63
+convnext_large_384_in22ft1k,160.02,1599.775,256,384,101.09,126.74,197.77
+cait_xxs36_384,158.26,1617.553,256,384,14.35,183.7,17.37
+xcit_medium_24_p8_224,157.17,1628.769,256,224,63.53,121.23,84.32
+xcit_medium_24_p8_224_dist,157.13,1629.205,256,224,63.53,121.23,84.32
+eca_nfnet_l3,156.17,1639.21,256,448,52.55,118.4,72.04
+resnetrs350,147.0,1741.501,256,384,77.59,154.74,163.96
+ig_resnext101_32x32d,144.49,1771.686,256,224,87.29,91.12,468.53
+xcit_tiny_24_p8_384_dist,141.04,1815.079,256,384,27.05,132.95,12.11
+xcit_small_12_p8_384_dist,137.26,1865.112,256,384,54.92,138.29,26.21
+vit_huge_patch14_224,131.97,1939.888,256,224,167.4,139.41,632.05
+cait_s24_384,129.91,1970.595,256,384,32.17,245.31,47.06
+xcit_large_24_p16_384_dist,127.3,2011.004,256,384,105.35,137.17,189.1
+efficientnetv2_xl,125.14,2045.696,256,512,93.85,247.32,208.12
+tf_efficientnetv2_xl_in21ft1k,124.18,2061.565,256,512,93.85,247.32,208.12
+resnest269e,119.71,2138.483,256,416,77.69,171.98,110.93
+nfnet_f3s,116.49,2197.633,256,416,115.58,141.78,254.92
+nfnet_f3,115.68,2213.012,256,416,115.58,141.78,254.92
+convnext_xlarge_384_in22ft1k,109.41,1169.883,128,384,179.18,168.99,350.2
+efficientnet_b7,108.34,590.697,64,600,38.33,289.94,66.35
+swin_large_patch4_window12_384,107.91,1186.161,128,384,104.08,202.16,196.74
+resnetrs420,106.92,2394.315,256,416,108.45,213.79,191.89
+tf_efficientnet_b7_ap,105.47,606.819,64,600,38.33,289.94,66.35
+tf_efficientnet_b7_ns,105.47,606.797,64,600,38.33,289.94,66.35
+tf_efficientnet_b7,105.43,607.002,64,600,38.33,289.94,66.35
+dm_nfnet_f3,102.2,2504.865,256,416,115.58,141.78,254.92
+xcit_large_24_p8_224_dist,94.99,2695.138,256,224,141.23,181.56,188.93
+xcit_large_24_p8_224,94.97,2695.444,256,224,141.23,181.56,188.93
+resnetv2_152x2_bit_teacher_384,94.51,2708.696,256,384,136.16,132.56,236.34
+resnetv2_50x3_bitm,93.18,1373.716,128,448,145.7,133.37,217.32
+vit_large_patch16_384,88.86,2880.826,256,384,191.21,270.24,304.72
+vit_giant_patch14_224,87.24,2934.557,256,224,267.18,192.64,1012.61
+cait_s36_384,87.06,2940.456,256,384,47.99,367.4,68.37
+ig_resnext101_32x48d,85.45,1497.919,128,224,153.57,131.06,828.41
+beit_large_patch16_384,77.25,3314.013,256,384,191.21,270.24,305.0
+xcit_small_24_p8_384_dist,72.02,3554.648,256,384,105.24,265.91,47.63
+resnetv2_152x2_bitm,69.3,1847.055,128,448,184.99,180.43,236.34
+efficientnet_b8,67.89,942.729,64,672,63.48,442.89,87.41
+tf_efficientnet_b8,66.48,962.617,64,672,63.48,442.89,87.41
+tf_efficientnet_b8_ap,66.47,962.833,64,672,63.48,442.89,87.41
+nfnet_f4s,65.05,3935.454,256,512,216.26,262.26,316.07
+nfnet_f4,64.4,3975.384,256,512,216.26,262.26,316.07
+resnetv2_101x3_bitm,57.01,2245.355,128,448,280.33,194.78,387.93
+dm_nfnet_f4,56.77,4509.206,256,512,216.26,262.26,316.07
+xcit_medium_24_p8_384_dist,53.32,4800.797,256,384,186.67,354.73,84.32
+vit_gigantic_patch14_224,53.03,4827.242,256,224,483.95,275.37,1844.44
+nfnet_f5s,47.06,5439.402,256,544,290.97,349.71,377.21
+nfnet_f5,46.62,5491.767,256,544,290.97,349.71,377.21
+dm_nfnet_f5,41.31,6196.83,256,544,290.97,349.71,377.21
+tf_efficientnet_l2_ns_475,38.18,1676.113,64,475,172.11,609.89,480.31
+nfnet_f6s,33.85,7563.356,256,576,378.69,452.2,438.36
+nfnet_f6,33.53,7634.937,256,576,378.69,452.2,438.36
+xcit_large_24_p8_384_dist,32.23,7943.828,256,384,415.0,531.82,188.93
+beit_large_patch16_512,31.3,2044.665,64,512,362.24,656.39,305.67
+cait_m36_384,30.09,8507.481,256,384,173.11,734.81,271.22
+dm_nfnet_f6,30.02,8526.734,256,576,378.69,452.2,438.36
+nfnet_f7s,26.4,9698.058,256,608,480.39,570.85,499.5
+nfnet_f7,26.15,9791.252,256,608,480.39,570.85,499.5
+resnetv2_152x4_bitm,18.07,3541.734,64,480,844.84,414.26,936.53
+efficientnet_l2,13.68,2266.184,31,800,479.12,1707.39,480.31
+tf_efficientnet_l2_ns,13.51,2295.302,31,800,479.12,1707.39,480.31
+cait_m48_448,13.05,9810.688,128,448,329.41,1708.23,356.46

results/benchmark-infer-amp-nhwc-pt110-cu113-rtx3090.csv

@@ -0,0 +1,696 @@
+model,infer_samples_per_sec,infer_step_time,infer_batch_size,infer_img_size,infer_gmacs,infer_macts,param_count
+tinynet_e,52443.33,4.87,256,106,0.03,0.69,2.04
+lcnet_035,41836.65,6.108,256,224,0.03,1.04,1.64
+mobilenetv3_small_050,40675.32,6.283,256,224,0.03,0.92,1.59
+lcnet_050,37031.22,6.902,256,224,0.05,1.26,1.88
+mobilenetv3_small_075,33134.36,7.715,256,224,0.05,1.3,2.04
+mobilenetv3_small_100,30332.75,8.428,256,224,0.06,1.42,2.54
+tinynet_d,29729.64,8.599,256,152,0.05,1.42,2.34
+tf_mobilenetv3_small_minimal_100,28698.97,8.909,256,224,0.06,1.41,2.04
+tf_mobilenetv3_small_075,27515.61,9.292,256,224,0.05,1.3,2.04
+tf_mobilenetv3_small_100,25529.73,10.016,256,224,0.06,1.42,2.54
+lcnet_075,25351.21,10.087,256,224,0.1,1.99,2.36
+lcnet_100,21251.26,12.035,256,224,0.16,2.52,2.95
+mnasnet_small,20610.12,12.41,256,224,0.07,2.16,2.03
+levit_128s,18679.12,13.694,256,224,0.31,1.88,7.78
+mobilenetv2_035,18422.82,13.885,256,224,0.07,2.86,1.68
+ghostnet_050,17065.19,14.99,256,224,0.05,1.77,2.59
+regnetx_002,17008.51,15.04,256,224,0.2,2.16,2.68
+mnasnet_050,16943.72,15.098,256,224,0.11,3.07,2.22
+tinynet_c,16089.67,15.9,256,184,0.11,2.87,2.46
+mobilenetv2_050,15226.35,16.802,256,224,0.1,3.64,1.97
+regnety_002,15182.41,16.851,256,224,0.2,2.17,3.16
+semnasnet_050,14956.97,17.104,256,224,0.11,3.44,2.08
+lcnet_150,14301.96,17.889,256,224,0.34,3.79,4.5
+regnetx_004,13656.57,18.735,256,224,0.4,3.14,5.16
+gernet_s,13307.64,19.226,256,224,0.75,2.65,8.17
+mobilenetv3_large_075,12875.59,19.872,256,224,0.16,4.0,3.99
+levit_128,12869.05,19.881,256,224,0.41,2.71,9.21
+mnasnet_075,11851.07,21.591,256,224,0.23,4.77,3.17
+mobilenetv3_rw,11749.18,21.778,256,224,0.23,4.41,5.48
+hardcorenas_a,11724.1,21.825,256,224,0.23,4.38,5.26
+mobilenetv3_large_100,11546.8,22.16,256,224,0.23,4.41,5.48
+mobilenetv3_large_100_miil,11527.85,22.196,256,224,0.23,4.41,5.48
+levit_192,11521.32,22.208,256,224,0.66,3.2,10.95
+tf_mobilenetv3_large_075,11097.44,23.056,256,224,0.16,4.0,3.99
+ese_vovnet19b_slim_dw,11011.34,23.237,256,224,0.4,5.28,1.9
+tf_mobilenetv3_large_minimal_100,10807.48,23.676,256,224,0.22,4.4,3.92
+tinynet_b,10794.3,23.705,256,188,0.21,4.44,3.73
+hardcorenas_b,10649.59,24.027,256,224,0.26,5.09,5.18
+hardcorenas_c,10616.11,24.103,256,224,0.28,5.01,5.52
+mnasnet_100,10429.97,24.534,256,224,0.33,5.46,4.38
+mnasnet_b1,10405.42,24.591,256,224,0.33,5.46,4.38
+mixer_s32_224,10314.02,24.809,256,224,1.0,2.28,19.1
+ssl_resnet18,10228.31,25.016,256,224,1.82,2.48,11.69
+gluon_resnet18_v1b,10218.86,25.04,256,224,1.82,2.48,11.69
+swsl_resnet18,10183.71,25.126,256,224,1.82,2.48,11.69
+resnet18,10180.53,25.135,256,224,1.82,2.48,11.69
+tf_mobilenetv3_large_100,10049.81,25.462,256,224,0.23,4.41,5.48
+ghostnet_100,9984.48,25.628,256,224,0.15,3.55,5.18
+mobilenetv2_075,9980.04,25.64,256,224,0.22,5.86,2.64
+semnasnet_075,9964.54,25.68,256,224,0.23,5.54,2.91
+hardcorenas_d,9736.47,26.282,256,224,0.3,4.93,7.5
+seresnet18,9448.37,27.084,256,224,1.82,2.49,11.78
+spnasnet_100,9378.19,27.285,256,224,0.35,6.03,4.42
+vit_small_patch32_224,9330.47,27.426,256,224,1.15,2.5,22.88
+regnety_006,9290.2,27.545,256,224,0.61,4.33,6.06
+legacy_seresnet18,9195.77,27.828,256,224,1.82,2.49,11.78
+mobilenetv2_100,8931.8,28.65,256,224,0.31,6.68,3.5
+regnety_004,8913.44,28.71,256,224,0.41,3.89,4.34
+semnasnet_100,8878.3,28.823,256,224,0.32,6.23,3.89
+mnasnet_a1,8872.5,28.841,256,224,0.32,6.23,3.89
+hardcorenas_f,8746.45,29.258,256,224,0.35,5.57,8.2
+tinynet_a,8705.83,29.394,256,192,0.35,5.41,6.19
+efficientnet_lite0,8678.08,29.489,256,224,0.4,6.74,4.65
+levit_256,8605.17,29.738,256,224,1.13,4.23,18.89
+fbnetc_100,8589.17,29.794,256,224,0.4,6.51,5.57
+hardcorenas_e,8587.1,29.801,256,224,0.35,5.65,8.07
+resnet18d,8454.18,30.27,256,224,2.06,3.29,11.71
+tf_efficientnetv2_b0,8451.14,30.28,256,224,0.73,4.77,7.14
+ese_vovnet19b_slim,8409.91,30.429,256,224,1.69,3.52,3.17
+regnetx_008,8190.92,31.242,256,224,0.81,5.15,7.26
+vit_tiny_r_s16_p8_224,8188.54,31.252,256,224,0.44,2.06,6.34
+regnetx_006,7654.81,33.432,256,224,0.61,3.98,6.2
+ghostnet_130,7516.11,34.049,256,224,0.24,4.6,7.36
+tf_efficientnet_lite0,7513.34,34.061,256,224,0.4,6.74,4.65
+regnety_008,7334.93,34.89,256,224,0.81,5.25,6.26
+mnasnet_140,7212.43,35.483,256,224,0.6,7.71,7.12
+efficientnet_b0,7189.76,35.595,256,224,0.4,6.75,5.29
+rexnetr_100,6816.78,37.543,256,224,0.43,7.72,4.88
+mobilenetv2_110d,6785.21,37.718,256,224,0.45,8.71,4.52
+tf_efficientnet_b0_ns,6397.82,40.002,256,224,0.4,6.75,5.29
+tf_efficientnet_b0_ap,6366.01,40.201,256,224,0.4,6.75,5.29
+tf_efficientnet_b0,6345.74,40.33,256,224,0.4,6.75,5.29
+hrnet_w18_small,6240.1,41.014,256,224,1.61,5.72,13.19
+semnasnet_140,6194.36,41.317,256,224,0.6,8.87,6.11
+ese_vovnet19b_dw,6016.74,42.537,256,224,1.34,8.25,6.54
+mobilenetv2_140,6010.52,42.581,256,224,0.6,9.57,6.11
+resnet34,5949.09,43.021,256,224,3.67,3.74,21.8
+gluon_resnet34_v1b,5944.0,43.057,256,224,3.67,3.74,21.8
+tv_resnet34,5920.52,43.227,256,224,3.67,3.74,21.8
+gernet_m,5909.18,43.311,256,224,3.02,5.24,21.14
+efficientnet_lite1,5815.72,44.007,256,240,0.62,10.14,5.42
+selecsls42,5752.79,44.489,256,224,2.94,4.62,30.35
+selecsls42b,5752.33,44.492,256,224,2.98,4.62,32.46
+fbnetv3_b,5567.33,45.971,256,256,0.55,9.1,8.6
+efficientnet_b1_pruned,5558.39,46.045,256,240,0.4,6.21,6.33
+resnet26,5532.69,46.26,256,224,2.36,7.35,16.0
+skresnet18,5529.57,46.284,256,224,1.82,3.24,11.96
+efficientnet_es_pruned,5519.91,46.366,256,224,1.81,8.73,5.44
+efficientnet_es,5518.49,46.378,256,224,1.81,8.73,5.44
+tf_efficientnetv2_b1,5515.51,46.402,256,240,1.21,7.34,8.14
+rexnet_100,5500.83,46.527,256,224,0.41,7.44,4.8
+dla46_c,5464.03,46.84,256,224,0.58,4.5,1.3
+seresnet34,5439.78,47.05,256,224,3.67,3.74,21.96
+resnet34d,5325.4,48.06,256,224,3.91,4.54,21.82
+resnetblur18,5301.3,48.279,256,224,2.34,3.39,11.69
+legacy_seresnet34,5280.08,48.473,256,224,3.67,3.74,21.96
+rexnetr_130,5239.92,48.845,256,224,0.68,9.81,7.61
+tf_efficientnet_es,5226.26,48.971,256,224,1.81,8.73,5.44
+tf_efficientnet_lite1,5192.67,49.288,256,240,0.62,10.14,5.42
+levit_384,5136.88,49.824,256,224,2.36,6.26,39.13
+nf_regnet_b0,5081.73,50.365,256,256,0.64,5.58,8.76
+selecsls60,5015.37,51.032,256,224,3.59,5.52,30.67
+selecsls60b,5008.51,51.102,256,224,3.63,5.52,32.77
+mobilenetv2_120d,5006.42,51.123,256,224,0.69,11.97,5.83
+repvgg_b0,4831.81,52.969,256,224,3.41,6.15,15.82
+resnet26d,4774.01,53.612,256,224,2.6,8.15,16.01
+rexnetr_150,4621.39,55.383,256,224,0.89,11.13,9.78
+fbnetv3_d,4543.95,56.327,256,256,0.68,11.1,10.31
+nf_resnet26,4522.01,56.601,256,224,2.41,7.35,16.0
+deit_tiny_patch16_224,4516.94,56.664,256,224,1.26,5.97,5.72
+visformer_tiny,4514.36,56.695,256,224,1.27,5.72,10.32
+vit_tiny_patch16_224,4509.39,56.76,256,224,1.26,5.97,5.72
+efficientnet_lite2,4476.87,57.171,256,260,0.89,12.9,6.09
+xcit_nano_12_p16_224_dist,4461.13,57.373,256,224,0.56,4.17,3.05
+deit_tiny_distilled_patch16_224,4450.37,57.513,256,224,1.27,6.01,5.91
+xcit_nano_12_p16_224,4443.48,57.6,256,224,0.56,4.17,3.05
+pit_ti_distilled_224,4267.82,59.972,256,224,0.71,6.23,5.1
+pit_ti_224,4256.93,60.125,256,224,0.7,6.19,4.85
+resmlp_12_distilled_224,4163.6,61.475,256,224,3.01,5.5,15.35
+resmlp_12_224,4160.41,61.521,256,224,3.01,5.5,15.35
+gernet_l,4096.96,62.474,256,256,4.57,8.0,31.08
+tf_efficientnetv2_b2,4081.75,62.706,256,260,1.72,9.84,10.1
+legacy_seresnext26_32x4d,4060.33,63.037,256,224,2.49,9.39,16.79
+tf_efficientnet_b1,4046.06,63.259,256,240,0.71,10.88,7.79
+tf_efficientnet_b1_ap,4044.6,63.282,256,240,0.71,10.88,7.79
+tf_efficientnet_b1_ns,4042.64,63.313,256,240,0.71,10.88,7.79
+tf_efficientnet_lite2,4027.76,63.547,256,260,0.89,12.9,6.09
+resnext26ts,3956.89,64.685,256,256,2.43,10.52,10.3
+dla46x_c,3913.74,65.398,256,224,0.54,5.66,1.07
+efficientnet_b1,3903.13,65.577,256,256,0.77,12.22,7.79
+dla34,3901.31,65.607,256,224,3.07,5.02,15.74
+mixer_b32_224,3872.36,66.099,256,224,3.24,6.29,60.29
+rexnet_130,3847.96,66.517,256,224,0.68,9.71,7.56
+vit_base_patch32_224_sam,3843.13,66.599,256,224,4.41,5.01,88.22
+nf_seresnet26,3833.86,66.763,256,224,2.41,7.36,17.4
+eca_resnext26ts,3823.14,66.95,256,256,2.43,10.52,10.3
+vit_base_patch32_224,3821.26,66.98,256,224,4.41,5.01,88.22
+nf_ecaresnet26,3821.0,66.987,256,224,2.41,7.36,16.0
+seresnext26ts,3814.04,67.108,256,256,2.43,10.52,10.39
+dla60x_c,3762.66,68.025,256,224,0.59,6.01,1.32
+efficientnet_b2_pruned,3734.36,68.542,256,260,0.73,9.13,8.31
+regnetx_016,3682.98,69.498,256,224,1.62,7.93,9.19
+resnet26t,3677.69,69.597,256,256,3.35,10.52,16.01
+gcresnext26ts,3645.26,70.216,256,256,2.43,10.53,10.48
+rexnet_150,3639.9,70.32,256,224,0.9,11.21,9.73
+seresnext26tn_32x4d,3635.04,70.413,256,224,2.7,10.09,16.81
+seresnext26t_32x4d,3629.32,70.524,256,224,2.7,10.09,16.81
+ecaresnext50t_32x4d,3625.17,70.606,256,224,2.7,10.09,15.41
+seresnext26d_32x4d,3620.48,70.696,256,224,2.73,10.19,16.81
+ecaresnet50d_pruned,3618.55,70.735,256,224,2.53,6.43,19.94
+ecaresnext26t_32x4d,3617.96,70.747,256,224,2.7,10.09,15.41
+resnetv2_50,3573.69,71.622,256,224,4.11,11.11,25.55
+eca_botnext26ts_256,3538.36,72.339,256,256,2.46,11.6,10.59
+repvgg_a2,3525.89,72.594,256,224,5.7,6.26,28.21
+pit_xs_224,3490.24,73.336,256,224,1.4,7.71,10.62
+mixer_s16_224,3485.97,73.426,256,224,3.79,5.97,18.53
+pit_xs_distilled_224,3470.1,73.761,256,224,1.41,7.76,11.0
+eca_halonext26ts,3433.36,74.552,256,256,2.44,11.46,10.76
+gluon_resnet50_v1b,3424.93,74.734,256,224,4.11,11.11,25.56
+rexnetr_200,3422.25,74.791,256,224,1.59,15.11,16.52
+ssl_resnet50,3418.93,74.864,256,224,4.11,11.11,25.56
+resnet50,3416.87,74.911,256,224,4.11,11.11,25.56
+tv_resnet50,3416.58,74.915,256,224,4.11,11.11,25.56
+swsl_resnet50,3415.95,74.931,256,224,4.11,11.11,25.56
+ecaresnetlight,3410.59,75.049,256,224,4.11,8.42,30.16
+efficientnet_em,3345.86,76.5,256,240,3.04,14.34,6.9
+dpn68b,3321.6,77.058,256,224,2.35,10.47,12.61
+dpn68,3321.37,77.064,256,224,2.35,10.47,12.61
+resnet32ts,3318.56,77.131,256,256,4.63,11.58,17.96
+botnet26t_256,3309.73,77.335,256,256,3.32,11.98,12.49
+resnet33ts,3269.93,78.278,256,256,4.76,11.66,19.68
+resnetv2_50t,3252.33,78.701,256,224,4.32,11.82,25.57
+halonet26t,3247.75,78.812,256,256,3.19,11.69,12.48
+resnetv2_50d,3238.36,79.041,256,224,4.35,11.92,25.57
+gluon_resnet50_v1c,3234.34,79.139,256,224,4.35,11.92,25.58
+nf_regnet_b1,3220.71,79.473,256,288,1.02,9.2,10.22
+tf_efficientnet_em,3211.19,79.71,256,240,3.04,14.34,6.9
+nf_regnet_b2,3198.33,80.03,256,272,1.22,9.27,14.31
+xcit_tiny_12_p16_224_dist,3183.42,80.405,256,224,1.24,6.29,6.72
+xcit_tiny_12_p16_224,3177.02,80.567,256,224,1.24,6.29,6.72
+regnety_016,3168.83,80.775,256,224,1.63,8.04,11.2
+gmixer_12_224,3164.43,80.888,256,224,2.67,7.26,12.7
+tf_efficientnet_b2_ap,3144.11,81.41,256,260,1.02,13.83,9.11
+tf_efficientnet_b2,3133.02,81.698,256,260,1.02,13.83,9.11
+eca_resnet33ts,3132.75,81.706,256,256,4.76,11.66,19.68
+seresnet33ts,3132.5,81.713,256,256,4.76,11.66,19.78
+skresnet34,3122.38,81.975,256,224,3.67,5.13,22.28
+resnet50t,3114.92,82.173,256,224,4.32,11.82,25.57
+tf_efficientnet_b2_ns,3112.24,82.244,256,260,1.02,13.83,9.11
+gluon_resnet50_v1d,3111.66,82.259,256,224,4.35,11.92,25.58
+resnet50d,3107.36,82.374,256,224,4.35,11.92,25.58
+vovnet39a,3097.06,82.648,256,224,7.09,6.73,22.6
+bat_resnext26ts,3052.18,83.861,256,256,2.53,12.51,10.73
+legacy_seresnet50,3043.79,84.094,256,224,3.88,10.6,28.09
+cspresnext50,3003.53,85.222,256,224,3.1,12.14,20.57
+gcresnet33ts,2986.89,85.696,256,256,4.76,11.68,19.88
+selecsls84,2981.3,85.857,256,224,5.9,7.57,50.95
+vit_small_patch32_384,2981.2,85.86,256,384,3.45,8.25,22.92
+ese_vovnet39b,2975.03,86.038,256,224,7.09,6.74,24.57
+efficientnet_b2a,2964.53,86.343,256,288,1.12,16.2,9.11
+efficientnet_b2,2962.37,86.406,256,288,1.12,16.2,9.11
+eca_vovnet39b,2962.13,86.413,256,224,7.09,6.74,22.6
+res2net50_48w_2s,2946.94,86.858,256,224,4.18,11.72,25.29
+seresnet50,2942.22,86.998,256,224,4.11,11.13,28.09
+efficientnet_b3_pruned,2917.98,87.72,256,300,1.04,11.86,9.86
+haloregnetz_b,2882.69,88.795,256,224,1.97,11.94,11.68
+vit_small_resnet26d_224,2880.85,88.851,256,224,5.07,11.12,63.61
+vgg11,2809.5,91.106,256,224,7.61,7.44,132.86
+gluon_resnext50_32x4d,2797.04,91.514,256,224,4.26,14.4,25.03
+swsl_resnext50_32x4d,2795.88,91.552,256,224,4.26,14.4,25.03
+tv_resnext50_32x4d,2795.77,91.554,256,224,4.26,14.4,25.03
+mixnet_s,2795.45,91.566,256,224,0.25,6.25,4.13
+ssl_resnext50_32x4d,2795.35,91.567,256,224,4.26,14.4,25.03
+resnext50_32x4d,2795.04,91.578,256,224,4.26,14.4,25.03
+resnetaa50d,2786.19,91.87,256,224,5.39,12.44,25.58
+ecaresnet101d_pruned,2770.2,92.4,256,224,3.48,7.69,24.88
+densenet121,2765.59,92.554,256,224,2.87,6.9,7.98
+tv_densenet121,2754.97,92.91,256,224,2.87,6.9,7.98
+gluon_resnet50_v1s,2746.47,93.198,256,224,5.47,13.52,25.68
+rexnet_200,2741.96,93.353,256,224,1.56,14.91,16.37
+cspresnet50,2739.95,93.42,256,256,4.54,11.5,21.62
+seresnet50t,2715.96,94.246,256,224,4.32,11.83,28.1
+resnetrs50,2709.47,94.471,256,224,4.48,12.14,35.69
+crossvit_tiny_240,2698.8,94.845,256,240,1.57,9.08,7.01
+ecaresnet50d,2695.97,94.946,256,224,4.35,11.93,25.58
+tf_mixnet_s,2668.56,95.92,256,224,0.25,6.25,4.13
+dla60,2658.15,96.295,256,224,4.26,10.16,22.04
+crossvit_9_240,2646.19,96.732,256,240,1.85,9.52,8.55
+densenet121d,2643.5,96.83,256,224,3.11,7.7,8.0
+efficientnet_lite3,2630.67,97.302,256,300,1.65,21.85,8.2
+convit_tiny,2612.77,97.969,256,224,1.26,7.94,5.71
+hrnet_w18_small_v2,2612.23,97.989,256,224,2.62,9.65,15.6
+crossvit_9_dagger_240,2607.67,98.16,256,240,1.99,9.97,8.78
+regnetz_b16,2592.86,98.722,256,288,2.39,16.43,9.72
+resnext50d_32x4d,2582.77,99.107,256,224,4.5,15.2,25.05
+cspresnet50d,2577.78,99.299,256,256,4.86,12.55,21.64
+cspresnet50w,2560.3,99.977,256,256,5.04,12.19,28.12
+vgg11_bn,2537.64,100.869,256,224,7.62,7.44,132.87
+vovnet57a,2533.31,101.041,256,224,8.95,7.52,36.64
+resnetblur50,2528.53,101.233,256,224,5.16,12.02,25.56
+gmlp_ti16_224,2479.86,103.22,256,224,1.34,7.55,5.87
+seresnext50_32x4d,2470.29,103.618,256,224,4.26,14.42,27.56
+legacy_seresnext50_32x4d,2466.47,103.78,256,224,4.26,14.42,27.56
+seresnetaa50d,2460.49,104.031,256,224,5.4,12.46,28.11
+gluon_seresnext50_32x4d,2457.31,104.168,256,224,4.26,14.42,27.56
+fbnetv3_g,2441.71,104.833,256,288,1.77,21.09,16.62
+res2net50_26w_4s,2439.77,104.916,256,224,4.28,12.61,25.7
+ese_vovnet57b,2429.1,105.377,256,224,8.95,7.52,38.61
+vit_tiny_r_s16_p8_384,2420.04,105.772,256,384,1.34,6.49,6.36
+gcresnet50t,2406.95,106.347,256,256,5.42,14.67,25.9
+adv_inception_v3,2387.53,107.209,256,299,5.73,8.97,23.83
+inception_v3,2378.4,107.624,256,299,5.73,8.97,23.83
+gluon_inception_v3,2378.14,107.635,256,299,5.73,8.97,23.83
+dla60x,2376.48,107.71,256,224,3.54,13.8,17.35
+efficientnetv2_rw_t,2371.59,107.933,256,288,3.19,16.42,13.65
+tf_inception_v3,2369.67,108.018,256,299,5.73,8.97,23.83
+densenetblur121d,2366.44,108.168,256,224,3.11,7.9,8.0
+tf_efficientnet_lite3,2359.22,108.498,256,300,1.65,21.85,8.2
+resnetblur50d,2356.51,108.624,256,224,5.4,12.82,25.58
+nf_seresnet50,2321.05,110.283,256,224,4.21,11.13,28.09
+lambda_resnet26rpt_256,2319.42,110.361,256,256,3.16,11.87,10.99
+tf_efficientnetv2_b3,2313.61,110.636,256,300,3.04,15.74,14.36
+nf_ecaresnet50,2312.19,110.706,256,224,4.21,11.13,25.56
+resnest14d,2311.92,110.718,256,224,2.76,7.33,10.61
+deit_small_patch16_224,2305.42,111.031,256,224,4.61,11.95,22.05
+pit_s_224,2300.71,111.258,256,224,2.88,11.56,23.46
+vit_small_patch16_224,2295.85,111.494,256,224,4.61,11.95,22.05
+pit_s_distilled_224,2281.8,112.18,256,224,2.9,11.64,24.04
+densenet169,2278.65,112.335,256,224,3.4,7.3,14.15
+vit_base_resnet26d_224,2268.07,112.857,256,224,6.97,13.16,101.4
+deit_small_distilled_patch16_224,2267.44,112.892,256,224,4.63,12.02,22.44
+skresnet50,2244.08,114.064,256,224,4.11,12.5,25.8
+darknet53,2213.23,115.656,256,256,9.31,12.39,41.61
+sehalonet33ts,2208.53,115.902,256,256,3.55,14.7,13.69
+res2net50_14w_8s,2201.46,116.274,256,224,4.21,13.28,25.06
+gc_efficientnetv2_rw_t,2198.9,116.41,256,288,3.2,16.45,13.68
+gcresnext50ts,2187.09,117.039,256,256,3.75,15.46,15.67
+resnetv2_101,2146.17,119.27,256,224,7.83,16.23,44.54
+resmlp_24_224,2127.24,120.334,256,224,5.96,10.91,30.02
+resmlp_24_distilled_224,2125.15,120.451,256,224,5.96,10.91,30.02
+skresnet50d,2105.03,121.599,256,224,4.36,13.31,25.82
+gluon_resnet101_v1b,2098.13,122.001,256,224,7.83,16.23,44.55
+tv_resnet101,2095.51,122.152,256,224,7.83,16.23,44.55
+resnet101,2094.89,122.19,256,224,7.83,16.23,44.55
+res2next50,2089.9,122.482,256,224,4.2,13.71,24.67
+ecaresnet26t,2087.45,122.626,256,320,5.24,16.44,16.01
+dla60_res2net,2078.29,123.165,256,224,4.15,12.34,20.85
+nf_regnet_b3,2039.85,125.487,256,320,2.05,14.61,18.59
+gluon_resnet101_v1c,2026.41,126.32,256,224,8.08,17.04,44.57
+resnetv2_101d,2021.62,126.619,256,224,8.07,17.04,44.56
+gluon_resnet101_v1d,1976.83,129.488,256,224,8.08,17.04,44.57
+vgg13,1972.22,129.79,256,224,11.31,12.25,133.05
+wide_resnet50_2,1932.56,132.455,256,224,11.43,14.4,68.88
+sebotnet33ts_256,1892.63,135.25,256,256,3.89,17.46,13.7
+repvgg_b1,1864.64,137.281,256,224,13.16,10.64,57.42
+resnetaa101d,1840.38,139.09,256,224,9.12,17.56,44.57
+legacy_seresnet101,1838.54,139.229,256,224,7.61,15.74,49.33
+resnet51q,1828.83,139.968,256,288,8.07,20.94,35.7
+gluon_resnet101_v1s,1822.02,140.491,256,224,9.19,18.64,44.67
+dla102,1820.91,140.577,256,224,7.19,14.18,33.27
+dla60_res2next,1817.78,140.818,256,224,3.49,13.17,17.03
+coat_lite_tiny,1811.49,141.309,256,224,1.6,11.65,5.72
+vit_base_resnet50d_224,1807.55,141.614,256,224,8.73,16.92,110.97
+tf_efficientnet_b3,1803.3,141.948,256,300,1.87,23.83,12.23
+tf_efficientnet_b3_ap,1802.94,141.978,256,300,1.87,23.83,12.23
+tf_efficientnet_b3_ns,1802.1,142.043,256,300,1.87,23.83,12.23
+cspdarknet53,1801.31,142.107,256,256,6.57,16.81,27.64
+seresnet101,1796.92,142.453,256,224,7.84,16.27,49.33
+efficientnet_b3,1780.99,143.727,256,320,2.01,26.52,12.23
+efficientnet_b3a,1780.57,143.761,256,320,2.01,26.52,12.23
+densenet201,1754.6,145.89,256,224,4.34,7.85,20.01
+vgg13_bn,1750.09,146.264,256,224,11.33,12.25,133.05
+ssl_resnext101_32x4d,1737.01,147.366,256,224,8.01,21.23,44.18
+resnext101_32x4d,1736.55,147.407,256,224,8.01,21.23,44.18
+gluon_resnext101_32x4d,1736.42,147.419,256,224,8.01,21.23,44.18
+swsl_resnext101_32x4d,1735.28,147.514,256,224,8.01,21.23,44.18
+halonet50ts,1731.56,147.831,256,256,5.3,19.2,22.73
+res2net50_26w_6s,1726.13,148.297,256,224,6.33,15.28,37.05
+nf_resnet101,1725.55,148.347,256,224,8.01,16.23,44.55
+coat_lite_mini,1724.07,148.474,256,224,2.0,12.25,11.01
+regnetz_c16,1702.47,150.358,256,320,3.92,25.88,13.46
+nf_resnet50,1701.7,150.426,256,288,6.88,18.37,25.56
+ecaresnet101d,1699.81,150.594,256,224,8.08,17.07,44.57
+xcit_tiny_24_p16_224_dist,1697.25,150.82,256,224,2.34,11.82,12.12
+xcit_tiny_24_p16_224,1694.56,151.06,256,224,2.34,11.82,12.12
+xcit_small_12_p16_224_dist,1690.19,151.45,256,224,4.82,12.58,26.25
+xcit_small_12_p16_224,1687.09,151.729,256,224,4.82,12.58,26.25
+resnest26d,1686.65,151.769,256,224,3.64,9.97,17.07
+convnext_tiny,1676.02,152.729,256,224,4.47,13.44,28.59
+convnext_tiny_hnf,1674.55,152.865,256,224,4.47,13.44,28.59
+vit_small_r26_s32_224,1656.11,154.567,256,224,3.56,9.85,36.43
+resnetblur101d,1639.84,156.101,256,224,9.12,17.94,44.57
+dla102x,1638.98,156.183,256,224,5.89,19.42,26.31
+resnet61q,1638.11,156.265,256,288,9.87,21.52,36.85
+gmixer_24_224,1615.83,158.42,256,224,5.28,14.45,24.72
+vgg16,1592.01,160.79,256,224,15.47,13.56,138.36
+xcit_nano_12_p16_384_dist,1580.63,161.949,256,384,1.64,12.15,3.05
+regnetx_032,1571.21,162.919,256,224,3.2,11.37,15.3
+repvgg_b1g4,1566.48,163.412,256,224,8.15,10.64,39.97
+regnetx_040,1550.05,165.143,256,224,3.99,12.2,22.12
+twins_svt_small,1531.98,167.089,256,224,2.94,13.75,24.06
+seresnext101_32x4d,1528.6,167.459,256,224,8.02,21.26,48.96
+gluon_seresnext101_32x4d,1525.5,167.801,256,224,8.02,21.26,48.96
+legacy_seresnext101_32x4d,1523.33,168.041,256,224,8.02,21.26,48.96
+res2net101_26w_4s,1522.04,168.183,256,224,8.1,18.45,45.21
+xception,1515.79,168.877,256,299,8.4,35.83,22.86
+visformer_small,1505.04,170.081,256,224,4.88,11.43,40.22
+botnet50ts_256,1504.07,170.19,256,256,5.54,22.23,22.74
+resnest50d_1s4x24d,1495.52,171.166,256,224,4.43,13.57,25.68
+resnetv2_50x1_bit_distilled,1491.91,171.579,256,224,4.23,11.11,25.55
+resnetv2_152,1488.31,171.994,256,224,11.55,22.56,60.19
+efficientnet_el,1485.73,172.293,256,300,8.0,30.7,10.59
+efficientnet_el_pruned,1484.88,172.392,256,300,8.0,30.7,10.59
+gluon_resnet152_v1b,1466.32,174.574,256,224,11.56,22.56,60.19
+crossvit_small_240,1465.43,174.68,256,240,5.63,18.17,26.86
+tv_resnet152,1464.76,174.758,256,224,11.56,22.56,60.19
+resnet152,1463.7,174.886,256,224,11.56,22.56,60.19
+res2net50_26w_8s,1440.77,177.671,256,224,8.37,17.95,48.4
+tf_efficientnet_el,1439.79,177.79,256,300,8.0,30.7,10.59
+halo2botnet50ts_256,1435.45,178.33,256,256,5.02,21.78,22.64
+hrnet_w32,1431.59,178.81,256,224,8.97,22.02,41.23
+gluon_resnet152_v1c,1431.14,178.867,256,224,11.8,23.36,60.21
+nf_seresnet101,1430.41,178.958,256,224,8.02,16.27,49.33
+vgg16_bn,1430.18,178.985,256,224,15.5,13.56,138.37
+resnetv2_152d,1427.39,179.335,256,224,11.8,23.36,60.2
+resmlp_36_224,1426.81,179.41,256,224,8.91,16.33,44.69
+resmlp_36_distilled_224,1426.22,179.484,256,224,8.91,16.33,44.69
+nf_ecaresnet101,1423.4,179.839,256,224,8.01,16.27,44.55
+mixnet_m,1415.37,180.859,256,224,0.36,8.19,5.01
+gluon_resnet152_v1d,1405.52,182.126,256,224,11.8,23.36,60.21
+ese_vovnet99b,1395.24,183.468,256,224,16.51,11.27,63.2
+gmlp_s16_224,1389.58,184.216,256,224,4.42,15.1,19.42
+tf_mixnet_m,1379.18,185.603,256,224,0.36,8.19,5.01
+regnety_040,1367.09,187.246,256,224,4.0,12.29,20.65
+vit_large_patch32_224,1365.88,187.414,256,224,15.39,13.3,306.54
+hrnet_w18,1363.25,187.774,256,224,4.32,16.31,21.3
+vit_base_r26_s32_224,1359.83,188.244,256,224,6.81,12.36,101.38
+crossvit_15_240,1344.51,190.391,256,240,5.81,19.77,27.53
+resnetv2_50d_evob,1343.28,190.567,256,224,4.33,11.92,25.59
+ecaresnet50t,1341.42,190.83,256,320,8.82,24.13,25.57
+mixer_b16_224,1341.1,190.877,256,224,12.62,14.53,59.88
+mixer_b16_224_miil,1340.95,190.898,256,224,12.62,14.53,59.88
+vgg19,1334.38,191.835,256,224,19.63,14.86,143.67
+regnety_032,1331.17,192.299,256,288,5.29,18.61,19.44
+cait_xxs24_224,1325.98,193.051,256,224,2.53,20.29,11.96
+gluon_resnet152_v1s,1323.82,193.367,256,224,12.92,24.96,60.32
+mixer_l32_224,1320.53,193.849,256,224,11.27,19.86,206.94
+skresnext50_32x4d,1317.3,194.322,256,224,4.5,17.18,27.48
+xception41,1314.21,194.782,256,299,9.28,39.86,26.97
+crossvit_15_dagger_240,1313.12,194.944,256,240,6.13,20.43,28.21
+densenet161,1297.86,197.236,256,224,7.79,11.06,28.68
+dpn92,1283.16,199.496,256,224,6.54,18.21,37.67
+efficientnet_lite4,1279.6,200.05,256,380,4.04,45.66,13.01
+tresnet_m,1272.03,201.24,256,224,5.74,7.31,31.39
+legacy_seresnet152,1268.99,201.723,256,224,11.33,22.08,66.82
+dla169,1257.87,203.505,256,224,11.6,20.2,53.39
+seresnet152,1252.8,204.325,256,224,11.57,22.61,66.82
+swin_tiny_patch4_window7_224,1250.47,204.71,256,224,4.51,17.06,28.29
+vit_base_patch32_384,1245.01,205.608,256,384,13.06,16.5,88.3
+repvgg_b2,1240.95,206.281,256,224,20.45,12.9,89.02
+twins_pcpvt_small,1237.75,206.812,256,224,3.83,18.08,24.11
+inception_v4,1232.0,207.78,256,299,12.28,15.09,42.68
+regnetx_080,1230.26,208.073,256,224,8.02,14.06,39.57
+convit_small,1215.14,210.662,256,224,5.76,17.87,27.78
+vgg19_bn,1208.18,211.876,256,224,19.66,14.86,143.68
+xcit_nano_12_p8_224_dist,1199.12,213.478,256,224,2.16,15.71,3.05
+xcit_nano_12_p8_224,1193.48,214.486,256,224,2.16,15.71,3.05
+hrnet_w30,1182.47,216.483,256,224,8.15,21.21,37.71
+tf_efficientnet_lite4,1173.72,218.096,256,380,4.04,45.66,13.01
+resnetv2_50d_gn,1171.09,218.587,256,224,4.38,11.92,25.57
+vit_tiny_patch16_384,1165.29,219.677,256,384,4.7,25.39,5.79
+dla102x2,1162.53,220.196,256,224,9.34,29.91,41.28
+regnetx_064,1154.46,221.737,256,224,6.49,16.37,26.21
+vit_small_resnet50d_s16_224,1148.12,222.96,256,224,13.48,24.82,57.53
+resnest50d,1142.3,224.096,256,224,5.4,14.36,27.48
+xcit_tiny_12_p16_384_dist,1120.72,228.412,256,384,3.64,18.26,6.72
+nf_regnet_b4,1118.05,228.958,256,384,4.7,28.61,30.21
+mixnet_l,1111.69,230.267,256,224,0.58,10.84,7.33
+efficientnetv2_s,1109.52,230.718,256,384,8.44,35.77,21.46
+resnet50_gn,1084.83,235.97,256,224,4.14,11.11,25.56
+tf_efficientnetv2_s,1082.56,236.462,256,384,8.44,35.77,21.46
+tf_efficientnetv2_s_in21ft1k,1080.83,236.841,256,384,8.44,35.77,21.46
+tf_mixnet_l,1080.28,236.961,256,224,0.58,10.84,7.33
+wide_resnet101_2,1080.17,236.987,256,224,22.8,21.23,126.89
+gluon_resnext101_64x4d,1075.35,238.048,256,224,15.52,31.21,83.46
+dpn98,1075.31,238.059,256,224,11.73,25.2,61.57
+eca_nfnet_l0,1073.57,238.443,256,288,7.12,17.29,24.14
+resnext101_64x4d,1073.4,238.483,256,224,15.52,31.21,83.46
+regnetz_d8,1072.66,238.647,256,320,6.19,37.08,23.37
+convnext_small,1068.35,239.61,256,224,8.7,21.56,50.22
+nfnet_l0,1067.0,239.911,256,288,7.13,17.29,35.07
+resnet200,1062.15,241.009,256,224,15.07,32.19,64.67
+resnetrs101,1051.01,243.563,256,288,13.56,28.53,63.62
+inception_resnet_v2,1033.59,247.668,256,299,13.18,25.06,55.84
+ens_adv_inception_resnet_v2,1033.26,247.747,256,299,13.18,25.06,55.84
+efficientnetv2_rw_s,1032.33,247.972,256,384,8.72,38.03,23.94
+nest_tiny,1020.47,250.854,256,224,5.83,25.48,17.06
+coat_lite_small,1020.33,250.888,256,224,3.96,22.09,19.84
+jx_nest_tiny,1008.4,253.854,256,224,5.83,25.48,17.06
+vit_base_patch16_224_miil,996.21,256.96,256,224,17.58,23.9,86.54
+crossvit_18_240,991.56,258.167,256,240,9.05,26.26,43.27
+gluon_seresnext101_64x4d,990.21,258.518,256,224,15.53,31.25,88.23
+swsl_resnext101_32x8d,977.82,261.793,256,224,16.48,31.21,88.79
+ssl_resnext101_32x8d,977.78,261.803,256,224,16.48,31.21,88.79
+resnext101_32x8d,976.71,262.092,256,224,16.48,31.21,88.79
+resnet101d,976.38,262.181,256,320,16.48,34.77,44.57
+ig_resnext101_32x8d,976.0,262.282,256,224,16.48,31.21,88.79
+crossvit_18_dagger_240,971.37,263.533,256,240,9.5,27.03,44.27
+regnetz_d32,970.52,263.764,256,320,9.33,37.08,27.58
+vit_base_patch16_224_sam,969.72,263.98,256,224,17.58,23.9,86.57
+deit_base_patch16_224,967.9,264.478,256,224,17.58,23.9,86.57
+vit_base_patch16_224,966.99,264.723,256,224,17.58,23.9,86.57
+deit_base_distilled_patch16_224,962.72,265.901,256,224,17.68,24.05,87.34
+repvgg_b3,962.17,266.052,256,224,29.16,15.1,123.09
+resnest50d_4s2x40d,957.24,267.424,256,224,4.4,17.94,30.42
+gluon_xception65,952.59,268.729,256,299,13.96,52.48,39.92
+efficientnet_b4,950.53,269.31,256,384,4.51,50.04,19.34
+xception65,944.4,271.059,256,299,13.96,52.48,39.92
+ese_vovnet39b_evos,937.23,273.135,256,224,7.07,6.74,24.58
+hrnet_w40,919.71,278.336,256,224,12.75,25.29,57.56
+seresnext101_32x8d,905.52,282.696,256,224,16.48,31.25,93.57
+twins_pcpvt_base,904.7,282.951,256,224,6.68,25.25,43.83
+tf_efficientnet_b4,899.69,284.527,256,380,4.49,49.49,19.34
+tf_efficientnet_b4_ap,899.17,284.692,256,380,4.49,49.49,19.34
+tf_efficientnet_b4_ns,898.81,284.808,256,380,4.49,49.49,19.34
+xcit_small_24_p16_224_dist,897.88,285.104,256,224,9.1,23.64,47.67
+xcit_small_24_p16_224,897.08,285.357,256,224,9.1,23.64,47.67
+beit_base_patch16_224,894.85,286.068,256,224,17.58,23.9,86.53
+cait_xxs36_224,891.78,287.055,256,224,3.77,30.34,17.3
+twins_svt_base,885.5,289.087,256,224,8.59,26.33,56.07
+regnety_080,865.03,295.931,256,224,8.0,17.97,39.18
+hrnet_w48,862.23,296.892,256,224,17.34,28.56,77.47
+xcit_tiny_12_p8_224_dist,851.38,300.676,256,224,4.81,23.6,6.71
+xcit_tiny_12_p8_224,850.38,301.029,256,224,4.81,23.6,6.71
+pit_b_224,831.73,307.778,256,224,12.42,32.94,73.76
+pit_b_distilled_224,829.76,308.512,256,224,12.5,33.07,74.79
+nfnet_f0s,822.42,311.265,256,256,12.62,18.05,71.49
+nfnet_f0,820.01,312.18,256,256,12.62,18.05,71.49
+swin_small_patch4_window7_224,808.39,316.666,256,224,8.77,27.47,49.61
+repvgg_b2g4,805.1,317.961,256,224,12.63,12.9,61.76
+resnetv2_50d_evos,799.75,320.087,256,224,4.33,11.92,25.59
+convnext_base,795.35,321.858,256,224,15.38,28.75,88.59
+convnext_base_in22ft1k,795.35,321.856,256,224,15.38,28.75,88.59
+dpn131,790.54,323.815,256,224,16.09,32.97,79.25
+regnetx_120,762.35,335.792,256,224,12.13,21.37,46.11
+regnety_064,760.83,336.463,256,224,6.39,16.41,30.58
+hrnet_w44,758.65,337.428,256,224,14.94,26.92,67.06
+densenet264,753.54,339.715,256,224,12.95,12.8,72.69
+mixnet_xl,746.77,342.795,256,224,0.93,14.57,11.9
+cait_s24_224,746.74,342.81,256,224,9.35,40.58,46.92
+regnety_120,732.77,349.348,256,224,12.14,21.38,51.82
+dm_nfnet_f0,732.71,349.374,256,256,12.62,18.05,71.49
+xception71,724.87,353.154,256,299,18.09,69.92,42.34
+coat_tiny,707.44,361.857,256,224,4.35,27.2,5.5
+resnet152d,694.55,368.574,256,320,24.08,47.67,60.21
+dpn107,691.54,370.175,256,224,18.38,33.46,86.92
+hrnet_w64,684.53,373.966,256,224,28.97,35.09,128.06
+nest_small,681.18,375.805,256,224,10.35,40.04,38.35
+vit_large_r50_s32_224,676.89,378.187,256,224,19.58,24.41,328.99
+jx_nest_small,675.63,378.895,256,224,10.35,40.04,38.35
+twins_svt_large,674.75,379.385,256,224,15.15,35.1,99.27
+seresnet200d,673.13,380.295,256,256,20.01,43.15,71.86
+ecaresnet200d,672.08,380.893,256,256,20.0,43.15,64.69
+repvgg_b3g4,671.77,381.071,256,224,17.89,15.1,83.83
+cspresnext50_iabn,667.96,383.242,256,256,4.02,15.86,20.57
+gmlp_b16_224,667.23,383.66,256,224,15.78,30.21,73.08
+crossvit_base_240,662.98,386.123,256,240,21.22,36.33,105.03
+xcit_medium_24_p16_224_dist,661.18,387.174,256,224,16.13,31.71,84.4
+xcit_medium_24_p16_224,660.34,387.669,256,224,16.13,31.71,84.4
+tresnet_l,657.41,389.391,256,224,10.88,11.9,55.99
+twins_pcpvt_large,657.27,389.474,256,224,9.84,35.82,60.99
+gluon_senet154,656.15,390.143,256,224,20.77,38.69,115.09
+senet154,655.06,390.791,256,224,20.77,38.69,115.09
+legacy_senet154,654.75,390.977,256,224,20.77,38.69,115.09
+convit_base,638.4,400.989,256,224,17.52,31.77,86.54
+swin_base_patch4_window7_224,617.48,414.571,256,224,15.47,36.63,87.77
+ese_vovnet99b_iabn,611.75,418.46,256,224,16.49,11.27,63.2
+regnetx_160,608.26,420.859,256,224,15.99,25.52,54.28
+cspdarknet53_iabn,603.85,423.936,256,256,6.53,16.81,27.64
+resnetrs152,602.36,424.984,256,320,24.34,48.14,86.62
+seresnet152d,600.43,426.343,256,320,24.09,47.72,66.84
+xcit_tiny_24_p16_384_dist,596.99,428.807,256,384,6.87,34.29,12.12
+xcit_small_12_p16_384_dist,594.31,430.737,256,384,14.14,36.51,26.25
+coat_mini,592.69,431.92,256,224,6.82,33.68,10.34
+vit_small_patch16_384,592.64,431.956,256,384,15.52,50.78,22.2
+regnetz_e8,575.64,444.705,256,320,15.46,63.94,57.7
+convmixer_768_32,563.29,454.463,256,224,19.55,25.95,21.11
+resnest101e,556.07,460.362,256,256,13.38,28.66,48.28
+efficientnetv2_m,555.39,460.921,256,416,18.6,67.5,54.14
+vit_base_r50_s16_224,554.34,461.796,256,224,21.66,35.29,98.66
+vit_small_r26_s32_384,549.37,465.972,256,384,10.43,29.85,36.47
+nf_regnet_b5,533.77,479.592,256,456,11.7,61.95,49.74
+seresnet269d,524.13,488.41,256,256,26.59,53.6,113.67
+nest_base,514.16,497.89,256,224,17.96,53.39,67.72
+jx_nest_base,510.9,501.061,256,224,17.96,53.39,67.72
+resnet200d,509.84,502.105,256,320,31.25,67.33,64.69
+tnt_s_patch16_224,490.75,521.641,256,224,5.24,24.37,23.76
+mixnet_xxl,486.35,526.355,256,224,2.04,23.43,23.96
+tresnet_xl,476.44,537.307,256,224,15.17,15.34,78.44
+convnext_large,476.21,537.564,256,224,34.4,43.13,197.77
+eca_nfnet_l1,475.22,538.689,256,320,14.92,34.42,41.41
+convnext_large_in22ft1k,475.1,538.817,256,224,34.4,43.13,197.77
+efficientnetv2_rw_m,464.7,550.875,256,416,21.49,79.62,53.24
+efficientnet_b5,456.59,560.666,256,456,10.46,98.86,30.39
+xcit_small_12_p8_224,453.25,564.8,256,224,18.69,47.21,26.21
+xcit_small_12_p8_224_dist,452.33,565.946,256,224,18.69,47.21,26.21
+xcit_tiny_24_p8_224,450.37,568.411,256,224,9.21,45.39,12.11
+mixer_l16_224,449.08,570.035,256,224,44.6,41.69,208.2
+xcit_tiny_24_p8_224_dist,448.81,570.379,256,224,9.21,45.39,12.11
+halonet_h1,448.78,570.426,256,256,3.0,51.17,8.1
+vit_large_patch32_384,441.35,580.019,256,384,45.31,43.86,306.63
+resnetrs200,436.51,586.457,256,320,31.51,67.81,93.21
+tf_efficientnet_b5,434.35,589.365,256,456,10.46,98.86,30.39
+tf_efficientnet_b5_ap,433.75,590.18,256,456,10.46,98.86,30.39
+tf_efficientnet_b5_ns,433.48,590.557,256,456,10.46,98.86,30.39
+tf_efficientnetv2_m_in21ft1k,410.56,623.518,256,480,24.76,89.84,54.14
+tf_efficientnetv2_m,410.47,623.661,256,480,24.76,89.84,54.14
+xcit_nano_12_p8_384_dist,410.15,624.154,256,384,6.34,46.08,3.05
+regnety_320,406.99,628.99,256,224,32.34,30.26,145.05
+regnety_160,405.46,631.367,256,288,26.37,38.07,83.59
+xcit_large_24_p16_224_dist,400.4,639.34,256,224,35.86,47.27,189.1
+xcit_large_24_p16_224,400.19,639.684,256,224,35.86,47.27,189.1
+regnetz_d8_evob,388.06,659.67,256,320,6.12,37.08,23.41
+tnt_b_patch16_224,385.58,663.922,256,224,14.09,39.01,65.41
+swin_large_patch4_window7_224,383.73,667.122,256,224,34.53,54.94,196.53
+resnetv2_50x1_bitm,374.41,683.729,256,448,16.62,44.46,25.55
+ssl_resnext101_32x16d,338.57,756.102,256,224,36.27,51.18,194.03
+swsl_resnext101_32x16d,338.32,756.674,256,224,36.27,51.18,194.03
+ig_resnext101_32x16d,338.18,756.969,256,224,36.27,51.18,194.03
+vit_large_patch16_224,326.85,783.223,256,224,61.6,63.52,304.33
+convnext_xlarge_in22ft1k,321.97,795.088,256,224,60.97,57.5,350.2
+tresnet_m_448,321.59,796.023,256,448,22.94,29.21,31.39
+xcit_small_24_p16_384_dist,315.9,810.38,256,384,26.72,68.58,47.67
+crossvit_15_dagger_408,307.27,833.13,256,408,21.45,95.05,28.5
+nasnetalarge,305.09,839.069,256,331,23.89,90.56,88.75
+beit_large_patch16_224,301.03,850.396,256,224,61.6,63.52,304.43
+pnasnet5large,294.97,867.879,256,331,25.04,92.89,86.06
+xcit_tiny_12_p8_384_dist,290.12,882.382,256,384,14.13,69.14,6.71
+ecaresnet269d,282.41,906.484,256,352,50.25,101.25,102.09
+nfnet_f1s,278.33,919.751,256,320,35.97,46.77,132.63
+nfnet_f1,277.93,921.079,256,320,35.97,46.77,132.63
+resnetrs270,274.74,931.762,256,352,51.13,105.48,129.86
+convnext_base_384_in22ft1k,274.29,933.296,256,384,45.2,84.49,88.59
+resnetv2_152x2_bit_teacher,268.11,954.835,256,224,46.95,45.11,236.34
+efficientnet_b6,261.2,490.039,128,528,19.4,167.39,43.04
+regnetx_320,257.17,995.447,256,224,31.81,36.3,107.81
+vit_base_patch16_384,256.27,998.923,256,384,55.54,101.56,86.86
+deit_base_patch16_384,256.23,999.078,256,384,55.54,101.56,86.86
+deit_base_distilled_patch16_384,253.03,1011.71,256,384,55.65,101.82,87.63
+dm_nfnet_f1,251.34,1018.532,256,320,35.97,46.77,132.63
+tf_efficientnet_b6_ns,249.3,513.425,128,528,19.4,167.39,43.04
+tf_efficientnet_b6_ap,249.28,513.468,128,528,19.4,167.39,43.04
+tf_efficientnet_b6,249.27,513.475,128,528,19.4,167.39,43.04
+cait_xxs24_384,240.96,1062.389,256,384,9.63,122.66,12.03
+resnetv2_101x1_bitm,237.63,1077.288,256,448,31.65,64.93,44.54
+xcit_small_24_p8_224,236.59,1082.038,256,224,35.81,90.78,47.63
+xcit_small_24_p8_224_dist,236.51,1082.395,256,224,35.81,90.78,47.63
+efficientnetv2_l,230.67,1109.79,256,480,56.4,157.99,118.52
+xcit_medium_24_p16_384_dist,229.71,1114.433,256,384,47.39,91.64,84.4
+crossvit_18_dagger_408,228.55,560.041,128,408,32.47,124.87,44.61
+tf_efficientnetv2_l,228.05,1122.527,256,480,56.4,157.99,118.52
+eca_nfnet_l2,227.63,1124.623,256,384,30.05,68.28,56.72
+tf_efficientnetv2_l_in21ft1k,227.22,1126.624,256,480,56.4,157.99,118.52
+vit_large_r50_s32_384,223.83,1143.726,256,384,57.43,76.52,329.09
+beit_base_patch16_384,221.26,1157.006,256,384,55.54,101.56,86.74
+regnetz_d8_evos,211.13,1212.504,256,320,7.03,38.92,23.46
+resmlp_big_24_224,198.32,1290.833,256,224,100.23,87.31,129.14
+resmlp_big_24_224_in22ft1k,197.59,1295.599,256,224,100.23,87.31,129.14
+resmlp_big_24_distilled_224,197.33,1297.278,256,224,100.23,87.31,129.14
+resnest200e,191.82,1334.548,256,320,35.69,82.78,70.2
+resnetrs350,182.63,1401.718,256,384,77.59,154.74,163.96
+swin_base_patch4_window12_384,171.61,745.879,128,384,47.19,134.78,87.9
+cait_xs24_384,171.6,1491.866,256,384,19.28,183.98,26.67
+xcit_medium_24_p8_224,171.09,1496.317,256,224,63.53,121.23,84.32
+xcit_medium_24_p8_224_dist,171.08,1496.368,256,224,63.53,121.23,84.32
+vit_base_patch8_224,170.14,1504.628,256,224,78.22,161.69,86.58
+tresnet_l_448,163.98,1561.153,256,448,43.5,47.56,55.99
+convnext_large_384_in22ft1k,162.15,1578.727,256,384,101.09,126.74,197.77
+cait_xxs36_384,161.38,1586.306,256,384,14.35,183.7,17.37
+vit_base_r50_s16_384,161.14,1588.636,256,384,67.43,135.03,98.95
+vit_base_resnet50_384,161.06,1589.462,256,384,67.43,135.03,98.95
+nfnet_f2s,156.52,1635.579,256,352,63.22,79.06,193.78
+efficientnet_b7,156.11,409.945,64,600,38.33,289.94,66.35
+nfnet_f2,155.79,1643.217,256,352,63.22,79.06,193.78
+xcit_tiny_24_p8_384_dist,152.9,1674.228,256,384,27.05,132.95,12.11
+densenet264d_iabn,151.23,1692.723,256,224,13.47,14.0,72.74
+xcit_small_12_p8_384_dist,150.53,1700.627,256,384,54.92,138.29,26.21
+tf_efficientnet_b7_ns,150.44,425.41,64,600,38.33,289.94,66.35
+tf_efficientnet_b7_ap,150.35,425.668,64,600,38.33,289.94,66.35
+tf_efficientnet_b7,150.33,425.703,64,600,38.33,289.94,66.35
+efficientnetv2_xl,146.15,1751.589,256,512,93.85,247.32,208.12
+tf_efficientnetv2_xl_in21ft1k,144.77,1768.278,256,512,93.85,247.32,208.12
+dm_nfnet_f2,141.71,1806.539,256,352,63.22,79.06,193.78
+xcit_large_24_p16_384_dist,138.32,1850.744,256,384,105.35,137.17,189.1
+cait_s24_384,132.94,1925.665,256,384,32.17,245.31,47.06
+resnetrs420,132.69,1929.303,256,416,108.45,213.79,191.89
+vit_huge_patch14_224,131.53,1946.246,256,224,167.4,139.41,632.05
+eca_nfnet_l3,127.14,2013.536,256,448,52.55,118.4,72.04
+tresnet_xl_448,120.79,2119.391,256,448,60.65,61.31,78.44
+efficientnet_cc_b0_8e,117.25,8.519,1,224,0.42,9.42,24.01
+efficientnet_cc_b0_4e,114.89,8.693,1,224,0.41,9.42,13.31
+convnext_xlarge_384_in22ft1k,110.42,1159.152,128,384,179.18,168.99,350.2
+tf_efficientnet_cc_b0_8e,109.13,9.153,1,224,0.42,9.42,24.01
+swin_large_patch4_window12_384,108.07,1184.415,128,384,104.08,202.16,196.74
+tf_efficientnet_cc_b0_4e,106.38,9.39,1,224,0.41,9.42,13.31
+xcit_large_24_p8_224,103.21,2480.29,256,224,141.23,181.56,188.93
+xcit_large_24_p8_224_dist,103.09,2483.315,256,224,141.23,181.56,188.93
+efficientnet_b8,95.86,667.618,64,672,63.48,442.89,87.41
+resnetv2_152x2_bit_teacher_384,93.91,2726.129,256,384,136.16,132.56,236.34
+tf_efficientnet_b8,93.03,687.97,64,672,63.48,442.89,87.41
+tf_efficientnet_b8_ap,92.99,688.234,64,672,63.48,442.89,87.41
+resnetv2_50x3_bitm,90.18,1419.305,128,448,145.7,133.37,217.32
+resnest269e,90.1,2841.126,256,416,77.69,171.98,110.93
+cait_s36_384,88.82,2882.347,256,384,47.99,367.4,68.37
+vit_large_patch16_384,88.44,2894.698,256,384,191.21,270.24,304.72
+vit_giant_patch14_224,87.02,2941.98,256,224,267.18,192.64,1012.61
+nfnet_f3s,85.07,3009.339,256,416,115.58,141.78,254.92
+nfnet_f3,84.65,3024.233,256,416,115.58,141.78,254.92
+efficientnet_cc_b1_8e,83.46,11.971,1,240,0.75,15.44,39.72
+convmixer_1024_20_ks9_p14,81.44,3143.542,256,224,5.55,5.51,24.38
+xcit_small_24_p8_384_dist,78.98,3241.259,256,384,105.24,265.91,47.63
+beit_large_patch16_384,77.03,3323.174,256,384,191.21,270.24,305.0
+dm_nfnet_f3,76.93,3327.777,256,416,115.58,141.78,254.92
+tf_efficientnet_cc_b1_8e,76.09,13.131,1,240,0.75,15.44,39.72
+resnetv2_152x2_bitm,68.83,1859.681,128,448,184.99,180.43,236.34
+xcit_medium_24_p8_384_dist,58.64,4365.679,256,384,186.67,354.73,84.32
+resnetv2_101x3_bitm,55.09,2323.244,128,448,280.33,194.78,387.93
+vit_gigantic_patch14_224,52.92,4837.231,256,224,483.95,275.37,1844.44
+nfnet_f4s,45.21,5662.39,256,512,216.26,262.26,316.07
+nfnet_f4,45.13,5672.08,256,512,216.26,262.26,316.07
+dm_nfnet_f4,41.12,6225.397,256,512,216.26,262.26,316.07
+xcit_large_24_p8_384_dist,35.18,3638.243,128,384,415.0,531.82,188.93
+nfnet_f5s,33.4,7663.628,256,544,290.97,349.71,377.21
+nfnet_f5,33.22,7705.647,256,544,290.97,349.71,377.21
+beit_large_patch16_512,31.23,2049.459,64,512,362.24,656.39,305.67
+cait_m36_384,30.6,8365.416,256,384,173.11,734.81,271.22
+dm_nfnet_f5,30.51,8392.012,256,544,290.97,349.71,377.21
+nfnet_f6s,24.87,10292.837,256,576,378.69,452.2,438.36
+nfnet_f6,24.74,10346.138,256,576,378.69,452.2,438.36
+dm_nfnet_f6,23.05,11108.286,256,576,378.69,452.2,438.36
+nfnet_f7s,19.54,13099.713,256,608,480.39,570.85,499.5
+nfnet_f7,19.47,13146.917,256,608,480.39,570.85,499.5
+resnetv2_152x4_bitm,17.96,3562.952,64,480,844.84,414.26,936.53
+cait_m48_448,13.23,9672.182,128,448,329.41,1708.23,356.46
+convmixer_1536_20,13.15,19467.539,256,224,48.68,33.03,51.63

results/benchmark-train-amp-nchw-pt110-cu113-rtx3090.csv

@@ -0,0 +1,703 @@
+model,train_samples_per_sec,train_step_time,train_batch_size,train_img_size,param_count
+tinynet_e,8788.08,57.613,512,106,2.04
+mobilenetv3_small_050,6451.43,78.69,512,224,1.59
+tf_mobilenetv3_small_minimal_100,6231.46,81.639,512,224,2.04
+levit_128s,5345.65,94.962,512,224,7.78
+lcnet_035,5326.51,95.668,512,224,1.64
+mobilenetv3_small_075,5149.15,98.754,512,224,2.04
+lcnet_050,4879.43,104.466,512,224,1.88
+tf_mobilenetv3_small_075,4839.2,105.1,512,224,2.04
+mobilenetv3_small_100,4741.69,107.315,512,224,2.54
+tf_mobilenetv3_small_100,4474.54,113.738,512,224,2.54
+tinynet_d,4465.05,113.942,512,152,2.34
+mixer_s32_224,3957.52,128.859,512,224,19.1
+levit_128,3868.72,131.359,512,224,9.21
+lcnet_075,3674.09,138.887,512,224,2.36
+vit_small_patch32_224,3657.34,139.252,512,224,22.88
+vit_tiny_r_s16_p8_224,3407.13,149.506,512,224,6.34
+regnetx_002,3372.01,151.204,512,224,2.68
+levit_192,3357.19,151.505,512,224,10.95
+mnasnet_small,3158.42,161.274,512,224,2.03
+regnety_002,3119.2,163.263,512,224,3.16
+lcnet_100,3075.72,165.994,512,224,2.95
+mobilenetv2_035,3007.95,169.496,512,224,1.68
+gernet_s,2901.28,175.82,512,224,8.17
+gluon_resnet18_v1b,2671.91,191.302,512,224,11.69
+swsl_resnet18,2669.15,191.488,512,224,11.69
+ssl_resnet18,2668.71,191.54,512,224,11.69
+resnet18,2662.83,191.947,512,224,11.69
+mobilenetv2_050,2596.25,196.482,512,224,1.97
+levit_256,2559.67,199.033,512,224,18.89
+regnetx_004,2499.65,203.84,512,224,5.16
+mnasnet_050,2494.39,204.524,512,224,2.22
+semnasnet_050,2494.01,204.433,512,224,2.08
+seresnet18,2463.7,207.356,512,224,11.78
+ese_vovnet19b_slim_dw,2437.98,209.545,512,224,1.9
+legacy_seresnet18,2413.99,211.632,512,224,11.78
+tinynet_c,2389.53,213.326,512,184,2.46
+tf_mobilenetv3_large_minimal_100,2278.25,224.06,512,224,3.92
+lcnet_150,2248.73,227.214,512,224,4.5
+mobilenetv3_large_075,2139.42,238.494,512,224,3.99
+ghostnet_050,2137.33,238.296,512,224,2.59
+resnet18d,2046.03,249.892,512,224,11.71
+tf_mobilenetv3_large_075,2027.78,251.65,512,224,3.99
+regnetx_006,2023.49,252.29,512,224,6.2
+ese_vovnet19b_slim,1970.7,259.464,512,224,3.17
+mobilenetv3_rw,1940.58,263.024,512,224,5.48
+mobilenetv3_large_100,1936.66,263.562,512,224,5.48
+mobilenetv3_large_100_miil,1936.5,263.563,512,224,5.48
+mnasnet_075,1879.23,271.718,512,224,3.17
+mobilenetv2_075,1868.15,273.318,512,224,2.64
+tf_efficientnetv2_b0,1853.32,275.137,512,224,7.14
+tf_mobilenetv3_large_100,1838.66,277.647,512,224,5.48
+regnety_004,1819.61,280.334,512,224,4.34
+resnetblur18,1808.21,282.828,512,224,11.69
+semnasnet_075,1791.94,284.855,512,224,2.91
+regnety_006,1780.58,286.558,512,224,6.06
+mobilenetv2_100,1744.9,292.687,512,224,3.5
+skresnet18,1742.33,293.29,512,224,11.96
+mnasnet_b1,1686.64,302.825,512,224,4.38
+mnasnet_100,1686.2,302.919,512,224,4.38
+deit_tiny_patch16_224,1678.5,304.281,512,224,5.72
+vit_tiny_patch16_224,1673.2,305.238,512,224,5.72
+tinynet_b,1669.81,305.621,512,188,3.73
+hardcorenas_a,1656.9,308.354,512,224,5.26
+deit_tiny_distilled_patch16_224,1654.85,308.608,512,224,5.91
+hardcorenas_b,1648.45,309.73,512,224,5.18
+regnetx_008,1647.09,310.092,512,224,7.26
+mnasnet_a1,1641.47,311.08,512,224,3.89
+semnasnet_100,1640.14,311.319,512,224,3.89
+levit_384,1616.42,315.722,512,224,39.13
+gluon_resnet34_v1b,1598.5,319.781,512,224,21.8
+resnet34,1593.87,320.693,512,224,21.8
+tv_resnet34,1592.81,320.898,512,224,21.8
+mixer_b32_224,1579.25,323.449,512,224,60.29
+hardcorenas_c,1569.06,325.387,512,224,5.52
+visformer_tiny,1563.94,326.764,512,224,10.32
+spnasnet_100,1551.19,329.16,512,224,4.42
+pit_ti_distilled_224,1547.24,330.092,512,224,5.1
+pit_ti_224,1546.92,330.186,512,224,4.85
+hardcorenas_d,1543.77,330.549,512,224,7.5
+ghostnet_100,1541.6,330.868,512,224,5.18
+vit_base_patch32_224,1529.64,333.95,512,224,88.22
+vit_base_patch32_224_sam,1527.96,334.33,512,224,88.22
+regnety_008,1525.57,334.693,512,224,6.26
+resmlp_12_224,1513.1,337.634,512,224,15.35
+resmlp_12_distilled_224,1513.05,337.643,512,224,15.35
+seresnet34,1461.69,349.453,512,224,21.96
+nf_regnet_b0,1450.04,351.967,512,256,8.76
+legacy_seresnet34,1428.59,357.543,512,224,21.96
+gernet_m,1427.9,357.926,512,224,21.14
+ese_vovnet19b_dw,1411.33,362.322,512,224,6.54
+nf_resnet26,1408.11,363.145,512,224,16.0
+dla46_c,1390.48,367.526,512,224,1.3
+tinynet_a,1383.86,368.826,512,192,6.19
+efficientnet_lite0,1368.92,373.303,512,224,4.65
+mobilenetv2_110d,1368.66,279.598,384,224,4.52
+rexnetr_100,1367.14,279.805,384,224,4.88
+rexnet_100,1361.11,281.051,384,224,4.8
+resnet34d,1349.98,378.71,512,224,21.82
+xcit_nano_12_p16_224_dist,1340.16,380.375,512,224,3.05
+xcit_nano_12_p16_224,1338.66,380.8,512,224,3.05
+hardcorenas_f,1338.06,381.608,512,224,8.2
+selecsls42,1328.24,384.874,512,224,30.35
+ghostnet_130,1326.71,384.661,512,224,7.36
+selecsls42b,1324.18,386.065,512,224,32.46
+mixer_s16_224,1308.0,390.939,512,224,18.53
+resnet26,1305.0,391.896,512,224,16.0
+tf_efficientnet_lite0,1300.06,393.134,512,224,4.65
+hardcorenas_e,1287.59,396.566,512,224,8.07
+hrnet_w18_small,1275.09,400.255,512,224,13.19
+pit_xs_distilled_224,1245.81,410.186,512,224,11.0
+pit_xs_224,1244.79,410.521,512,224,10.62
+mnasnet_140,1243.07,411.14,512,224,7.12
+fbnetc_100,1241.42,411.527,512,224,5.57
+mobilenetv2_140,1236.2,309.893,384,224,6.11
+vit_small_patch32_384,1210.76,422.138,512,384,22.92
+efficientnet_b0,1207.97,316.872,384,224,5.29
+semnasnet_140,1199.21,426.076,512,224,6.11
+nf_seresnet26,1183.4,432.047,512,224,17.4
+tf_efficientnetv2_b1,1182.26,323.455,384,240,8.14
+vit_tiny_r_s16_p8_384,1176.2,325.707,384,384,6.36
+repvgg_b0,1166.7,437.858,512,224,15.82
+gmixer_12_224,1164.26,439.045,512,224,12.7
+tf_efficientnet_b0_ns,1155.19,331.423,384,224,5.29
+tf_efficientnet_b0,1155.11,331.42,384,224,5.29
+tf_efficientnet_b0_ap,1154.21,331.702,384,224,5.29
+efficientnet_b1_pruned,1141.45,447.176,512,240,6.33
+selecsls60,1132.77,451.165,512,224,30.67
+selecsls60b,1130.1,452.225,512,224,32.77
+resnet26d,1110.35,460.655,512,224,16.01
+nf_ecaresnet26,1108.5,461.384,512,224,16.0
+rexnetr_130,1084.92,234.909,256,224,7.61
+mixnet_s,1079.62,473.081,512,224,4.13
+dla34,1055.21,484.649,512,224,15.74
+dla46x_c,1044.64,489.413,512,224,1.07
+rexnet_130,1038.12,245.54,256,224,7.56
+regnetx_016,1027.51,497.437,512,224,9.19
+mobilenetv2_120d,1026.61,248.294,256,224,5.83
+ecaresnet50d_pruned,1019.51,501.3,512,224,19.94
+tf_mixnet_s,1018.74,501.408,512,224,4.13
+skresnet34,1008.77,506.519,512,224,22.28
+dla60x_c,999.33,511.452,512,224,1.32
+gernet_l,994.19,514.18,512,256,31.08
+xcit_tiny_12_p16_224,974.89,523.578,512,224,6.72
+xcit_tiny_12_p16_224_dist,974.56,523.762,512,224,6.72
+rexnetr_150,962.03,265.059,256,224,9.78
+crossvit_tiny_240,946.09,404.397,384,240,7.01
+crossvit_9_240,942.0,406.34,384,240,8.55
+rexnet_150,920.27,277.114,256,224,9.73
+repvgg_a2,916.8,557.644,512,224,28.21
+legacy_seresnext26_32x4d,911.29,561.256,512,224,16.79
+efficientnet_lite1,907.6,281.167,256,240,5.42
+crossvit_9_dagger_240,907.56,421.794,384,240,8.78
+regnety_016,905.38,563.868,512,224,11.2
+convit_tiny,905.04,423.366,384,224,5.71
+gmlp_ti16_224,895.37,427.435,384,224,5.87
+efficientnet_es_pruned,894.51,571.683,512,224,5.44
+efficientnet_es,894.06,571.94,512,224,5.44
+resnest14d,890.17,574.739,512,224,10.61
+deit_small_patch16_224,879.08,436.051,384,224,22.05
+vit_small_patch16_224,877.74,436.745,384,224,22.05
+tf_efficientnet_es,877.36,582.875,512,224,5.44
+resnetv2_50,874.29,584.904,512,224,25.55
+resnext26ts,872.05,439.874,384,256,10.3
+tf_efficientnet_lite1,870.75,293.088,256,240,5.42
+resnet26t,867.6,589.639,512,256,16.01
+tf_efficientnetv2_b2,866.58,294.02,256,260,10.1
+deit_small_distilled_patch16_224,865.21,443.046,384,224,22.44
+nf_regnet_b1,863.6,591.527,512,288,10.22
+nf_regnet_b2,857.0,595.977,512,272,14.31
+eca_resnext26ts,843.14,303.098,256,256,10.3
+seresnext26ts,838.62,304.653,256,256,10.39
+resnet50,826.48,463.878,384,224,25.56
+ssl_resnet50,826.14,464.04,384,224,25.56
+botnet26t_256,825.99,464.363,384,256,12.49
+tv_resnet50,825.99,464.131,384,224,25.56
+swsl_resnet50,825.61,464.367,384,224,25.56
+gluon_resnet50_v1b,825.26,464.548,384,224,25.56
+gcresnext26ts,820.94,311.077,256,256,10.48
+seresnext26t_32x4d,819.08,468.216,384,224,16.81
+seresnext26tn_32x4d,818.65,468.46,384,224,16.81
+seresnext26d_32x4d,816.35,469.756,384,224,16.81
+pit_s_224,814.17,313.637,256,224,23.46
+pit_s_distilled_224,813.41,313.92,256,224,24.04
+vgg11,798.47,641.065,512,224,132.86
+halonet26t,796.75,481.425,384,256,12.48
+eca_botnext26ts_256,794.14,321.819,256,256,10.59
+resnetv2_50t,783.96,652.33,512,224,25.57
+vit_small_resnet26d_224,783.75,488.981,384,224,63.61
+ecaresnext26t_32x4d,782.53,490.179,384,224,15.41
+ecaresnext50t_32x4d,782.45,490.233,384,224,15.41
+resnetv2_50d,781.21,654.649,512,224,25.57
+tresnet_m,773.27,659.381,512,224,31.39
+fbnetv3_d,771.82,329.985,256,256,10.31
+eca_halonext26ts,770.01,331.906,256,256,10.76
+cspresnet50,766.64,500.094,384,256,21.62
+fbnetv3_b,765.95,499.824,384,256,8.6
+ecaresnet101d_pruned,765.08,667.536,512,224,24.88
+vovnet39a,764.37,669.255,512,224,22.6
+gluon_resnet50_v1c,761.03,503.811,384,224,25.58
+cspresnext50,757.53,506.115,384,224,20.57
+efficientnet_cc_b0_4e,755.85,506.911,384,224,13.31
+ecaresnetlight,754.13,678.075,512,224,30.16
+efficientnet_cc_b0_8e,750.91,510.273,384,224,24.01
+mixnet_m,749.54,510.912,384,224,5.01
+efficientnet_b2_pruned,748.23,340.751,256,260,8.31
+resnet50t,745.06,514.605,384,224,25.57
+resmlp_24_distilled_224,744.01,342.726,256,224,30.02
+gluon_resnet50_v1d,743.98,515.37,384,224,25.58
+resnet50d,743.76,515.521,384,224,25.58
+resmlp_24_224,743.43,342.965,256,224,30.02
+legacy_seresnet50,739.53,518.192,384,224,28.09
+resnet32ts,736.71,346.942,256,256,17.96
+ese_vovnet39b,733.2,523.109,384,224,24.57
+tf_efficientnet_cc_b0_4e,732.92,522.856,384,224,13.31
+tf_efficientnet_cc_b0_8e,731.1,524.143,384,224,24.01
+dpn68b,728.18,526.241,384,224,12.61
+resnet33ts,727.42,351.376,256,256,19.68
+selecsls84,719.68,710.264,512,224,50.95
+nf_seresnet50,718.4,533.452,384,224,28.09
+rexnetr_200,715.44,267.313,192,224,16.52
+visformer_small,713.57,537.5,384,224,40.22
+res2net50_48w_2s,712.35,538.286,384,224,25.29
+tf_mixnet_m,711.23,538.464,384,224,5.01
+dpn68,711.07,539.016,384,224,12.61
+seresnet50,710.55,539.38,384,224,28.09
+lambda_resnet26t,708.6,541.35,384,256,10.96
+bat_resnext26ts,706.1,361.453,256,256,10.73
+tf_efficientnet_b1,705.85,361.254,256,240,7.79
+tf_efficientnet_b1_ns,705.53,361.468,256,240,7.79
+tf_efficientnet_b1_ap,705.49,361.504,256,240,7.79
+seresnet33ts,700.84,364.51,256,256,19.78
+eca_resnet33ts,700.22,364.953,256,256,19.68
+resnetblur50,694.99,551.777,384,224,25.56
+cspresnet50d,694.84,551.823,384,256,21.64
+rexnet_200,693.57,275.752,192,224,16.37
+cspresnet50w,686.88,558.19,384,256,28.12
+eca_vovnet39b,683.56,748.398,512,224,22.6
+densenet121,682.9,373.209,256,224,7.98
+dla60,682.79,561.488,384,224,22.04
+efficientnet_b1,682.7,373.555,256,256,7.79
+tv_densenet121,681.14,374.178,256,224,7.98
+gcresnet33ts,680.99,374.966,256,256,19.88
+efficientnet_lite2,680.43,375.329,256,260,6.09
+resnest26d,672.85,570.012,384,224,17.07
+resnetaa50d,669.52,572.758,384,224,25.58
+vgg11_bn,669.19,382.326,256,224,132.87
+gluon_resnet50_v1s,664.71,576.903,384,224,25.68
+nf_ecaresnet50,663.54,577.843,384,224,25.56
+hrnet_w18_small_v2,663.05,769.959,512,224,15.6
+tf_efficientnet_lite2,656.12,389.28,256,260,6.09
+vit_small_r26_s32_224,654.12,390.216,256,224,36.43
+lambda_resnet26rpt_256,650.73,196.144,128,256,10.99
+vit_base_resnet26d_224,650.56,589.099,384,224,101.4
+seresnet50t,645.62,395.439,256,224,28.1
+gluon_resnext50_32x4d,643.1,596.344,384,224,25.03
+ssl_resnext50_32x4d,642.69,596.753,384,224,25.03
+resnext50_32x4d,642.56,596.871,384,224,25.03
+tv_resnext50_32x4d,642.26,597.116,384,224,25.03
+swsl_resnext50_32x4d,642.22,597.166,384,224,25.03
+haloregnetz_b,639.45,398.971,256,224,11.68
+resnetrs50,639.32,599.548,384,224,35.69
+densenet121d,637.87,399.654,256,224,8.0
+regnetx_032,635.48,603.11,384,224,15.3
+resnetblur50d,634.6,402.62,256,224,25.58
+res2net50_26w_4s,625.68,407.946,256,224,25.7
+skresnet50,620.44,411.364,256,224,25.8
+ese_vovnet57b,618.38,620.149,384,224,38.61
+densenetblur121d,611.64,416.815,256,224,8.0
+ecaresnet50d,601.62,424.614,256,224,25.58
+adv_inception_v3,592.83,646.406,384,299,23.83
+resnext50d_32x4d,590.58,432.684,256,224,25.05
+gluon_inception_v3,589.45,650.157,384,299,23.83
+inception_v3,588.55,651.172,384,299,23.83
+tf_inception_v3,588.36,651.336,384,299,23.83
+seresnetaa50d,587.38,434.758,256,224,28.11
+sehalonet33ts,585.23,436.688,256,256,13.69
+coat_lite_tiny,580.33,660.876,384,224,5.72
+gmixer_24_224,577.9,441.611,256,224,24.72
+gcresnet50t,577.25,442.093,256,256,25.9
+efficientnet_b3_pruned,576.96,442.1,256,300,9.86
+mixnet_l,576.91,442.34,256,224,7.33
+resnetv2_50x1_bit_distilled,575.09,333.113,192,224,25.55
+skresnet50d,572.85,445.576,256,224,25.82
+vovnet57a,569.57,898.133,512,224,36.64
+res2next50,568.73,448.924,256,224,24.67
+seresnext50_32x4d,567.85,449.777,256,224,27.56
+gluon_seresnext50_32x4d,567.27,450.224,256,224,27.56
+cspresnext50_iabn,565.5,676.906,384,256,20.57
+legacy_seresnext50_32x4d,565.44,451.632,256,224,27.56
+resnest50d_1s4x24d,561.18,454.967,256,224,25.68
+res2net50_14w_8s,560.86,454.355,256,224,25.06
+repvgg_b1g4,560.47,912.475,512,224,39.97
+densenet169,558.29,456.216,256,224,14.15
+convnext_tiny,558.25,457.708,256,224,28.59
+convnext_tiny_hnf,555.94,459.59,256,224,28.59
+coat_lite_mini,554.95,691.16,384,224,11.01
+regnety_040,554.71,690.835,384,224,20.65
+dla60_res2net,550.65,463.555,256,224,20.85
+darknet53,549.0,465.546,256,256,41.61
+vgg13,548.68,699.681,384,224,133.05
+tf_mixnet_l,547.03,466.516,256,224,7.33
+dla60x,544.8,469.001,256,224,17.35
+efficientnet_em,542.04,707.524,384,240,6.9
+regnetx_040,541.76,707.76,384,224,22.12
+nf_resnet101,541.63,707.493,384,224,44.55
+tf_efficientnet_em,539.79,473.368,256,240,6.9
+nf_regnet_b3,535.57,476.345,256,320,18.59
+resnetv2_101,535.41,476.74,256,224,44.54
+resnetv2_50d_gn,535.21,357.946,192,224,25.57
+tf_efficientnet_b2_ns,534.14,358.079,192,260,9.11
+tf_efficientnet_b2,533.77,358.295,192,260,9.11
+tf_efficientnet_b2_ap,533.7,358.382,192,260,9.11
+gcresnext50ts,533.26,358.674,192,256,15.67
+tf_efficientnetv2_b3,532.84,358.667,192,300,14.36
+nf_resnet50,527.83,726.718,384,288,25.56
+sebotnet33ts_256,526.55,242.351,128,256,13.7
+xcit_small_12_p16_224,524.79,486.162,256,224,26.25
+xcit_small_12_p16_224_dist,524.65,486.278,256,224,26.25
+resnet50_gn,521.4,367.482,192,224,25.56
+efficientnetv2_rw_t,520.2,367.035,192,288,13.65
+resnet101,518.79,491.983,256,224,44.55
+tv_resnet101,518.57,492.187,256,224,44.55
+gluon_resnet101_v1b,518.13,492.648,256,224,44.55
+crossvit_small_240,516.01,370.633,192,240,26.86
+twins_svt_small,514.62,496.02,256,224,24.06
+vit_base_r26_s32_224,510.65,374.824,192,224,101.38
+xcit_tiny_24_p16_224_dist,510.27,498.69,256,224,12.12
+xcit_tiny_24_p16_224,510.13,498.896,256,224,12.12
+gmlp_s16_224,504.22,379.373,192,224,19.42
+resnetv2_101d,500.13,510.457,256,224,44.56
+mixer_b16_224,499.95,511.32,256,224,59.88
+mixer_b16_224_miil,499.11,512.204,256,224,59.88
+vit_base_resnet50d_224,497.25,513.299,256,224,110.97
+dla60_res2next,494.84,516.012,256,224,17.03
+gluon_resnet101_v1c,492.43,518.398,256,224,44.57
+vit_base_patch32_384,491.61,519.97,256,384,88.3
+repvgg_b1,491.0,1041.743,512,224,57.42
+resmlp_36_224,490.1,389.77,192,224,44.69
+resmlp_36_distilled_224,490.1,389.707,192,224,44.69
+efficientnet_b2,489.47,390.855,192,288,9.11
+efficientnet_b2a,489.25,391.025,192,288,9.11
+vit_large_patch32_224,487.93,523.265,256,224,306.54
+gluon_resnet101_v1d,484.72,526.651,256,224,44.57
+gc_efficientnetv2_rw_t,482.91,395.05,192,288,13.68
+wide_resnet50_2,482.59,529.718,256,224,68.88
+mixer_l32_224,479.14,399.369,192,224,206.94
+cait_xxs24_224,478.85,532.347,256,224,11.96
+cspdarknet53,478.49,534.051,256,256,27.64
+efficientnet_cc_b1_8e,478.04,534.008,256,240,39.72
+crossvit_15_240,474.46,403.054,192,240,27.53
+resnest50d,473.17,539.784,256,224,27.48
+res2net50_26w_6s,470.73,542.208,256,224,37.05
+ecaresnet26t,470.49,543.574,256,320,16.01
+dla102,468.81,544.602,256,224,33.27
+regnetz_b16,467.13,409.622,192,288,9.72
+xcit_nano_12_p16_384_dist,466.59,547.027,256,384,3.05
+tf_efficientnet_cc_b1_8e,466.17,547.634,256,240,39.72
+cspdarknet53_iabn,464.37,824.482,384,256,27.64
+crossvit_15_dagger_240,458.06,417.527,192,240,28.21
+regnetx_080,454.82,843.271,384,224,39.57
+vgg13_bn,453.43,564.307,256,224,133.05
+halonet50ts,453.0,422.945,192,256,22.73
+lambda_resnet50ts,451.97,423.868,192,256,21.54
+legacy_seresnet101,451.72,564.699,256,224,49.33
+resnetaa101d,451.69,565.275,256,224,44.57
+gluon_resnet101_v1s,450.48,566.841,256,224,44.67
+vgg16,449.51,854.057,384,224,138.36
+densenet201,445.52,428.155,192,224,20.01
+swin_tiny_patch4_window7_224,443.41,576.481,256,224,28.29
+seresnet101,441.23,578.213,256,224,49.33
+nf_seresnet101,438.29,581.969,256,224,49.33
+resnetblur101d,436.36,585.216,256,224,44.57
+mixnet_xl,424.84,450.181,192,224,11.9
+twins_pcpvt_small,421.68,605.595,256,224,24.11
+nfnet_l0,417.1,612.673,256,288,35.07
+botnet50ts_256,413.47,308.688,128,256,22.74
+skresnext50_32x4d,412.36,619.54,256,224,27.48
+gluon_resnext101_32x4d,412.2,619.638,256,224,44.18
+swsl_resnext101_32x4d,412.03,619.892,256,224,44.18
+resnext101_32x4d,411.85,620.12,256,224,44.18
+ssl_resnext101_32x4d,411.66,620.437,256,224,44.18
+vit_tiny_patch16_384,409.64,311.71,128,384,5.79
+convit_small,402.9,475.62,192,224,27.78
+halo2botnet50ts_256,398.01,320.711,128,256,22.64
+res2net101_26w_4s,397.75,480.37,192,224,45.21
+efficientnet_lite3,397.55,320.952,128,300,8.2
+tresnet_l,396.67,1287.459,512,224,55.99
+eca_nfnet_l0,396.55,644.544,256,288,24.14
+resnet51q,393.31,487.375,192,288,35.7
+nf_ecaresnet101,390.17,654.38,256,224,44.55
+dla102x,386.87,494.82,192,224,26.31
+res2net50_26w_8s,386.73,494.414,192,224,48.4
+lamhalobotnet50ts_256,386.04,330.65,128,256,22.57
+regnetx_064,385.03,664.106,256,224,26.21
+regnety_032,384.2,665.001,256,288,19.44
+tf_efficientnet_lite3,383.37,332.84,128,300,8.2
+vgg19,381.34,1006.734,384,224,143.67
+xception,378.7,337.254,128,299,22.86
+vgg16_bn,376.28,680.016,256,224,138.37
+ecaresnet101d,375.12,680.773,256,224,44.57
+resnetv2_152,373.91,682.585,256,224,60.19
+resnest50d_4s2x40d,372.15,686.672,256,224,30.42
+regnety_064,370.51,689.324,256,224,30.58
+repvgg_b2g4,370.32,1381.565,512,224,61.76
+gluon_resnet152_v1b,366.03,522.463,192,224,60.19
+tv_resnet152,365.47,523.229,192,224,60.19
+resnet152,365.43,523.289,192,224,60.19
+convnext_small,365.2,524.035,192,224,50.22
+ese_vovnet99b,363.8,702.311,256,224,63.2
+xcit_nano_12_p8_224,362.09,705.393,256,224,3.05
+xcit_nano_12_p8_224_dist,361.89,705.792,256,224,3.05
+gluon_seresnext101_32x4d,361.32,529.345,192,224,48.96
+seresnext101_32x4d,361.3,529.348,192,224,48.96
+legacy_seresnext101_32x4d,359.71,531.693,192,224,48.96
+nfnet_f0,359.11,711.802,256,256,71.49
+vit_base_patch16_224_miil,357.29,357.529,128,224,86.54
+nest_tiny,356.86,357.893,128,224,17.06
+nfnet_f0s,355.27,719.527,256,256,71.49
+resnet61q,355.26,359.368,128,288,36.85
+resnetv2_152d,355.12,538.597,192,224,60.2
+jx_nest_tiny,353.84,360.954,128,224,17.06
+vit_small_resnet50d_s16_224,353.26,542.366,192,224,57.53
+hrnet_w18,352.9,721.041,256,224,21.3
+deit_base_patch16_224,352.4,362.443,128,224,86.57
+gluon_resnet152_v1c,352.05,543.263,192,224,60.21
+vit_base_patch16_224_sam,351.83,363.047,128,224,86.57
+vit_base_patch16_224,351.68,363.189,128,224,86.57
+ese_vovnet99b_iabn,349.3,1096.435,384,224,63.2
+gluon_resnet152_v1d,348.27,549.187,192,224,60.21
+deit_base_distilled_patch16_224,347.05,368.041,128,224,87.34
+crossvit_18_240,345.05,369.11,128,240,43.27
+xcit_tiny_12_p16_384_dist,335.91,569.966,192,384,6.72
+fbnetv3_g,335.24,379.926,128,288,16.62
+regnety_080,335.22,762.533,256,224,39.18
+crossvit_18_dagger_240,333.95,381.492,128,240,44.27
+repvgg_b2,331.54,1543.333,512,224,89.02
+gluon_resnet152_v1s,330.47,578.863,192,224,60.32
+densenet161,327.93,388.08,128,224,28.68
+dla169,324.94,588.567,192,224,53.39
+beit_base_patch16_224,323.79,394.303,128,224,86.53
+coat_lite_small,323.43,592.328,192,224,19.84
+vgg19_bn,322.88,792.477,256,224,143.68
+dm_nfnet_f0,322.48,792.711,256,256,71.49
+repvgg_b3g4,313.61,1223.425,384,224,83.83
+tf_efficientnet_b3,313.56,406.655,128,300,12.23
+tf_efficientnet_b3_ns,313.33,406.959,128,300,12.23
+tf_efficientnet_b3_ap,313.29,407.029,128,300,12.23
+legacy_seresnet152,311.06,614.195,192,224,66.82
+ese_vovnet39b_evos,310.12,411.897,128,224,24.58
+dpn92,308.99,827.182,256,224,37.67
+efficientnet_b3,306.79,415.635,128,320,12.23
+efficientnet_b3a,306.46,416.067,128,320,12.23
+cait_xxs36_224,306.42,414.392,128,224,17.3
+inception_v4,305.84,625.702,192,299,42.68
+twins_pcpvt_base,304.9,627.165,192,224,43.83
+regnetx_120,304.64,839.453,256,224,46.11
+ecaresnet50t,303.47,420.896,128,320,25.57
+hrnet_w32,302.01,631.305,192,224,41.23
+seresnet152,300.58,422.864,128,224,66.82
+hrnet_w30,299.22,851.13,256,224,37.71
+convmixer_1024_20_ks9_p14,299.07,855.183,256,224,24.38
+regnetz_c16,296.99,429.626,128,320,13.46
+twins_svt_base,296.14,430.333,128,224,56.07
+tresnet_xl,289.45,1323.227,384,224,78.44
+swin_small_patch4_window7_224,288.11,442.639,128,224,49.61
+xception41,287.58,443.922,128,299,26.97
+regnety_120,286.74,668.36,192,224,51.82
+pit_b_224,286.53,334.205,96,224,73.76
+pit_b_distilled_224,285.98,334.794,96,224,74.79
+nf_regnet_b4,282.38,451.32,128,384,30.21
+wide_resnet101_2,281.79,679.914,192,224,126.89
+mixnet_xxl,275.31,463.133,128,224,23.96
+resnetv2_50d_evob,273.52,700.95,192,224,25.59
+xcit_small_24_p16_224_dist,271.4,468.703,128,224,47.67
+xcit_small_24_p16_224,271.18,469.033,128,224,47.67
+repvgg_b3,269.71,1422.719,384,224,123.09
+convnext_base,268.96,474.28,128,224,88.59
+convnext_base_in22ft1k,268.69,474.728,128,224,88.59
+resnetv2_50d_evos,264.08,483.67,128,224,25.59
+resnext101_64x4d,263.45,484.429,128,224,83.46
+gluon_resnext101_64x4d,263.28,484.77,128,224,83.46
+cait_s24_224,262.8,484.707,128,224,46.92
+dla102x2,261.95,487.178,128,224,41.28
+resnet200,261.21,487.362,128,224,64.67
+xcit_tiny_12_p8_224_dist,256.95,496.557,128,224,6.71
+xcit_tiny_12_p8_224,256.64,497.116,128,224,6.71
+regnetx_160,256.42,747.752,192,224,54.28
+inception_resnet_v2,251.21,761.158,192,299,55.84
+ens_adv_inception_resnet_v2,251.14,761.388,192,299,55.84
+resnetrs101,248.83,512.362,128,288,63.62
+efficientnet_el,248.41,514.229,128,300,10.59
+efficientnet_el_pruned,248.32,514.423,128,300,10.59
+swsl_resnext101_32x8d,247.35,516.019,128,224,88.79
+ssl_resnext101_32x8d,247.27,516.22,128,224,88.79
+ig_resnext101_32x8d,247.19,516.409,128,224,88.79
+resnext101_32x8d,247.05,516.7,128,224,88.79
+dpn98,245.05,520.932,128,224,61.57
+tf_efficientnet_el,244.21,523.119,128,300,10.59
+efficientnetv2_s,241.82,394.917,96,384,21.46
+tf_efficientnetv2_s_in21ft1k,240.69,396.775,96,384,21.46
+tf_efficientnetv2_s,239.59,398.639,96,384,21.46
+gluon_seresnext101_64x4d,239.24,533.021,128,224,88.23
+gmlp_b16_224,238.08,401.808,96,224,73.08
+resnet101d,236.98,538.626,128,320,44.57
+nest_small,234.62,407.715,96,224,38.35
+resnest101e,234.0,544.633,128,256,48.28
+jx_nest_small,233.02,410.534,96,224,38.35
+crossvit_base_240,230.08,415.793,96,240,105.03
+efficientnetv2_rw_s,228.99,277.434,64,384,23.94
+twins_svt_large,227.78,560.084,128,224,99.27
+coat_tiny,226.98,562.116,128,224,5.5
+seresnext101_32x8d,225.04,566.768,128,224,93.57
+vit_large_r50_s32_224,220.85,432.537,96,224,328.99
+swin_base_patch4_window7_224,220.19,579.694,128,224,87.77
+gluon_xception65,220.18,434.23,96,299,39.92
+xception65,219.22,436.124,96,299,39.92
+hrnet_w40,218.29,875.173,192,224,57.56
+twins_pcpvt_large,217.89,583.814,128,224,60.99
+vit_small_r26_s32_384,215.21,296.22,64,384,36.47
+convit_base,210.68,454.737,96,224,86.54
+vit_small_patch16_384,209.29,305.058,64,384,22.2
+hrnet_w44,203.55,938.867,192,224,67.06
+hrnet_w48,203.33,625.151,128,224,77.47
+tresnet_m_448,203.05,943.084,192,448,31.39
+efficientnet_lite4,198.26,321.517,64,380,13.01
+xcit_medium_24_p16_224,197.63,482.811,96,224,84.4
+xcit_medium_24_p16_224_dist,197.58,482.902,96,224,84.4
+vit_base_r50_s16_224,195.39,326.034,64,224,98.66
+densenet264,194.98,488.653,96,224,72.69
+regnetz_d8,194.31,327.837,64,320,23.37
+tf_efficientnet_lite4,192.8,330.742,64,380,13.01
+coat_mini,192.49,663.128,128,224,10.34
+eca_nfnet_l1,186.17,685.824,128,320,41.41
+regnetz_d32,182.61,348.904,64,320,27.58
+dpn131,182.53,699.301,128,224,79.25
+xcit_small_12_p16_384_dist,180.8,529.313,96,384,26.25
+nest_base,176.01,362.129,64,224,67.72
+xcit_tiny_24_p16_384_dist,175.08,545.374,96,384,12.12
+jx_nest_base,174.7,364.806,64,224,67.72
+dpn107,171.15,746.299,128,224,86.92
+resnet152d,170.12,562.135,96,320,60.21
+hrnet_w64,167.69,758.827,128,224,128.06
+efficientnet_b4,167.63,379.88,64,384,19.34
+xception71,165.9,383.797,64,299,42.34
+densenet264d_iabn,163.71,1168.658,192,224,72.74
+regnety_320,163.11,783.43,128,224,145.05
+halonet_h1,161.83,394.259,64,256,8.1
+convnext_large,159.74,599.371,96,224,197.77
+convnext_large_in22ft1k,159.74,599.406,96,224,197.77
+tnt_s_patch16_224,158.13,807.71,128,224,23.76
+tf_efficientnet_b4,157.41,404.653,64,380,19.34
+tf_efficientnet_b4_ap,157.35,404.822,64,380,19.34
+tf_efficientnet_b4_ns,157.32,404.855,64,380,19.34
+mixer_l16_224,155.84,614.71,96,224,208.2
+seresnet200d,155.16,614.74,96,256,71.86
+ecaresnet200d,149.12,640.536,96,256,64.69
+vit_large_patch32_384,146.32,435.992,64,384,306.63
+regnetx_320,143.56,890.603,128,224,107.81
+seresnet152d,141.0,450.755,64,320,66.84
+resnetrs152,139.45,455.893,64,320,86.62
+resnetv2_50x1_bitm,138.96,344.659,48,448,25.55
+gluon_senet154,138.12,691.987,96,224,115.09
+senet154,138.03,692.329,96,224,115.09
+xcit_small_12_p8_224_dist,137.35,464.322,64,224,26.21
+xcit_small_12_p8_224,137.29,464.503,64,224,26.21
+legacy_senet154,136.93,698.045,96,224,115.09
+regnety_160,133.17,719.682,96,288,83.59
+xcit_tiny_24_p8_224,132.53,721.469,96,224,12.11
+xcit_tiny_24_p8_224_dist,132.41,722.018,96,224,12.11
+swin_large_patch4_window7_224,132.32,482.098,64,224,196.53
+tnt_b_patch16_224,126.43,757.574,96,224,65.41
+resnet200d,123.09,517.162,64,320,64.69
+nfnet_f1,122.2,783.645,96,320,132.63
+xcit_nano_12_p8_384_dist,122.19,522.144,64,384,3.05
+nfnet_f1s,121.0,791.465,96,320,132.63
+xcit_large_24_p16_224_dist,119.24,533.65,64,224,189.1
+xcit_large_24_p16_224,119.19,534.033,64,224,189.1
+efficientnetv2_m,117.91,404.134,48,416,54.14
+seresnet269d,115.13,550.403,64,256,113.67
+nf_regnet_b5,109.6,581.496,64,456,49.74
+dm_nfnet_f1,109.28,583.486,64,320,132.63
+vit_large_patch16_224,106.61,448.786,48,224,304.33
+convnext_xlarge_in22ft1k,104.28,612.171,64,224,350.2
+crossvit_15_dagger_408,103.01,308.973,32,408,28.5
+efficientnetv2_rw_m,101.82,311.114,32,416,53.24
+regnetz_e8,100.87,473.939,48,320,57.7
+tresnet_l_448,99.64,1281.316,128,448,55.99
+beit_large_patch16_224,98.41,485.874,48,224,304.43
+resnetrs200,97.6,487.799,48,320,93.21
+ig_resnext101_32x16d,96.96,988.697,96,224,194.03
+swsl_resnext101_32x16d,96.88,989.444,96,224,194.03
+ssl_resnext101_32x16d,96.83,989.971,96,224,194.03
+xcit_small_24_p16_384_dist,92.8,514.221,48,384,47.67
+eca_nfnet_l2,92.75,687.481,64,384,56.72
+convnext_base_384_in22ft1k,92.09,519.541,48,384,88.59
+deit_base_patch16_384,90.25,353.802,32,384,86.86
+vit_base_patch16_384,90.2,353.987,32,384,86.86
+deit_base_distilled_patch16_384,88.49,360.855,32,384,87.63
+tf_efficientnetv2_m,88.2,359.811,32,480,54.14
+tf_efficientnetv2_m_in21ft1k,88.09,360.345,32,480,54.14
+xcit_tiny_12_p8_384_dist,87.2,548.793,48,384,6.71
+cait_xxs24_384,86.38,368.153,32,384,12.03
+resnetv2_101x1_bitm,85.15,374.386,32,448,44.54
+resnetv2_152x2_bit_teacher,85.01,374.341,32,224,236.34
+convmixer_768_32,84.46,1135.466,96,224,21.11
+resnest200e,77.86,611.828,48,320,70.2
+crossvit_18_dagger_408,77.59,307.496,24,408,44.61
+beit_base_patch16_384,76.84,311.346,24,384,86.74
+tresnet_xl_448,75.89,1261.775,96,448,78.44
+efficientnet_b5,75.33,422.385,32,456,30.39
+vit_large_r50_s32_384,73.95,430.573,32,384,329.09
+tf_efficientnet_b5_ap,73.87,430.874,32,456,30.39
+tf_efficientnet_b5_ns,73.84,431.06,32,456,30.39
+tf_efficientnet_b5,73.76,431.443,32,456,30.39
+regnetz_d8_evob,69.9,684.907,48,320,23.41
+nfnet_f2,69.7,685.799,48,352,193.78
+xcit_small_24_p8_224,69.53,457.315,32,224,47.63
+xcit_small_24_p8_224_dist,69.49,457.517,32,224,47.63
+nfnet_f2s,69.09,691.894,48,352,193.78
+xcit_medium_24_p16_384_dist,67.17,473.231,32,384,84.4
+resmlp_big_24_distilled_224,65.96,483.773,32,224,129.14
+resmlp_big_24_224,65.95,483.848,32,224,129.14
+resmlp_big_24_224_in22ft1k,65.95,483.866,32,224,129.14
+regnetz_d8_evos,64.24,496.339,32,320,23.46
+dm_nfnet_f2,62.49,765.031,48,352,193.78
+ecaresnet269d,60.75,522.332,32,352,102.09
+swin_base_patch4_window12_384,60.1,397.741,24,384,87.9
+cait_xs24_384,59.89,398.428,24,384,26.67
+pnasnet5large,59.35,535.626,32,331,86.06
+resnetrs270,58.4,542.338,32,352,129.86
+vit_base_patch8_224,58.07,274.782,16,224,86.58
+nasnetalarge,57.07,556.194,32,331,88.75
+cait_xxs36_384,56.42,422.115,24,384,17.37
+convnext_large_384_in22ft1k,54.47,585.847,32,384,197.77
+vit_base_r50_s16_384,54.27,293.3,16,384,98.95
+vit_base_resnet50_384,54.19,293.736,16,384,98.95
+xcit_medium_24_p8_224,51.72,615.776,32,224,84.32
+xcit_medium_24_p8_224_dist,51.69,616.066,32,224,84.32
+ig_resnext101_32x32d,51.67,617.889,32,224,468.53
+eca_nfnet_l3,49.71,640.359,32,448,72.04
+convmixer_1536_20,48.09,997.399,48,224,51.63
+xcit_small_12_p8_384_dist,46.72,512.057,24,384,26.21
+efficientnetv2_l,46.55,339.739,16,480,118.52
+tf_efficientnetv2_l_in21ft1k,46.43,340.391,16,480,118.52
+tf_efficientnetv2_l,46.41,340.685,16,480,118.52
+xcit_tiny_24_p8_384_dist,44.75,712.159,32,384,12.11
+cait_s24_384,44.7,355.627,16,384,47.06
+efficientnet_b6,41.67,381.28,16,528,43.04
+tf_efficientnet_b6_ap,41.13,386.3,16,528,43.04
+tf_efficientnet_b6,41.11,386.574,16,528,43.04
+tf_efficientnet_b6_ns,40.99,387.676,16,528,43.04
+xcit_large_24_p16_384_dist,40.96,582.937,24,384,189.1
+vit_huge_patch14_224,37.91,420.175,16,224,632.05
+resnetrs350,36.74,646.103,24,384,163.96
+swin_large_patch4_window12_384,36.6,435.55,16,384,196.74
+nfnet_f3,34.36,694.502,24,416,254.92
+nfnet_f3s,34.26,696.883,24,416,254.92
+convnext_xlarge_384_in22ft1k,34.01,468.853,16,384,350.2
+resnest269e,32.96,721.9,24,416,110.93
+dm_nfnet_f3,31.6,755.444,24,416,254.92
+resnetv2_50x3_bitm,31.14,513.017,16,448,217.32
+xcit_large_24_p8_224_dist,30.37,523.887,16,224,188.93
+xcit_large_24_p8_224,30.32,524.75,16,224,188.93
+cait_s36_384,29.46,404.043,12,384,68.37
+resnetv2_152x2_bit_teacher_384,29.25,408.083,12,384,236.34
+tf_efficientnetv2_xl_in21ft1k,28.52,415.17,12,512,208.12
+efficientnetv2_xl,28.5,415.628,12,512,208.12
+ig_resnext101_32x48d,27.26,585.456,16,224,828.41
+vit_large_patch16_384,26.62,299.076,8,384,304.72
+efficientnet_b7,25.52,310.186,8,600,66.35
+tf_efficientnet_b7,25.25,313.569,8,600,66.35
+tf_efficientnet_b7_ap,25.18,314.571,8,600,66.35
+tf_efficientnet_b7_ns,25.12,315.222,8,600,66.35
+resnetrs420,24.94,632.993,16,416,191.89
+xcit_small_24_p8_384_dist,23.58,505.912,12,384,47.63
+beit_large_patch16_384,22.66,351.145,8,384,305.0
+vit_giant_patch14_224,21.59,368.118,8,224,1012.61
+resnetv2_152x2_bitm,21.27,373.944,8,448,236.34
+nfnet_f4,18.45,645.682,12,512,316.07
+nfnet_f4s,18.35,649.188,12,512,316.07
+resnetv2_101x3_bitm,17.19,464.077,8,448,387.93
+xcit_medium_24_p8_384_dist,17.03,466.878,8,384,84.32
+dm_nfnet_f4,16.71,713.246,12,512,316.07
+efficientnet_b8,12.85,463.389,6,672,87.41
+tf_efficientnet_b8_ap,12.75,466.866,6,672,87.41
+tf_efficientnet_b8,12.73,467.671,6,672,87.41
+nfnet_f5,12.01,660.317,8,544,377.21
+nfnet_f5s,11.94,664.267,8,544,377.21
+cait_m36_384,11.32,526.589,6,384,271.22
+dm_nfnet_f5,10.99,722.011,8,544,377.21
+xcit_large_24_p8_384_dist,10.37,575.649,6,384,188.93
+nfnet_f6,9.06,655.713,6,576,438.36
+nfnet_f6s,9.04,657.589,6,576,438.36
+tf_efficientnet_l2_ns_475,8.8,449.624,4,475,480.31
+beit_large_patch16_512,8.36,356.829,3,512,305.67
+dm_nfnet_f6,8.32,713.582,6,576,438.36
+nfnet_f7,6.44,613.323,4,608,499.5
+nfnet_f7s,6.38,619.793,4,608,499.5
+cait_m48_448,4.6,430.517,2,448,356.46
+resnetv2_152x4_bitm,4.42,450.22,2,480,936.53
+tf_efficientnet_l2_ns,2.55,387.178,1,800,480.31
+efficientnet_l2,2.52,391.57,1,800,480.31

results/benchmark-train-amp-nhwc-pt110-cu113-rtx3090.csv

@@ -0,0 +1,699 @@
+model,train_samples_per_sec,train_step_time,train_batch_size,train_img_size,param_count
+mobilenetv3_small_050,10124.2,49.732,512,224,1.59
+lcnet_035,9556.39,53.122,512,224,1.64
+tinynet_e,9259.87,53.163,512,106,2.04
+lcnet_050,8204.13,61.959,512,224,1.88
+tf_mobilenetv3_small_minimal_100,8115.25,62.582,512,224,2.04
+tinynet_d,7842.33,64.591,512,152,2.34
+mobilenetv3_small_075,7645.76,66.298,512,224,2.04
+mobilenetv3_small_100,7391.62,68.594,512,224,2.54
+tf_mobilenetv3_small_075,7004.36,72.432,512,224,2.04
+tf_mobilenetv3_small_100,6775.26,74.902,512,224,2.54
+levit_128s,6249.41,81.115,512,224,7.78
+lcnet_075,5706.1,89.275,512,224,2.36
+lcnet_100,5288.36,96.356,512,224,2.95
+mnasnet_small,4565.41,111.287,512,224,2.03
+mnasnet_050,4479.4,113.569,512,224,2.22
+levit_128,4339.94,116.946,512,224,9.21
+mobilenetv2_035,4152.95,122.551,512,224,1.68
+mixer_s32_224,4131.98,123.414,512,224,19.1
+tinynet_c,4071.23,124.835,512,184,2.46
+gernet_s,4039.7,126.102,512,224,8.17
+semnasnet_050,3974.04,127.989,512,224,2.08
+levit_192,3756.95,135.269,512,224,10.95
+vit_small_patch32_224,3671.47,138.697,512,224,22.88
+lcnet_150,3645.88,139.972,512,224,4.5
+ssl_resnet18,3513.72,145.398,512,224,11.69
+gluon_resnet18_v1b,3512.34,145.445,512,224,11.69
+swsl_resnet18,3508.0,145.642,512,224,11.69
+mobilenetv2_050,3501.31,145.498,512,224,1.97
+resnet18,3496.86,146.104,512,224,11.69
+vit_tiny_r_s16_p8_224,3335.03,152.754,512,224,6.34
+mobilenetv3_large_075,3296.24,154.517,512,224,3.99
+ese_vovnet19b_slim_dw,3204.84,159.295,512,224,1.9
+tf_mobilenetv3_large_minimal_100,3175.97,160.546,512,224,3.92
+seresnet18,3165.67,161.261,512,224,11.78
+legacy_seresnet18,3081.19,165.707,512,224,11.78
+mnasnet_075,3078.06,165.602,512,224,3.17
+tf_mobilenetv3_large_075,3052.6,166.905,512,224,3.99
+ghostnet_050,2997.88,169.549,512,224,2.59
+mobilenetv3_rw,2978.91,171.068,512,224,5.48
+mobilenetv3_large_100_miil,2954.41,172.485,512,224,5.48
+mobilenetv3_large_100,2949.17,172.802,512,224,5.48
+levit_256,2853.53,178.318,512,224,18.89
+hardcorenas_a,2833.05,180.061,512,224,5.26
+mnasnet_100,2794.45,182.489,512,224,4.38
+mnasnet_b1,2793.42,182.553,512,224,4.38
+resnet18d,2767.22,184.677,512,224,11.71
+tf_mobilenetv3_large_100,2745.99,185.646,512,224,5.48
+tinynet_b,2724.35,186.966,512,188,3.73
+hardcorenas_b,2641.84,192.939,512,224,5.18
+semnasnet_075,2627.64,193.989,512,224,2.91
+hardcorenas_c,2601.15,195.937,512,224,5.52
+ese_vovnet19b_slim,2546.07,200.754,512,224,3.17
+mobilenetv2_075,2520.15,202.421,512,224,2.64
+spnasnet_100,2483.13,205.29,512,224,4.42
+tf_efficientnetv2_b0,2462.53,206.791,512,224,7.14
+hardcorenas_d,2435.65,209.09,512,224,7.5
+mnasnet_a1,2380.33,214.252,512,224,3.89
+semnasnet_100,2377.81,214.477,512,224,3.89
+mobilenetv2_100,2363.97,215.837,512,224,3.5
+regnetx_002,2307.34,221.263,512,224,2.68
+tinynet_a,2237.07,227.684,512,192,6.19
+fbnetc_100,2225.25,229.187,512,224,5.57
+ghostnet_100,2211.89,230.191,512,224,5.18
+regnety_002,2204.25,231.393,512,224,3.16
+efficientnet_lite0,2189.09,233.188,512,224,4.65
+hardcorenas_f,2163.36,235.624,512,224,8.2
+hardcorenas_e,2139.96,238.217,512,224,8.07
+tf_efficientnet_lite0,2037.27,250.609,512,224,4.65
+resnet34,2037.12,250.805,512,224,21.8
+tv_resnet34,2035.85,250.969,512,224,21.8
+gluon_resnet34_v1b,2035.71,250.984,512,224,21.8
+skresnet18,2026.36,252.064,512,224,11.96
+resnet26,1936.86,263.917,512,224,16.0
+resnetblur18,1923.61,265.837,512,224,11.69
+gernet_m,1870.52,273.071,512,224,21.14
+ese_vovnet19b_dw,1864.29,274.174,512,224,6.54
+nf_resnet26,1859.82,274.839,512,224,16.0
+mnasnet_140,1826.42,279.575,512,224,7.12
+hrnet_w18_small,1825.55,279.196,512,224,13.19
+seresnet34,1821.14,280.321,512,224,21.96
+legacy_seresnet34,1770.27,288.393,512,224,21.96
+efficientnet_b0,1763.31,216.796,384,224,5.29
+resnet34d,1762.59,289.926,512,224,21.82
+mobilenetv2_110d,1752.61,218.148,384,224,4.52
+levit_384,1748.52,291.817,512,224,39.13
+rexnetr_100,1739.4,219.681,384,224,4.88
+selecsls42,1711.06,298.631,512,224,30.35
+selecsls42b,1703.95,299.892,512,224,32.46
+vit_tiny_patch16_224,1668.2,306.156,512,224,5.72
+deit_tiny_patch16_224,1666.4,306.494,512,224,5.72
+tf_efficientnet_b0_ap,1660.27,230.299,384,224,5.29
+tf_efficientnet_b0_ns,1659.78,230.359,384,224,5.29
+tf_efficientnet_b0,1659.12,230.455,384,224,5.29
+deit_tiny_distilled_patch16_224,1644.53,310.585,512,224,5.91
+semnasnet_140,1620.0,315.219,512,224,6.11
+efficientnet_es_pruned,1595.68,320.161,512,224,5.44
+efficientnet_es,1595.53,320.205,512,224,5.44
+mixer_b32_224,1593.71,320.545,512,224,60.29
+tf_efficientnet_es,1592.03,320.904,512,224,5.44
+resnet26d,1580.82,323.426,512,224,16.01
+tf_efficientnetv2_b1,1573.77,242.663,384,240,8.14
+ghostnet_130,1567.24,325.459,512,224,7.36
+repvgg_b0,1556.65,327.916,512,224,15.82
+pit_ti_distilled_224,1537.82,332.146,512,224,5.1
+pit_ti_224,1535.4,332.663,512,224,4.85
+mobilenetv2_140,1525.48,250.968,384,224,6.11
+resmlp_12_224,1516.23,336.972,512,224,15.35
+resmlp_12_distilled_224,1515.94,336.996,512,224,15.35
+vit_base_patch32_224,1491.42,342.538,512,224,88.22
+vit_base_patch32_224_sam,1490.35,342.802,512,224,88.22
+selecsls60,1450.31,352.185,512,224,30.67
+selecsls60b,1447.33,352.935,512,224,32.77
+nf_seresnet26,1446.95,353.243,512,224,17.4
+xcit_nano_12_p16_224,1428.16,356.892,512,224,3.05
+xcit_nano_12_p16_224_dist,1427.07,357.115,512,224,3.05
+efficientnet_lite1,1402.56,181.634,256,240,5.42
+mixer_s16_224,1387.35,368.544,512,224,18.53
+efficientnet_b1_pruned,1356.21,376.138,512,240,6.33
+dla46_c,1347.45,379.298,512,224,1.3
+rexnetr_130,1335.58,190.622,256,224,7.61
+nf_ecaresnet26,1333.44,383.458,512,224,16.0
+tf_efficientnet_lite1,1330.92,191.449,256,240,5.42
+mobilenetv2_120d,1299.02,195.983,256,224,5.83
+resnetv2_50,1277.54,400.053,512,224,25.55
+rexnet_100,1259.4,303.853,384,224,4.8
+gernet_l,1250.61,408.582,512,256,31.08
+pit_xs_224,1243.31,411.016,512,224,10.62
+pit_xs_distilled_224,1243.18,411.013,512,224,11.0
+gmixer_12_224,1232.53,414.68,512,224,12.7
+resnet26t,1208.16,423.321,512,256,16.01
+vit_small_patch32_384,1207.26,423.36,512,384,22.92
+tv_resnet50,1193.37,321.016,384,224,25.56
+ssl_resnet50,1193.18,321.076,384,224,25.56
+gluon_resnet50_v1b,1192.95,321.138,384,224,25.56
+resnet50,1192.59,321.245,384,224,25.56
+swsl_resnet50,1191.64,321.479,384,224,25.56
+rexnetr_150,1171.61,217.431,256,224,9.78
+fbnetv3_b,1163.46,328.537,384,256,8.6
+regnetx_004,1151.84,443.491,512,224,5.16
+botnet26t_256,1146.74,334.32,384,256,12.49
+skresnet34,1146.67,445.45,512,224,22.28
+tf_efficientnetv2_b2,1137.18,223.699,256,260,10.1
+fbnetv3_d,1123.37,226.192,256,256,10.31
+resnetv2_50t,1114.22,458.774,512,224,25.57
+gluon_resnet50_v1c,1114.01,343.942,384,224,25.58
+repvgg_a2,1113.97,458.853,512,224,28.21
+resnetv2_50d,1111.46,459.91,512,224,25.57
+halonet26t,1094.74,350.238,384,256,12.48
+efficientnet_lite2,1085.96,234.846,256,260,6.09
+dla34,1074.06,476.134,512,224,15.74
+mixnet_s,1060.15,481.782,512,224,4.13
+resnet50d,1054.46,363.395,384,224,25.58
+gluon_resnet50_v1d,1053.62,363.691,384,224,25.58
+resnet50t,1052.26,364.133,384,224,25.57
+vit_tiny_r_s16_p8_384,1046.6,366.108,384,384,6.36
+xcit_tiny_12_p16_224_dist,1038.68,491.328,512,224,6.72
+xcit_tiny_12_p16_224,1038.06,491.593,512,224,6.72
+tf_efficientnet_lite2,1032.64,247.013,256,260,6.09
+legacy_seresnext26_32x4d,1027.19,497.849,512,224,16.79
+resnet32ts,1007.64,253.517,256,256,17.96
+tf_efficientnet_b1_ap,1002.39,253.996,256,240,7.79
+tf_efficientnet_b1,1001.08,254.315,256,240,7.79
+tf_efficientnet_b1_ns,1000.92,254.412,256,240,7.79
+res2net50_48w_2s,999.09,383.596,384,224,25.29
+vit_small_resnet26d_224,998.6,383.579,384,224,63.61
+resnet33ts,996.53,256.337,256,256,19.68
+seresnext26d_32x4d,978.97,391.64,384,224,16.81
+vovnet39a,977.19,523.377,512,224,22.6
+seresnext26tn_32x4d,976.94,392.442,384,224,16.81
+seresnext26t_32x4d,976.02,392.82,384,224,16.81
+legacy_seresnet50,965.33,396.753,384,224,28.09
+tf_efficientnet_em,961.83,265.228,256,240,6.9
+efficientnet_em,961.29,265.407,256,240,6.9
+tf_mixnet_s,959.89,532.224,512,224,4.13
+dla46x_c,958.36,533.526,512,224,1.07
+eca_resnet33ts,943.71,270.653,256,256,19.68
+seresnet33ts,943.39,270.622,256,256,19.78
+crossvit_tiny_240,940.57,406.823,384,240,7.01
+crossvit_9_240,937.22,408.452,384,240,8.55
+efficientnet_b1,935.15,272.379,256,256,7.79
+seresnet50,934.74,409.786,384,224,28.09
+regnety_004,931.71,548.466,512,224,4.34
+gluon_resnet50_v1s,928.8,412.664,384,224,25.68
+ecaresnetlight,928.23,550.753,512,224,30.16
+resnetaa50d,926.64,413.614,384,224,25.58
+dla60,926.01,413.77,384,224,22.04
+ecaresnext26t_32x4d,925.94,414.209,384,224,15.41
+ecaresnext50t_32x4d,924.8,414.714,384,224,15.41
+ese_vovnet39b,919.92,555.952,512,224,24.57
+convit_tiny,917.45,417.618,384,224,5.71
+crossvit_9_dagger_240,912.87,419.313,384,240,8.78
+gcresnet33ts,912.73,279.541,256,256,19.88
+eca_vovnet39b,910.35,561.796,512,224,22.6
+gmlp_ti16_224,908.97,421.011,384,224,5.87
+dla60x_c,908.47,562.709,512,224,1.32
+cspresnet50,908.32,421.991,384,256,21.62
+vgg11,905.72,565.146,512,224,132.86
+resnetblur50,901.02,425.435,384,224,25.56
+ecaresnet50d_pruned,892.5,572.812,512,224,19.94
+rexnetr_200,885.64,215.704,192,224,16.52
+lambda_resnet26rpt_256,881.41,217.279,192,256,10.99
+vit_small_patch16_224,879.33,435.939,384,224,22.05
+deit_small_patch16_224,876.6,437.293,384,224,22.05
+nf_seresnet50,866.69,441.953,384,224,28.09
+dpn68b,864.02,443.325,384,224,12.61
+cspresnext50,863.58,443.862,384,224,20.57
+cspresnet50w,858.14,446.674,384,256,28.12
+selecsls84,852.15,599.694,512,224,50.95
+hrnet_w18_small_v2,849.12,600.799,512,224,15.6
+cspresnet50d,847.29,452.406,384,256,21.64
+deit_small_distilled_patch16_224,846.66,301.609,256,224,22.44
+seresnet50t,844.69,453.54,384,224,28.1
+resnetrs50,843.24,454.312,384,224,35.69
+densenet121,835.01,304.929,256,224,7.98
+tv_densenet121,832.77,305.766,256,224,7.98
+rexnet_150,829.93,307.376,256,224,9.73
+gluon_resnext50_32x4d,824.86,464.778,384,224,25.03
+tv_resnext50_32x4d,824.06,465.237,384,224,25.03
+ssl_resnext50_32x4d,823.91,465.298,384,224,25.03
+swsl_resnext50_32x4d,823.16,465.728,384,224,25.03
+resnext50_32x4d,823.01,465.815,384,224,25.03
+dpn68,816.18,469.476,384,224,12.61
+res2net50_26w_4s,812.07,314.023,256,224,25.7
+efficientnet_b2_pruned,811.6,314.06,256,260,8.31
+resnetblur50d,810.86,314.915,256,224,25.58
+pit_s_224,809.09,315.601,256,224,23.46
+pit_s_distilled_224,808.91,315.694,256,224,24.04
+vovnet57a,808.86,473.9,384,224,36.64
+skresnet50,802.36,317.782,256,224,25.8
+densenet121d,795.36,320.183,256,224,8.0
+vit_base_resnet26d_224,792.88,483.157,384,224,101.4
+rexnet_130,792.23,322.087,256,224,7.56
+tf_efficientnet_b2_ns,791.33,241.259,192,260,9.11
+tf_efficientnet_b2_ap,790.24,241.588,192,260,9.11
+regnetx_006,788.72,648.389,512,224,6.2
+tf_efficientnet_b2,788.26,242.2,192,260,9.11
+vgg11_bn,788.01,487.085,384,224,132.87
+nf_ecaresnet50,787.45,486.772,384,224,25.56
+regnety_006,779.2,656.099,512,224,6.06
+gcresnet50t,776.26,328.367,256,256,25.9
+ese_vovnet57b,765.66,500.673,384,224,38.61
+ecaresnet50d,761.68,335.218,256,224,25.58
+resnext50d_32x4d,757.82,337.002,256,224,25.05
+adv_inception_v3,756.25,506.326,384,299,23.83
+tf_inception_v3,754.65,507.545,384,299,23.83
+gluon_inception_v3,754.57,507.572,384,299,23.83
+inception_v3,753.64,508.224,384,299,23.83
+seresnetaa50d,751.96,339.385,256,224,28.11
+res2net50_14w_8s,748.86,339.851,256,224,25.06
+resmlp_24_distilled_224,743.59,342.912,256,224,30.02
+resmlp_24_224,743.53,342.937,256,224,30.02
+resnetv2_101,742.52,343.363,256,224,44.54
+skresnet50d,738.05,345.557,256,224,25.82
+sehalonet33ts,731.23,349.345,256,256,13.69
+dla60_res2net,730.56,349.091,256,224,20.85
+densenetblur121d,729.66,349.185,256,224,8.0
+resnet101,719.46,354.412,256,224,44.55
+tv_resnet101,719.29,354.488,256,224,44.55
+gluon_resnet101_v1b,719.28,354.459,256,224,44.55
+efficientnet_b2a,707.95,269.81,192,288,9.11
+efficientnet_b2,707.16,270.111,192,288,9.11
+nf_resnet50,703.13,545.338,384,288,25.56
+nf_resnet101,702.76,544.906,384,224,44.55
+ecaresnet101d_pruned,698.97,730.858,512,224,24.88
+densenet169,691.74,367.774,256,224,14.15
+gluon_resnet101_v1c,690.21,369.47,256,224,44.57
+gluon_seresnext50_32x4d,689.84,370.073,256,224,27.56
+legacy_seresnext50_32x4d,688.18,370.947,256,224,27.56
+seresnext50_32x4d,688.18,370.943,256,224,27.56
+resnetv2_101d,686.94,371.274,256,224,44.56
+darknet53,683.71,373.676,256,256,41.61
+dla60x,680.87,375.081,256,224,17.35
+nf_regnet_b0,680.47,751.331,512,256,8.76
+efficientnetv2_rw_t,669.58,284.789,192,288,13.65
+gluon_resnet101_v1d,666.43,382.723,256,224,44.57
+regnety_008,657.21,778.104,512,224,6.26
+rexnet_200,656.53,291.353,192,224,16.37
+tf_efficientnetv2_b3,655.17,291.405,192,300,14.36
+vit_small_r26_s32_224,652.39,391.27,256,224,36.43
+sebotnet33ts_256,650.1,196.14,128,256,13.7
+wide_resnet50_2,646.12,593.536,384,224,68.88
+vgg13,645.37,594.836,384,224,133.05
+dla102,640.56,398.166,256,224,33.27
+vit_base_resnet50d_224,619.99,411.439,256,224,110.97
+gluon_resnet101_v1s,614.53,415.124,256,224,44.67
+resnext26ts,613.38,625.548,384,256,10.3
+resnetaa101d,612.93,416.224,256,224,44.57
+ecaresnet26t,610.98,418.477,256,320,16.01
+gmixer_24_224,610.59,417.896,256,224,24.72
+repvgg_b1,609.77,838.686,512,224,57.42
+eca_botnext26ts_256,609.21,419.672,256,256,10.59
+eca_halonext26ts,598.72,427.016,256,256,10.76
+gc_efficientnetv2_rw_t,597.14,319.011,192,288,13.68
+eca_resnext26ts,593.02,431.168,256,256,10.3
+seresnext26ts,592.35,431.553,256,256,10.39
+regnetx_008,587.44,870.811,512,224,7.26
+convnext_tiny,584.52,437.049,256,224,28.59
+coat_lite_tiny,583.0,657.845,384,224,5.72
+convnext_tiny_hnf,582.18,438.813,256,224,28.59
+gcresnext26ts,581.42,439.532,256,256,10.48
+mixnet_m,576.71,664.443,384,224,5.01
+resnetv2_50x1_bit_distilled,575.65,332.777,192,224,25.55
+cspdarknet53,570.42,447.825,256,256,27.64
+legacy_seresnet101,570.04,447.058,256,224,49.33
+halonet50ts,568.71,336.701,192,256,22.73
+res2net50_26w_6s,565.22,451.306,256,224,37.05
+resnetblur101d,562.81,453.413,256,224,44.57
+seresnet101,560.51,454.74,256,224,49.33
+coat_lite_mini,559.21,685.884,384,224,11.01
+xcit_small_12_p16_224,558.21,456.962,256,224,26.25
+xcit_small_12_p16_224_dist,557.95,457.219,256,224,26.25
+vgg13_bn,555.55,460.55,256,224,133.05
+tf_efficientnet_lite3,555.29,229.476,128,300,8.2
+efficientnet_lite3,554.26,229.924,128,300,8.2
+tf_mixnet_m,541.25,708.06,384,224,5.01
+ssl_resnext101_32x4d,539.66,472.944,256,224,44.18
+gluon_resnext101_32x4d,539.41,473.189,256,224,44.18
+xcit_tiny_24_p16_224_dist,538.89,472.126,256,224,12.12
+resnext101_32x4d,538.71,473.739,256,224,44.18
+xcit_tiny_24_p16_224,538.48,472.477,256,224,12.12
+swsl_resnext101_32x4d,538.39,474.037,256,224,44.18
+densenet201,535.33,355.953,192,224,20.01
+bat_resnext26ts,534.3,478.021,256,256,10.73
+twins_svt_small,529.33,482.243,256,224,24.06
+mixer_b16_224,521.89,489.791,256,224,59.88
+mixer_b16_224_miil,521.33,490.33,256,224,59.88
+vgg16,519.79,738.569,384,224,138.36
+nf_seresnet101,519.55,490.612,256,224,49.33
+gmlp_s16_224,514.57,371.708,192,224,19.42
+crossvit_small_240,514.25,371.914,192,240,26.86
+resnetv2_152,514.07,495.927,256,224,60.19
+res2net101_26w_4s,512.94,372.022,192,224,45.21
+resnetv2_50d_gn,512.21,374.104,192,224,25.57
+botnet50ts_256,511.02,249.57,128,256,22.74
+vit_base_r26_s32_224,509.73,375.506,192,224,101.38
+efficientnet_b3_pruned,508.52,501.852,256,300,9.86
+nf_regnet_b3,507.41,502.895,256,320,18.59
+res2next50,500.31,510.5,256,224,24.67
+xcit_nano_12_p16_384_dist,497.26,513.197,256,384,3.05
+dla102x,495.72,385.887,192,224,26.31
+resnet50_gn,494.95,387.141,192,224,25.56
+mixnet_l,494.74,516.045,256,224,7.33
+resnet152,490.44,389.409,192,224,60.19
+gluon_resnet152_v1b,490.37,389.44,192,224,60.19
+tv_resnet152,490.23,389.552,192,224,60.19
+resmlp_36_224,489.84,389.965,192,224,44.69
+resmlp_36_distilled_224,489.66,390.129,192,224,44.69
+cait_xxs24_224,487.15,523.231,256,224,11.96
+mixer_l32_224,486.19,393.553,192,224,206.94
+halo2botnet50ts_256,485.95,262.504,128,256,22.64
+xception,485.66,262.809,128,299,22.86
+visformer_tiny,484.78,1055.527,512,224,10.32
+vit_large_patch32_224,484.04,527.472,256,224,306.54
+vit_base_patch32_384,478.93,533.779,256,384,88.3
+gluon_resnet152_v1c,476.8,400.507,192,224,60.21
+resnetv2_152d,475.38,401.77,192,224,60.2
+crossvit_15_240,472.73,404.523,192,240,27.53
+res2net50_26w_8s,468.77,407.558,192,224,48.4
+tf_mixnet_l,466.85,546.936,256,224,7.33
+gluon_resnet152_v1d,465.06,410.759,192,224,60.21
+ecaresnet101d,462.22,552.191,256,224,44.57
+crossvit_15_dagger_240,459.76,415.927,192,240,28.21
+vgg16_bn,457.66,559.065,256,224,138.37
+nf_ecaresnet101,455.3,560.589,256,224,44.55
+efficientnet_el,446.2,285.844,128,300,10.59
+efficientnet_el_pruned,445.59,286.28,128,300,10.59
+swin_tiny_patch4_window7_224,445.53,573.708,256,224,28.29
+fbnetv3_g,442.28,287.506,128,288,16.62
+gluon_resnet152_v1s,441.14,433.129,192,224,60.32
+gluon_seresnext101_32x4d,437.16,437.191,192,224,48.96
+tf_efficientnet_el,436.28,292.36,128,300,10.59
+vgg19,436.15,880.196,384,224,143.67
+seresnext101_32x4d,435.91,438.42,192,224,48.96
+legacy_seresnext101_32x4d,435.08,439.246,192,224,48.96
+ese_vovnet99b,434.62,587.614,256,224,63.2
+skresnext50_32x4d,434.44,587.996,256,224,27.48
+twins_pcpvt_small,433.1,589.648,256,224,24.11
+dla60_res2next,431.53,591.928,256,224,17.03
+dla169,427.18,447.18,192,224,53.39
+hrnet_w32,426.75,445.609,192,224,41.23
+hrnet_w18,413.83,614.275,256,224,21.3
+convit_small,409.81,467.592,192,224,27.78
+vit_tiny_patch16_384,409.3,311.972,128,384,5.79
+vit_small_resnet50d_s16_224,407.06,470.553,192,224,57.53
+vgg19_bn,388.74,658.182,256,224,143.68
+ecaresnet50t,388.23,328.837,128,320,25.57
+inception_v4,387.4,493.551,192,299,42.68
+tf_efficientnet_b3_ap,384.58,331.269,128,300,12.23
+tf_efficientnet_b3_ns,384.22,331.517,128,300,12.23
+tf_efficientnet_b3,384.13,331.655,128,300,12.23
+legacy_seresnet152,382.75,498.635,192,224,66.82
+xcit_nano_12_p8_224,378.83,674.154,256,224,3.05
+xcit_nano_12_p8_224_dist,378.65,674.433,256,224,3.05
+convnext_small,375.68,509.406,192,224,50.22
+regnetx_016,374.21,1367.378,512,224,9.19
+haloregnetz_b,372.27,686.285,256,224,11.68
+densenet161,371.62,342.252,128,224,28.68
+repvgg_b2,368.23,1389.47,512,224,89.02
+seresnet152,362.95,349.646,128,224,66.82
+nest_tiny,356.9,357.853,128,224,17.06
+dla102x2,355.19,358.924,128,224,41.28
+vit_base_patch16_224_miil,354.79,360.089,128,224,86.54
+jx_nest_tiny,353.97,360.795,128,224,17.06
+gluon_resnext101_64x4d,352.72,361.493,128,224,83.46
+resnext101_64x4d,352.52,361.632,128,224,83.46
+wide_resnet101_2,350.85,545.834,192,224,126.89
+regnety_016,350.64,1458.484,512,224,11.2
+vit_base_patch16_224,349.4,365.598,128,224,86.57
+deit_base_patch16_224,349.3,365.691,128,224,86.57
+vit_base_patch16_224_sam,349.14,365.832,128,224,86.57
+resnet200,348.81,364.208,128,224,64.67
+resnest14d,348.25,1469.78,512,224,10.61
+efficientnet_b3a,345.42,368.96,128,320,12.23
+efficientnet_b3,345.06,369.413,128,320,12.23
+deit_base_distilled_patch16_224,344.74,370.536,128,224,87.34
+crossvit_18_240,343.92,370.408,128,240,43.27
+hrnet_w30,341.47,745.312,256,224,37.71
+crossvit_18_dagger_240,335.19,380.045,128,240,44.27
+ens_adv_inception_resnet_v2,334.4,571.029,192,299,55.84
+inception_resnet_v2,334.03,571.741,192,299,55.84
+efficientnet_lite4,332.63,191.154,64,380,13.01
+nf_regnet_b1,327.56,1561.741,512,288,10.22
+coat_lite_small,324.96,589.506,192,224,19.84
+dpn92,324.94,786.493,256,224,37.67
+resnet101d,323.43,394.33,128,320,44.57
+mixnet_xl,322.47,593.686,192,224,11.9
+beit_base_patch16_224,320.26,398.682,128,224,86.53
+tf_efficientnet_lite4,319.09,199.328,64,380,13.01
+gcresnext50ts,314.79,608.558,192,256,15.67
+resnetrs101,314.59,404.798,128,288,63.62
+twins_pcpvt_base,313.15,610.712,192,224,43.83
+cait_xxs36_224,311.89,407.079,128,224,17.3
+gluon_seresnext101_64x4d,306.13,416.102,128,224,88.23
+twins_svt_base,305.99,416.477,128,224,56.07
+repvgg_b3,305.95,1254.102,384,224,123.09
+regnetz_d8,301.65,210.668,64,320,23.37
+xception41,300.49,424.83,128,299,26.97
+efficientnetv2_s,292.99,325.601,96,384,21.46
+ese_vovnet39b_evos,292.82,436.279,128,224,24.58
+efficientnetv2_rw_s,289.9,218.715,64,384,23.94
+tf_efficientnetv2_s,289.31,329.828,96,384,21.46
+swin_small_patch4_window7_224,289.15,441.029,128,224,49.61
+tf_efficientnetv2_s_in21ft1k,288.88,330.272,96,384,21.46
+xcit_small_24_p16_224_dist,286.23,444.237,128,224,47.67
+xcit_small_24_p16_224,286.12,444.416,128,224,47.67
+hrnet_w40,285.91,667.227,192,224,57.56
+pit_b_224,285.13,335.829,96,224,73.76
+pit_b_distilled_224,284.45,336.631,96,224,74.79
+nf_regnet_b2,282.04,1813.867,512,272,14.31
+gluon_xception65,276.41,345.52,96,299,39.92
+dpn98,275.63,462.928,128,224,61.57
+xception65,275.32,346.928,96,299,39.92
+resnet51q,274.2,699.442,192,288,35.7
+convnext_base_in22ft1k,273.43,466.529,128,224,88.59
+convnext_base,272.92,467.434,128,224,88.59
+nf_regnet_b4,271.03,470.306,128,384,30.21
+xcit_tiny_12_p8_224,270.45,471.679,128,224,6.71
+xcit_tiny_12_p8_224_dist,270.45,471.621,128,224,6.71
+cait_s24_224,268.53,474.373,128,224,46.92
+hrnet_w48,262.05,484.085,128,224,77.47
+resnest26d,254.99,1505.218,384,224,17.07
+resnest50d_1s4x24d,252.9,1011.055,256,224,25.68
+resnetv2_50d_evob,247.25,775.496,192,224,25.59
+regnetz_c16,246.47,517.947,128,320,13.46
+gmlp_b16_224,243.4,392.966,96,224,73.08
+hrnet_w44,236.76,806.557,192,224,67.06
+twins_svt_large,235.13,542.578,128,224,99.27
+efficientnet_b4,234.85,270.618,64,384,19.34
+nest_small,234.57,407.797,96,224,38.35
+resnetv2_50d_evos,233.18,410.671,96,224,25.59
+jx_nest_small,233.1,410.378,96,224,38.35
+tf_efficientnet_b4_ap,230.34,275.916,64,380,19.34
+tresnet_m,230.07,2222.819,512,224,31.39
+tf_efficientnet_b4,229.98,276.378,64,380,19.34
+tf_efficientnet_b4_ns,229.78,276.586,64,380,19.34
+visformer_small,229.49,1672.652,384,224,40.22
+crossvit_base_240,228.79,418.11,96,240,105.03
+nfnet_l0,228.47,1119.415,256,288,35.07
+coat_tiny,228.08,559.427,128,224,5.5
+twins_pcpvt_large,223.01,570.431,128,224,60.99
+eca_nfnet_l0,221.69,1153.795,256,288,24.14
+vit_large_r50_s32_224,220.98,432.297,96,224,328.99
+swin_base_patch4_window7_224,220.89,577.812,128,224,87.77
+densenet264,217.72,437.317,96,224,72.69
+hrnet_w64,216.27,587.374,128,224,128.06
+convit_base,213.62,448.479,96,224,86.54
+xception71,212.24,299.632,64,299,42.34
+vit_small_r26_s32_384,211.98,300.724,64,384,36.47
+resnet152d,211.82,451.04,96,320,60.21
+vit_small_patch16_384,209.76,304.352,64,384,22.2
+xcit_medium_24_p16_224,207.55,459.61,96,224,84.4
+xcit_medium_24_p16_224_dist,207.03,460.529,96,224,84.4
+resnet61q,206.68,618.373,128,288,36.85
+dpn131,205.78,620.133,128,224,79.25
+mixnet_xxl,204.23,624.949,128,224,23.96
+vit_base_r50_s16_224,195.65,325.59,64,224,98.66
+coat_mini,193.15,660.874,128,224,10.34
+regnety_040,193.14,1986.785,384,224,20.65
+xcit_small_12_p16_384_dist,191.88,498.659,96,384,26.25
+xcit_tiny_24_p16_384_dist,186.09,512.939,96,384,12.12
+seresnet200d,185.21,514.398,96,256,71.86
+regnetz_b16,179.85,1066.235,192,288,9.72
+repvgg_b1g4,177.2,2888.374,512,224,39.97
+ecaresnet200d,176.34,541.265,96,256,64.69
+nest_base,176.15,361.868,64,224,67.72
+convmixer_768_32,176.07,543.988,96,224,21.11
+resnest50d,175.35,1458.742,256,224,27.48
+regnety_032,175.25,1459.459,256,288,19.44
+jx_nest_base,174.95,364.333,64,224,67.72
+regnetx_032,174.19,2203.311,384,224,15.3
+senet154,172.45,553.616,96,224,115.09
+gluon_senet154,172.38,553.858,96,224,115.09
+legacy_senet154,171.96,555.183,96,224,115.09
+seresnet152d,169.43,374.655,64,320,66.84
+resnetrs152,167.98,377.927,64,320,86.62
+dpn107,167.97,570.011,96,224,86.92
+halonet_h1,167.27,381.366,64,256,8.1
+resnest50d_4s2x40d,163.46,1564.906,256,224,30.42
+resnet200d,163.41,388.892,64,320,64.69
+mixer_l16_224,161.74,592.23,96,224,208.2
+convnext_large_in22ft1k,160.8,595.418,96,224,197.77
+convnext_large,160.72,595.739,96,224,197.77
+regnetx_040,160.66,2389.077,384,224,22.12
+tnt_s_patch16_224,157.73,809.855,128,224,23.76
+regnetx_080,148.64,1721.236,256,224,39.57
+efficientnetv2_m,147.13,323.331,48,416,54.14
+resnext101_32x8d,145.87,876.049,128,224,88.79
+ssl_resnext101_32x8d,145.3,879.52,128,224,88.79
+vit_large_patch32_384,145.15,439.536,64,384,306.63
+xcit_small_12_p8_224_dist,145.04,439.625,64,224,26.21
+xcit_small_12_p8_224,144.99,439.796,64,224,26.21
+ig_resnext101_32x8d,144.47,884.578,128,224,88.79
+swsl_resnext101_32x8d,144.29,885.69,128,224,88.79
+regnetz_e8,140.91,338.814,48,320,57.7
+xcit_tiny_24_p8_224,139.28,686.242,96,224,12.11
+xcit_tiny_24_p8_224_dist,139.16,686.889,96,224,12.11
+resnetv2_50x1_bitm,137.88,347.358,48,448,25.55
+seresnext101_32x8d,136.74,934.059,128,224,93.57
+seresnet269d,134.18,471.701,64,256,113.67
+swin_large_patch4_window7_224,132.61,480.998,64,224,196.53
+xcit_nano_12_p8_384_dist,128.43,496.705,64,384,3.05
+tnt_b_patch16_224,126.43,757.626,96,224,65.41
+xcit_large_24_p16_224,124.67,510.437,64,224,189.1
+xcit_large_24_p16_224_dist,124.62,510.639,64,224,189.1
+efficientnetv2_rw_m,123.91,255.105,32,416,53.24
+regnetx_064,123.53,2071.55,256,224,26.21
+regnety_064,121.34,2108.152,256,224,30.58
+resnetrs200,119.86,396.447,48,320,93.21
+nfnet_f0,119.25,2145.698,256,256,71.49
+nfnet_f0s,118.17,2165.276,256,256,71.49
+ese_vovnet99b_iabn,116.31,3298.583,384,224,63.2
+regnety_080,108.46,2359.307,256,224,39.18
+tf_efficientnetv2_m_in21ft1k,107.43,294.9,32,480,54.14
+vit_large_patch16_224,106.81,447.947,48,224,304.33
+tf_efficientnetv2_m,106.77,296.673,32,480,54.14
+convnext_xlarge_in22ft1k,104.57,610.376,64,224,350.2
+crossvit_15_dagger_408,103.52,307.452,32,408,28.5
+tresnet_l,102.9,4972.484,512,224,55.99
+repvgg_b2g4,101.39,5048.708,512,224,61.76
+eca_nfnet_l1,99.64,1282.96,128,320,41.41
+xcit_small_24_p16_384_dist,98.62,483.709,48,384,47.67
+beit_large_patch16_224,98.03,487.815,48,224,304.43
+cspdarknet53_iabn,97.55,3934.064,384,256,27.64
+efficientnet_b5,94.83,335.072,32,456,30.39
+convnext_base_384_in22ft1k,93.76,510.323,48,384,88.59
+tf_efficientnet_b5_ap,93.42,340.255,32,456,30.39
+tf_efficientnet_b5_ns,93.39,340.338,32,456,30.39
+tf_efficientnet_b5,93.26,340.852,32,456,30.39
+xcit_tiny_12_p8_384_dist,92.1,519.585,48,384,6.71
+tresnet_xl,89.9,4267.972,384,224,78.44
+vit_base_patch16_384,89.9,355.185,32,384,86.86
+deit_base_patch16_384,89.87,355.331,32,384,86.86
+cspresnext50_iabn,89.4,4293.09,384,256,20.57
+resnest101e,88.34,1446.581,128,256,48.28
+deit_base_distilled_patch16_384,88.07,362.54,32,384,87.63
+cait_xxs24_384,87.17,364.76,32,384,12.03
+nf_regnet_b5,85.26,748.171,64,456,49.74
+resnetv2_101x1_bitm,84.86,375.607,32,448,44.54
+resnetv2_152x2_bit_teacher,84.71,375.537,32,224,236.34
+repvgg_b3g4,83.61,4591.876,384,224,83.83
+crossvit_18_dagger_408,77.92,306.154,24,408,44.61
+regnetx_120,77.0,3323.571,256,224,46.11
+beit_base_patch16_384,76.44,313.03,24,384,86.74
+ecaresnet269d,75.74,418.332,32,352,102.09
+regnety_120,75.43,2544.024,192,224,51.82
+pnasnet5large,75.01,423.127,32,331,86.06
+vit_large_r50_s32_384,73.58,432.769,32,384,329.09
+xcit_small_24_p8_224_dist,73.31,433.498,32,224,47.63
+xcit_small_24_p8_224,73.25,433.859,32,224,47.63
+resnetrs270,73.04,432.615,32,352,129.86
+regnetz_d32,71.19,897.484,64,320,27.58
+dm_nfnet_f0,71.06,3601.578,256,256,71.49
+xcit_medium_24_p16_384_dist,70.95,448.036,32,384,84.4
+regnetz_d8_evob,68.84,695.406,48,320,23.41
+regnety_320,66.69,1917.907,128,224,145.05
+nasnetalarge,65.99,480.339,32,331,88.75
+resmlp_big_24_distilled_224,65.96,483.761,32,224,129.14
+resmlp_big_24_224,65.94,483.953,32,224,129.14
+resmlp_big_24_224_in22ft1k,65.9,484.194,32,224,129.14
+regnetx_160,65.04,2951.208,192,224,54.28
+regnetz_d8_evos,62.47,510.384,32,320,23.46
+cait_xs24_384,60.56,393.982,24,384,26.67
+swin_base_patch4_window12_384,60.23,396.845,24,384,87.9
+efficientnetv2_l,60.16,261.949,16,480,118.52
+convmixer_1024_20_ks9_p14,59.76,4282.918,256,224,24.38
+tf_efficientnetv2_l,59.73,263.768,16,480,118.52
+tf_efficientnetv2_l_in21ft1k,59.31,265.732,16,480,118.52
+vit_base_patch8_224,57.86,275.754,16,224,86.58
+cait_xxs36_384,56.92,418.29,24,384,17.37
+convnext_large_384_in22ft1k,54.87,581.624,32,384,197.77
+xcit_medium_24_p8_224,54.52,584.052,32,224,84.32
+xcit_medium_24_p8_224_dist,54.51,583.988,32,224,84.32
+vit_base_resnet50_384,54.07,294.48,16,384,98.95
+vit_base_r50_s16_384,54.05,294.543,16,384,98.95
+tresnet_m_448,52.97,3622.14,192,448,31.39
+xcit_small_12_p8_384_dist,49.7,481.259,24,384,26.21
+regnety_160,48.21,1990.283,96,288,83.59
+xcit_tiny_24_p8_384_dist,47.11,676.387,32,384,12.11
+ssl_resnext101_32x16d,46.89,2045.721,96,224,194.03
+swsl_resnext101_32x16d,46.82,2049.006,96,224,194.03
+ig_resnext101_32x16d,46.52,2062.16,96,224,194.03
+resnetrs350,46.21,512.496,24,384,163.96
+regnetx_320,45.87,2789.512,128,224,107.81
+cait_s24_384,45.29,350.995,16,384,47.06
+eca_nfnet_l2,44.43,1438.009,64,384,56.72
+xcit_large_24_p16_384_dist,43.19,552.749,24,384,189.1
+efficientnet_b6,42.65,372.523,16,528,43.04
+tf_efficientnet_b6,41.27,384.949,16,528,43.04
+tf_efficientnet_b6_ap,41.07,386.945,16,528,43.04
+tf_efficientnet_b6_ns,41.05,387.133,16,528,43.04
+nfnet_f1,39.56,2424.624,96,320,132.63
+nfnet_f1s,39.31,2439.907,96,320,132.63
+vit_huge_patch14_224,38.12,417.919,16,224,632.05
+dm_nfnet_f1,37.77,1692.371,64,320,132.63
+swin_large_patch4_window12_384,36.73,433.938,16,384,196.74
+efficientnet_b7,35.91,219.519,8,600,66.35
+efficientnetv2_xl,35.88,329.245,12,512,208.12
+tf_efficientnetv2_xl_in21ft1k,35.79,329.897,12,512,208.12
+tf_efficientnet_b7,35.55,221.854,8,600,66.35
+tf_efficientnet_b7_ns,35.4,222.801,8,600,66.35
+tf_efficientnet_b7_ap,35.26,223.637,8,600,66.35
+convnext_xlarge_384_in22ft1k,34.11,467.466,16,384,350.2
+resnetrs420,31.93,492.311,16,416,191.89
+xcit_large_24_p8_224_dist,31.85,499.415,16,224,188.93
+xcit_large_24_p8_224,31.84,499.581,16,224,188.93
+resnest200e,31.57,1515.738,48,320,70.2
+densenet264d_iabn,31.46,4064.91,128,224,72.74
+resnetv2_50x3_bitm,30.97,515.904,16,448,217.32
+cait_s36_384,29.83,399.013,12,384,68.37
+resnetv2_152x2_bit_teacher_384,29.16,409.309,12,384,236.34
+vit_large_patch16_384,26.62,299.125,8,384,304.72
+tresnet_l_448,25.11,5094.194,128,448,55.99
+xcit_small_24_p8_384_dist,25.02,476.66,12,384,47.63
+eca_nfnet_l3,23.96,1332.494,32,448,72.04
+efficientnet_cc_b0_4e,23.7,40.469,1,224,13.31
+efficientnet_cc_b0_8e,22.94,41.864,1,224,24.01
+tresnet_xl_448,22.71,4223.097,96,448,78.44
+beit_large_patch16_384,22.62,351.795,8,384,305.0
+tf_efficientnet_cc_b0_8e,22.37,43.033,1,224,24.01
+nfnet_f2,21.97,2181.861,48,352,193.78
+nfnet_f2s,21.86,2193.026,48,352,193.78
+vit_giant_patch14_224,21.57,368.599,8,224,1012.61
+dm_nfnet_f2,21.42,1490.949,32,352,193.78
+tf_efficientnet_cc_b0_4e,21.27,45.221,1,224,13.31
+resnetv2_152x2_bitm,21.19,375.422,8,448,236.34
+efficientnet_cc_b1_8e,18.42,52.102,1,240,39.72
+xcit_medium_24_p8_384_dist,17.99,441.849,8,384,84.32
+ig_resnext101_32x32d,17.78,1798.209,32,224,468.53
+tf_efficientnet_cc_b1_8e,17.44,55.209,1,240,39.72
+resnetv2_101x3_bitm,17.08,466.831,8,448,387.93
+resnest269e,12.91,1852.797,24,416,110.93
+nfnet_f3,11.98,1999.215,24,416,254.92
+nfnet_f3s,11.88,2016.77,24,416,254.92
+dm_nfnet_f3,11.64,2058.707,24,416,254.92
+cait_m36_384,11.45,520.837,6,384,271.22
+efficientnet_b8,11.27,528.827,6,672,87.41
+tf_efficientnet_b8_ap,11.23,530.563,6,672,87.41
+tf_efficientnet_b8,11.1,537.074,6,672,87.41
+xcit_large_24_p8_384_dist,10.94,545.597,6,384,188.93
+convmixer_1536_20,9.34,5136.027,48,224,51.63
+tf_efficientnet_l2_ns_475,8.79,336.325,3,475,480.31
+ig_resnext101_32x48d,8.38,1908.292,16,224,828.41
+beit_large_patch16_512,8.36,357.133,3,512,305.67
+nfnet_f4s,6.37,1879.688,12,512,316.07
+nfnet_f4,6.34,1889.493,12,512,316.07
+dm_nfnet_f4,6.16,1941.969,12,512,316.07
+cait_m48_448,4.64,426.785,2,448,356.46
+nfnet_f5,4.56,1747.564,8,544,377.21
+nfnet_f5s,4.56,1750.367,8,544,377.21
+dm_nfnet_f5,4.43,1800.804,8,544,377.21
+nfnet_f6,3.47,1722.95,6,576,438.36
+nfnet_f6s,3.47,1723.469,6,576,438.36
+dm_nfnet_f6,3.34,1791.721,6,576,438.36
+resnetv2_152x4_bitm,3.17,313.58,1,480,936.53
+nfnet_f7,2.64,1507.256,4,608,499.5
+nfnet_f7s,2.63,1515.779,4,608,499.5
+efficientnet_l2,2.07,477.813,1,800,480.31
+tf_efficientnet_l2_ns,2.03,488.22,1,800,480.31

results/model_benchmark_amp_nchw_rtx3090.csv

@@ -1,528 +0,0 @@
-model,infer_samples_per_sec,infer_step_time,infer_batch_size,infer_img_size,infer_gmacs,infer_macts,param_count
-tf_mobilenetv3_small_minimal_100,23813.47,10.741,256,224,0.06,1.41,2.04
-tf_mobilenetv3_small_075,20226.39,12.646,256,224,0.05,1.3,2.04
-tf_mobilenetv3_small_100,18228.81,14.034,256,224,0.06,1.42,2.54
-levit_128s,17210.0,14.865,256,224,0.31,1.88,7.78
-regnetx_002,14906.56,17.164,256,224,0.2,2.16,2.68
-regnety_002,13464.5,19.003,256,224,0.2,2.17,3.16
-levit_128,11979.62,21.36,256,224,0.41,2.71,9.21
-levit_192,10405.48,24.593,256,224,0.66,3.2,10.95
-gernet_s,10172.35,25.156,256,224,0.75,2.65,8.17
-vit_small_patch32_224,9285.97,27.555,256,224,1.15,2.5,22.88
-regnetx_004,9188.53,27.851,256,224,0.4,3.14,5.16
-tf_mobilenetv3_large_minimal_100,9164.35,27.924,256,224,0.22,4.4,3.92
-tf_mobilenetv3_large_075,8667.81,29.525,256,224,0.16,4.0,3.99
-mobilenetv3_rw,8631.08,29.65,256,224,0.23,4.41,5.48
-vit_tiny_r_s16_p8_224,8546.38,29.942,256,224,0.44,2.06,6.34
-mobilenetv3_large_100_miil,8526.13,30.015,256,224,0.23,4.41,5.48
-mobilenetv3_large_100,8496.93,30.118,256,224,0.23,4.41,5.48
-gluon_resnet18_v1b,8461.63,30.244,256,224,1.82,2.48,11.69
-ssl_resnet18,8460.21,30.249,256,224,1.82,2.48,11.69
-resnet18,8386.36,30.514,256,224,1.82,2.48,11.69
-swsl_resnet18,8382.52,30.528,256,224,1.82,2.48,11.69
-ghostnet_100,8276.88,30.92,256,224,0.15,3.55,5.18
-levit_256,7773.58,32.922,256,224,1.13,4.23,18.89
-legacy_seresnet18,7701.78,33.229,256,224,1.82,2.49,11.78
-tf_mobilenetv3_large_100,7680.05,33.323,256,224,0.23,4.41,5.48
-regnetx_006,7603.45,33.658,256,224,0.61,3.98,6.2
-mobilenetv2_100,7541.65,33.934,256,224,0.31,6.68,3.5
-regnety_004,7368.29,34.733,256,224,0.41,3.89,4.34
-hardcorenas_a,7264.93,35.227,256,224,0.23,4.38,5.26
-hardcorenas_b,7208.11,35.505,256,224,0.26,5.09,5.18
-mnasnet_100,7142.85,35.829,256,224,0.33,5.46,4.38
-resnet18d,7065.39,36.221,256,224,2.06,3.29,11.71
-semnasnet_100,6753.19,37.897,256,224,0.32,6.23,3.89
-hardcorenas_c,6746.84,37.933,256,224,0.28,5.01,5.52
-spnasnet_100,6739.75,37.973,256,224,0.35,6.03,4.42
-regnety_006,6693.6,38.235,256,224,0.61,4.33,6.06
-hardcorenas_d,6572.55,38.939,256,224,0.3,4.93,7.5
-tf_efficientnetv2_b0,6314.13,40.533,256,224,0.73,4.77,7.14
-regnetx_008,6079.04,42.101,256,224,0.81,5.15,7.26
-efficientnet_lite0,5804.98,44.09,256,224,0.4,6.74,4.65
-dla46_c,5780.94,44.273,256,224,0.58,4.5,1.3
-mobilenetv2_110d,5723.57,44.717,256,224,0.45,8.71,4.52
-rexnet_100,5717.93,44.761,256,224,0.41,7.44,4.8
-hardcorenas_f,5617.23,45.564,256,224,0.35,5.57,8.2
-regnety_008,5508.59,46.462,256,224,0.81,5.25,6.26
-hardcorenas_e,5410.0,47.31,256,224,0.35,5.65,8.07
-fbnetc_100,5329.99,48.02,256,224,0.4,6.51,5.57
-skresnet18,5316.12,48.145,256,224,1.82,3.24,11.96
-tf_efficientnet_lite0,5240.32,48.842,256,224,0.4,6.74,4.65
-mobilenetv2_140,5070.12,50.481,256,224,0.6,9.57,6.11
-efficientnet_b0,5059.78,50.585,256,224,0.4,6.75,5.29
-ese_vovnet19b_dw,5050.16,50.68,256,224,1.34,8.25,6.54
-gluon_resnet34_v1b,4958.52,51.618,256,224,3.67,3.74,21.8
-efficientnet_b1_pruned,4954.57,51.658,256,240,0.4,6.21,6.33
-tv_resnet34,4937.29,51.84,256,224,3.67,3.74,21.8
-resnet34,4928.32,51.933,256,224,3.67,3.74,21.8
-hrnet_w18_small,4869.58,52.561,256,224,1.61,5.72,13.19
-levit_384,4654.29,54.993,256,224,2.36,6.26,39.13
-tf_efficientnet_b0_ap,4650.22,55.039,256,224,0.4,6.75,5.29
-tf_efficientnet_b0_ns,4646.91,55.076,256,224,0.4,6.75,5.29
-tf_efficientnet_b0,4644.25,55.108,256,224,0.4,6.75,5.29
-dla46x_c,4605.94,55.57,256,224,0.54,5.66,1.07
-selecsls42b,4570.17,56.005,256,224,2.98,4.62,32.46
-deit_tiny_patch16_224,4543.82,56.329,256,224,1.26,5.97,5.72
-vit_tiny_patch16_224,4538.07,56.399,256,224,1.26,5.97,5.72
-gernet_m,4516.03,56.676,256,224,3.02,5.24,21.14
-deit_tiny_distilled_patch16_224,4481.69,57.11,256,224,1.27,6.01,5.91
-legacy_seresnet34,4474.42,57.204,256,224,3.67,3.74,21.96
-resnet34d,4448.37,57.538,256,224,3.91,4.54,21.82
-pit_ti_distilled_224,4332.05,59.084,256,224,0.71,6.23,5.1
-pit_ti_224,4322.46,59.215,256,224,0.7,6.19,4.85
-dla60x_c,4302.12,59.495,256,224,0.59,6.01,1.32
-mixnet_s,4297.51,59.559,256,224,0.25,6.25,4.13
-rexnet_130,4236.61,60.415,256,224,0.68,9.71,7.56
-tf_efficientnetv2_b1,4231.93,60.481,256,240,1.21,7.34,8.14
-xcit_nano_12_p16_224_dist,4191.2,61.067,256,224,0.56,4.17,3.05
-xcit_nano_12_p16_224,4188.12,61.112,256,224,0.56,4.17,3.05
-resmlp_12_distilled_224,4137.96,61.855,256,224,3.01,5.5,15.35
-resmlp_12_224,4137.2,61.867,256,224,3.01,5.5,15.35
-resnet26,4135.23,61.895,256,224,2.36,7.35,16.0
-vit_base_patch32_sam_224,4111.04,62.261,256,224,4.41,5.01,88.22
-mobilenetv2_120d,4108.55,62.299,256,224,0.69,11.97,5.83
-vit_base_patch32_224,4102.94,62.384,256,224,4.41,5.01,88.22
-tf_mixnet_s,4033.2,63.462,256,224,0.25,6.25,4.13
-repvgg_b0,4020.07,63.669,256,224,3.41,6.15,15.82
-selecsls60b,3957.1,64.683,256,224,3.63,5.52,32.77
-selecsls60,3955.58,64.708,256,224,3.59,5.52,30.67
-resnet26d,3771.71,67.862,256,224,2.6,8.15,16.01
-dla34,3751.8,68.222,256,224,3.07,5.02,15.74
-rexnet_150,3693.73,69.295,256,224,0.9,11.21,9.73
-ecaresnet50d_pruned,3635.83,70.4,256,224,2.53,6.43,19.94
-tf_efficientnet_lite1,3541.96,72.266,256,240,0.62,10.14,5.42
-pit_xs_224,3506.77,72.991,256,224,1.4,7.71,10.62
-regnetx_016,3495.21,73.233,256,224,1.62,7.93,9.19
-pit_xs_distilled_224,3481.48,73.522,256,224,1.41,7.76,11.0
-efficientnet_es_pruned,3365.67,76.052,256,224,1.81,8.73,5.44
-efficientnet_es,3358.26,76.219,256,224,1.81,8.73,5.44
-efficientnet_b2_pruned,3344.4,76.535,256,260,0.73,9.13,8.31
-tf_efficientnet_es,3248.35,78.797,256,224,1.81,8.73,5.44
-tf_efficientnetv2_b2,3221.93,79.444,256,260,1.72,9.84,10.1
-resnest14d,3200.62,79.966,256,224,2.76,7.33,10.61
-gernet_l,3173.09,80.669,256,256,4.57,8.0,31.08
-regnety_016,3082.19,83.046,256,224,1.63,8.04,11.2
-tf_efficientnet_cc_b0_8e,3079.83,83.109,256,224,0.42,9.42,24.01
-tf_efficientnet_cc_b0_4e,3072.34,83.31,256,224,0.41,9.42,13.31
-mixnet_m,3041.78,84.15,256,224,0.36,8.19,5.01
-skresnet34,3025.8,84.595,256,224,3.67,5.13,22.28
-resnext26ts,2999.56,85.335,256,256,2.43,10.52,10.3
-repvgg_a2,2997.36,85.397,256,224,5.7,6.26,28.21
-legacy_seresnext26_32x4d,2982.2,85.832,256,224,2.49,9.39,16.79
-vit_tiny_r_s16_p8_384,2981.27,85.856,256,384,1.34,6.49,6.36
-vit_small_patch32_384,2975.01,86.035,256,384,3.45,8.25,22.92
-xcit_tiny_12_p16_224_dist,2962.26,86.406,256,224,1.24,6.29,6.72
-xcit_tiny_12_p16_224,2958.87,86.506,256,224,1.24,6.29,6.72
-resnet26t,2949.15,86.793,256,256,3.35,10.52,16.01
-seresnext26ts,2930.19,87.355,256,256,2.43,10.52,10.39
-eca_resnext26ts,2927.39,87.439,256,256,2.43,10.52,10.3
-tf_mixnet_m,2918.13,87.716,256,224,0.36,8.19,5.01
-tf_efficientnet_b1_ap,2909.55,87.973,256,240,0.71,10.88,7.79
-tf_efficientnet_b1_ns,2907.96,88.021,256,240,0.71,10.88,7.79
-tf_efficientnet_b1,2906.46,88.066,256,240,0.71,10.88,7.79
-gcresnext26ts,2860.35,89.489,256,256,2.43,10.53,10.48
-ecaresnet101d_pruned,2832.39,90.373,256,224,3.48,7.69,24.88
-efficientnet_b1,2816.51,90.881,256,256,0.77,12.22,7.79
-seresnext26t_32x4d,2803.22,91.313,256,224,2.7,10.09,16.81
-seresnext26d_32x4d,2791.34,91.701,256,224,2.73,10.19,16.81
-ecaresnetlight,2748.62,93.127,256,224,4.11,8.42,30.16
-tf_efficientnet_lite2,2734.03,93.624,256,260,0.89,12.9,6.09
-nf_regnet_b1,2722.26,94.028,256,288,1.02,9.2,10.22
-crossvit_tiny_240,2719.71,94.117,256,240,1.57,9.08,7.01
-rexnet_200,2697.36,94.896,256,224,1.56,14.91,16.37
-resnetv2_50,2678.82,95.552,256,224,4.11,11.11,25.55
-crossvit_9_240,2675.71,95.665,256,240,1.85,9.52,8.55
-eca_botnext26ts_256,2660.24,96.221,256,256,2.46,11.6,10.59
-tresnet_m,2657.69,96.314,256,224,5.74,7.31,31.39
-botnet26t_256,2648.39,96.646,256,256,3.32,11.98,12.49
-halonet26t,2611.82,98.005,256,256,3.19,11.69,12.48
-vgg11,2609.78,98.082,256,224,7.61,7.44,132.86
-eca_halonext26ts,2603.27,98.327,256,256,2.44,11.46,10.76
-gluon_resnet50_v1b,2595.8,98.61,256,224,4.11,11.11,25.56
-ssl_resnet50,2595.55,98.619,256,224,4.11,11.11,25.56
-efficientnet_b3_pruned,2594.03,98.677,256,300,1.04,11.86,9.86
-tv_resnet50,2591.78,98.763,256,224,4.11,11.11,25.56
-crossvit_9_dagger_240,2589.88,98.836,256,240,1.99,9.97,8.78
-resnet50,2584.36,99.039,256,224,4.11,11.11,25.56
-swsl_resnet50,2581.65,99.146,256,224,4.11,11.11,25.56
-convit_tiny,2544.18,100.61,256,224,1.26,7.94,5.71
-hrnet_w18_small_v2,2509.11,102.017,256,224,2.62,9.65,15.6
-resnet32ts,2503.04,102.263,256,256,4.63,11.58,17.96
-bat_resnext26ts,2479.68,103.219,256,256,2.53,12.51,10.73
-resnet33ts,2467.72,103.728,256,256,4.76,11.66,19.68
-gluon_resnet50_v1c,2464.02,103.884,256,224,4.35,11.92,25.58
-ese_vovnet39b,2457.44,104.162,256,224,7.09,6.74,24.57
-cspresnet50,2453.31,104.338,256,256,4.54,11.5,21.62
-cspresnext50,2451.69,104.407,256,224,3.1,12.14,20.57
-gluon_resnet50_v1d,2445.69,104.663,256,224,4.35,11.92,25.58
-resnet50d,2441.21,104.85,256,224,4.35,11.92,25.58
-dpn68b,2432.24,105.242,256,224,2.35,10.47,12.61
-legacy_seresnet50,2426.52,105.49,256,224,3.88,10.6,28.09
-dpn68,2406.35,106.374,256,224,2.35,10.47,12.61
-eca_resnet33ts,2405.01,106.434,256,256,4.76,11.66,19.68
-seresnet33ts,2402.61,106.54,256,256,4.76,11.66,19.78
-vgg11_bn,2383.08,107.413,256,224,7.62,7.44,132.87
-mixnet_l,2358.59,108.528,256,224,0.58,10.84,7.33
-lambda_resnet26t,2358.23,108.545,256,256,3.02,11.87,10.96
-gcresnet33ts,2347.52,109.04,256,256,4.76,11.68,19.88
-pit_s_224,2332.08,109.763,256,224,2.88,11.56,23.46
-dla60,2324.78,110.106,256,224,4.26,10.16,22.04
-seresnet50,2316.85,110.484,256,224,4.11,11.13,28.09
-resnest26d,2313.67,110.634,256,224,3.64,9.97,17.07
-pit_s_distilled_224,2311.97,110.718,256,224,2.9,11.64,24.04
-deit_small_patch16_224,2297.25,111.426,256,224,4.61,11.95,22.05
-vit_small_patch16_224,2293.15,111.622,256,224,4.61,11.95,22.05
-deit_small_distilled_patch16_224,2268.28,112.85,256,224,4.63,12.02,22.44
-tf_mixnet_l,2264.19,113.053,256,224,0.58,10.84,7.33
-tf_efficientnet_b2_ns,2256.47,113.438,256,260,1.02,13.83,9.11
-tf_efficientnet_b2_ap,2256.28,113.446,256,260,1.02,13.83,9.11
-tf_efficientnet_b2,2253.15,113.605,256,260,1.02,13.83,9.11
-tv_densenet121,2246.15,113.963,256,224,2.87,6.9,7.98
-densenet121,2241.22,114.212,256,224,2.87,6.9,7.98
-res2net50_48w_2s,2234.19,114.57,256,224,4.18,11.72,25.29
-ecaresnet50d,2188.72,116.953,256,224,4.35,11.93,25.58
-resnetblur50,2181.37,117.342,256,224,5.16,12.02,25.56
-haloregnetz_b,2181.01,117.365,256,224,1.97,11.94,11.68
-resnetrs50,2148.74,119.126,256,224,4.48,12.14,35.69
-gluon_resnet50_v1s,2124.85,120.469,256,224,5.47,13.52,25.68
-visformer_small,2123.42,120.549,256,224,4.88,11.43,40.22
-gluon_inception_v3,2118.5,120.829,256,299,5.73,8.97,23.83
-regnetx_032,2117.36,120.894,256,224,3.2,11.37,15.3
-resmlp_24_distilled_224,2109.62,121.337,256,224,5.96,10.91,30.02
-efficientnet_b2,2109.39,121.351,256,288,1.12,16.2,9.11
-inception_v3,2109.39,121.351,256,299,5.73,8.97,23.83
-resmlp_24_224,2108.77,121.386,256,224,5.96,10.91,30.02
-tf_inception_v3,2108.49,121.403,256,299,5.73,8.97,23.83
-adv_inception_v3,2107.2,121.474,256,299,5.73,8.97,23.83
-efficientnet_em,2078.91,123.131,256,240,3.04,14.34,6.9
-densenetblur121d,2047.01,125.05,256,224,3.11,7.9,8.0
-tf_efficientnet_em,2016.96,126.911,256,240,3.04,14.34,6.9
-ssl_resnext50_32x4d,2014.12,127.092,256,224,4.26,14.4,25.03
-tv_resnext50_32x4d,2011.71,127.243,256,224,4.26,14.4,25.03
-gluon_resnext50_32x4d,2008.79,127.428,256,224,4.26,14.4,25.03
-swsl_resnext50_32x4d,2003.89,127.736,256,224,4.26,14.4,25.03
-resnext50_32x4d,2003.33,127.776,256,224,4.26,14.4,25.03
-tf_efficientnet_cc_b1_8e,1964.78,130.28,256,240,0.75,15.44,39.72
-dla60x,1943.55,131.706,256,224,3.54,13.8,17.35
-regnetx_040,1936.18,132.208,256,224,3.99,12.2,22.12
-resnext50d_32x4d,1913.78,133.755,256,224,4.5,15.2,25.05
-res2net50_26w_4s,1909.57,134.046,256,224,4.28,12.61,25.7
-tf_efficientnetv2_b3,1906.95,134.234,256,300,3.04,15.74,14.36
-regnety_040,1904.74,134.39,256,224,4.0,12.29,20.65
-gcresnet50t,1896.24,134.993,256,256,5.42,14.67,25.9
-efficientnetv2_rw_t,1892.3,135.274,256,288,3.19,16.42,13.65
-sehalonet33ts,1872.56,136.7,256,256,3.55,14.7,13.69
-lambda_resnet26rpt_256,1852.61,138.171,256,256,3.16,11.87,10.99
-seresnext50_32x4d,1837.91,139.277,256,224,4.26,14.42,27.56
-gcresnext50ts,1836.72,139.369,256,256,3.75,15.46,15.67
-dla60_res2net,1835.75,139.44,256,224,4.15,12.34,20.85
-gluon_seresnext50_32x4d,1834.84,139.511,256,224,4.26,14.42,27.56
-legacy_seresnext50_32x4d,1831.03,139.801,256,224,4.26,14.42,27.56
-resnest50d_1s4x24d,1825.82,140.197,256,224,4.43,13.57,25.68
-repvgg_b1g4,1816.81,140.887,256,224,8.15,10.64,39.97
-densenet169,1814.19,141.099,256,224,3.4,7.3,14.15
-gc_efficientnetv2_rw_t,1807.69,141.606,256,288,3.2,16.45,13.68
-coat_lite_tiny,1802.74,141.995,256,224,1.6,11.65,5.72
-res2net50_14w_8s,1801.06,142.125,256,224,4.21,13.28,25.06
-res2next50,1791.13,142.906,256,224,4.2,13.71,24.67
-cspdarknet53,1752.54,146.063,256,256,6.57,16.81,27.64
-ecaresnet26t,1748.79,146.375,256,320,5.24,16.44,16.01
-dla60_res2next,1747.0,146.526,256,224,3.49,13.17,17.03
-regnetz_b,1724.41,148.443,256,288,2.39,16.43,9.72
-coat_lite_mini,1717.23,149.066,256,224,2.0,12.25,11.01
-vgg13,1708.54,149.824,256,224,11.31,12.25,133.05
-mixnet_xl,1689.17,151.542,256,224,0.93,14.57,11.9
-vit_small_r26_s32_224,1664.97,153.742,256,224,3.56,9.85,36.43
-resnetv2_101,1630.91,156.956,256,224,7.83,16.23,44.54
-tf_efficientnet_lite3,1624.63,157.563,256,300,1.65,21.85,8.2
-gluon_resnet101_v1b,1601.1,159.88,256,224,7.83,16.23,44.55
-tv_resnet101,1599.29,160.06,256,224,7.83,16.23,44.55
-repvgg_b1,1595.1,160.478,256,224,13.16,10.64,57.42
-xcit_tiny_24_p16_224_dist,1586.68,161.331,256,224,2.34,11.82,12.12
-xcit_tiny_24_p16_224,1585.43,161.458,256,224,2.34,11.82,12.12
-xcit_small_12_p16_224_dist,1555.24,164.59,256,224,4.82,12.58,26.25
-lambda_resnet50ts,1554.98,164.616,256,256,5.07,17.48,21.54
-xcit_small_12_p16_224,1551.41,164.997,256,224,4.82,12.58,26.25
-gluon_resnet101_v1c,1549.68,165.185,256,224,8.08,17.04,44.57
-resnest50d,1546.83,165.485,256,224,5.4,14.36,27.48
-vgg13_bn,1545.98,165.579,256,224,11.33,12.25,133.05
-gluon_resnet101_v1d,1542.35,165.971,256,224,8.08,17.04,44.57
-dla102,1540.91,166.124,256,224,7.19,14.18,33.27
-twins_svt_small,1527.64,167.568,256,224,2.94,13.75,24.06
-wide_resnet50_2,1520.69,168.321,256,224,11.43,14.4,68.88
-gmixer_24_224,1499.34,170.73,256,224,5.28,14.45,24.72
-resnetv2_50x1_bit_distilled,1491.85,171.586,256,224,4.23,11.11,25.55
-regnetx_080,1490.15,171.784,256,224,8.02,14.06,39.57
-xcit_nano_12_p16_384_dist,1485.24,172.347,256,384,1.64,12.15,3.05
-crossvit_small_240,1471.27,173.987,256,240,5.63,18.17,26.86
-legacy_seresnet101,1468.28,174.342,256,224,7.61,15.74,49.33
-halonet50ts,1455.95,175.819,256,256,5.3,19.2,22.73
-res2net50_26w_6s,1449.56,176.589,256,224,6.33,15.28,37.05
-regnetx_064,1419.83,180.292,256,224,6.49,16.37,26.21
-densenet201,1416.48,180.718,256,224,4.34,7.85,20.01
-resmlp_36_224,1415.51,180.84,256,224,8.91,16.33,44.69
-resmlp_36_distilled_224,1415.35,180.86,256,224,8.91,16.33,44.69
-gluon_resnet101_v1s,1409.69,181.589,256,224,9.19,18.64,44.67
-lamhalobotnet50ts_256,1390.79,184.054,256,256,5.02,18.44,22.57
-nf_resnet50,1387.87,184.443,256,288,6.88,18.37,25.56
-ecaresnet101d,1376.05,186.03,256,224,8.08,17.07,44.57
-vgg16,1375.43,186.112,256,224,15.47,13.56,138.36
-hrnet_w18,1354.53,188.983,256,224,4.32,16.31,21.3
-regnety_032,1350.9,189.492,256,288,5.29,18.61,19.44
-crossvit_15_240,1346.42,190.123,256,240,5.81,19.77,27.53
-tresnet_l,1344.95,190.33,256,224,10.88,11.9,55.99
-gmlp_s16_224,1341.95,190.757,256,224,4.42,15.1,19.42
-tf_efficientnet_b3_ap,1337.77,191.35,256,300,1.87,23.83,12.23
-tf_efficientnet_b3_ns,1337.21,191.429,256,300,1.87,23.83,12.23
-tf_efficientnet_b3,1336.96,191.464,256,300,1.87,23.83,12.23
-resnet51q,1335.36,191.695,256,288,8.07,20.94,35.7
-vit_base_patch32_384,1326.76,192.939,256,384,13.06,16.5,88.3
-xception,1314.05,194.804,256,299,8.4,35.83,22.86
-crossvit_15_dagger_240,1305.29,196.113,256,240,6.13,20.43,28.21
-dla102x,1302.51,196.531,256,224,5.89,19.42,26.31
-efficientnet_b3,1297.94,197.224,256,320,2.01,26.52,12.23
-mixer_b16_224,1285.59,199.119,256,224,12.62,14.53,59.88
-mixer_b16_224_miil,1283.81,199.396,256,224,12.62,14.53,59.88
-cait_xxs24_224,1283.41,199.457,256,224,2.53,20.29,11.96
-skresnext50_32x4d,1277.72,200.346,256,224,4.5,17.18,27.48
-regnety_064,1270.37,201.504,256,224,6.39,16.41,30.58
-ssl_resnext101_32x4d,1267.29,201.994,256,224,8.01,21.23,44.18
-gluon_resnext101_32x4d,1262.61,202.744,256,224,8.01,21.23,44.18
-swsl_resnext101_32x4d,1261.41,202.934,256,224,8.01,21.23,44.18
-vgg16_bn,1259.82,203.192,256,224,15.5,13.56,138.37
-repvgg_b2g4,1250.78,204.659,256,224,12.63,12.9,61.76
-halo2botnet50ts_256,1247.07,205.268,256,256,5.02,21.78,22.64
-swin_tiny_patch4_window7_224,1247.03,205.273,256,224,4.51,17.06,28.29
-twins_pcpvt_small,1237.18,206.911,256,224,3.83,18.08,24.11
-regnety_080,1231.05,207.941,256,224,8.0,17.97,39.18
-resnest50d_4s2x40d,1214.34,210.797,256,224,4.4,17.94,30.42
-resnet61q,1207.53,211.987,256,288,9.87,21.52,36.85
-eca_nfnet_l0,1191.71,214.805,256,288,7.12,17.29,24.14
-nfnet_l0,1190.26,215.067,256,288,7.13,17.29,35.07
-res2net101_26w_4s,1183.89,216.22,256,224,8.1,18.45,45.21
-dpn92,1178.74,217.169,256,224,6.54,18.21,37.67
-res2net50_26w_8s,1178.7,217.174,256,224,8.37,17.95,48.4
-vit_tiny_patch16_384,1172.78,218.271,256,384,4.7,25.39,5.79
-convit_small,1169.52,218.881,256,224,5.76,17.87,27.78
-vgg19,1152.13,222.185,256,224,19.63,14.86,143.67
-gluon_seresnext101_32x4d,1150.11,222.576,256,224,8.02,21.26,48.96
-legacy_seresnext101_32x4d,1149.76,222.644,256,224,8.02,21.26,48.96
-hrnet_w32,1142.24,224.109,256,224,8.97,22.02,41.23
-xcit_nano_12_p8_224,1127.98,226.94,256,224,2.16,15.71,3.05
-xcit_nano_12_p8_224_dist,1127.81,226.974,256,224,2.16,15.71,3.05
-hrnet_w30,1123.84,227.778,256,224,8.15,21.21,37.71
-tv_resnet152,1116.24,229.331,256,224,11.56,22.56,60.19
-gluon_resnet152_v1b,1115.71,229.44,256,224,11.56,22.56,60.19
-ecaresnet50t,1110.54,230.506,256,320,8.82,24.13,25.57
-regnetz_c,1099.38,232.841,256,320,3.92,25.88,13.46
-gluon_resnet152_v1c,1090.66,234.708,256,224,11.8,23.36,60.21
-gluon_resnet152_v1d,1087.27,235.441,256,224,11.8,23.36,60.21
-repvgg_b2,1087.24,235.444,256,224,20.45,12.9,89.02
-xception41,1068.11,239.655,256,299,9.28,39.86,26.97
-densenet161,1067.45,239.812,256,224,7.79,11.06,28.68
-vgg19_bn,1061.8,241.089,256,224,19.66,14.86,143.68
-inception_v4,1057.29,242.117,256,299,12.28,15.09,42.68
-convmixer_1024_20_ks9_p14,1048.55,244.135,256,224,5.55,5.51,24.38
-dla169,1043.9,245.223,256,224,11.6,20.2,53.39
-xcit_tiny_12_p16_384_dist,1041.14,245.869,256,384,3.64,18.26,6.72
-regnetx_120,1028.54,248.883,256,224,12.13,21.37,46.11
-gluon_resnet152_v1s,1019.74,251.033,256,224,12.92,24.96,60.32
-coat_lite_small,1019.01,251.211,256,224,3.96,22.09,19.84
-vit_base_patch16_224_miil,1015.29,252.134,256,224,17.58,23.9,86.54
-legacy_seresnet152,1014.16,252.414,256,224,11.33,22.08,66.82
-repvgg_b3g4,1009.72,253.523,256,224,17.89,15.1,83.83
-jx_nest_tiny,1009.0,253.705,256,224,5.83,25.48,17.06
-crossvit_18_240,995.7,257.092,256,240,9.05,26.26,43.27
-vit_base_patch16_224,987.74,259.165,256,224,17.58,23.9,86.57
-vit_base_patch16_sam_224,983.41,260.307,256,224,17.58,23.9,86.57
-tresnet_xl,983.24,260.352,256,224,15.17,15.34,78.44
-deit_base_patch16_224,981.95,260.694,256,224,17.58,23.9,86.57
-regnety_120,981.03,260.937,256,224,12.14,21.38,51.82
-deit_base_distilled_patch16_224,972.1,263.335,256,224,17.68,24.05,87.34
-crossvit_18_dagger_240,968.8,264.234,256,240,9.5,27.03,44.27
-efficientnet_el_pruned,927.76,275.921,256,300,8.0,30.7,10.59
-efficientnet_el,927.59,275.971,256,300,8.0,30.7,10.59
-tf_efficientnet_el,908.45,281.783,256,300,8.0,30.7,10.59
-beit_base_patch16_224,907.03,282.225,256,224,17.58,23.9,86.53
-dm_nfnet_f0,906.74,282.318,256,256,12.62,18.05,71.49
-beit_base_patch16_224_in22k,904.43,283.033,256,224,17.6,23.92,102.56
-twins_pcpvt_base,903.59,283.303,256,224,6.68,25.25,43.83
-dla102x2,892.39,286.858,256,224,9.34,29.91,41.28
-twins_svt_base,884.1,289.548,256,224,8.59,26.33,56.07
-wide_resnet101_2,879.83,290.95,256,224,22.8,21.23,126.89
-tf_efficientnetv2_s_in21ft1k,867.65,295.037,256,384,8.44,35.77,21.46
-tf_efficientnetv2_s,866.61,295.392,256,384,8.44,35.77,21.46
-cait_xxs36_224,864.05,296.268,256,224,3.77,30.34,17.3
-resnetrs101,850.72,300.905,256,288,13.56,28.53,63.62
-repvgg_b3,845.91,302.619,256,224,29.16,15.1,123.09
-efficientnetv2_rw_s,844.69,303.059,256,384,8.72,38.03,23.94
-dpn98,840.2,304.675,256,224,11.73,25.2,61.57
-pit_b_distilled_224,838.63,305.249,256,224,12.5,33.07,74.79
-pit_b_224,837.02,305.836,256,224,12.42,32.94,73.76
-ens_adv_inception_resnet_v2,834.35,306.814,256,299,13.18,25.06,55.84
-regnetx_160,833.02,307.301,256,224,15.99,25.52,54.28
-inception_resnet_v2,832.69,307.427,256,299,13.18,25.06,55.84
-xcit_small_24_p16_224_dist,830.32,308.297,256,224,9.1,23.64,47.67
-xcit_small_24_p16_224,828.74,308.886,256,224,9.1,23.64,47.67
-swin_small_patch4_window7_224,809.9,316.074,256,224,8.77,27.47,49.61
-gluon_resnext101_64x4d,803.65,318.536,256,224,15.52,31.21,83.46
-xcit_tiny_12_p8_224_dist,791.87,323.27,256,224,4.81,23.6,6.71
-xcit_tiny_12_p8_224,791.64,323.364,256,224,4.81,23.6,6.71
-gluon_xception65,784.11,326.474,256,299,13.96,52.48,39.92
-xception65,777.91,329.074,256,299,13.96,52.48,39.92
-resnet101d,776.37,329.723,256,320,16.48,34.77,44.57
-ig_resnext101_32x8d,775.21,330.222,256,224,16.48,31.21,88.79
-swsl_resnext101_32x8d,773.83,330.808,256,224,16.48,31.21,88.79
-resnext101_32x8d,773.08,331.13,256,224,16.48,31.21,88.79
-ssl_resnext101_32x8d,769.28,332.765,256,224,16.48,31.21,88.79
-tf_efficientnet_lite4,764.13,335.01,256,380,4.04,45.66,13.01
-hrnet_w40,760.18,336.752,256,224,12.75,25.29,57.56
-gluon_seresnext101_64x4d,756.43,338.418,256,224,15.53,31.25,88.23
-resnest101e,755.07,339.024,256,256,13.38,28.66,48.28
-hrnet_w48,716.4,357.329,256,224,17.34,28.56,77.47
-cait_s24_224,714.86,358.098,256,224,9.35,40.58,46.92
-tresnet_m_448,713.55,358.758,256,448,22.94,29.21,31.39
-coat_tiny,707.04,362.062,256,224,4.35,27.2,5.5
-regnetz_d,704.94,363.138,256,320,9.33,37.08,27.58
-hrnet_w44,702.53,364.385,256,224,14.94,26.92,67.06
-vit_large_r50_s32_224,680.11,376.394,256,224,19.58,24.41,328.99
-jx_nest_small,676.97,378.142,256,224,10.35,40.04,38.35
-twins_svt_large,673.22,380.252,256,224,15.15,35.1,99.27
-crossvit_base_240,671.16,381.416,256,240,21.22,36.33,105.03
-efficientnet_b4,667.18,383.692,256,384,4.51,50.04,19.34
-twins_pcpvt_large,657.69,389.226,256,224,9.84,35.82,60.99
-tf_efficientnet_b4,637.24,401.719,256,380,4.49,49.49,19.34
-tf_efficientnet_b4_ap,636.68,402.068,256,380,4.49,49.49,19.34
-tf_efficientnet_b4_ns,636.62,402.109,256,380,4.49,49.49,19.34
-convit_base,621.76,411.725,256,224,17.52,31.77,86.54
-dpn131,619.15,413.454,256,224,16.09,32.97,79.25
-swin_base_patch4_window7_224,617.4,414.625,256,224,15.47,36.63,87.77
-swin_base_patch4_window7_224_in22k,616.51,415.228,256,224,15.49,36.65,109.13
-xcit_medium_24_p16_224_dist,613.3,417.401,256,224,16.13,31.71,84.4
-xcit_medium_24_p16_224,612.89,417.675,256,224,16.13,31.71,84.4
-vit_small_patch16_384,594.17,430.838,256,384,15.52,50.78,22.2
-coat_mini,592.46,432.084,256,224,6.82,33.68,10.34
-xception71,590.14,433.78,256,299,18.09,69.92,42.34
-vit_small_r26_s32_384,574.75,445.396,256,384,10.43,29.85,36.47
-hrnet_w64,570.22,448.937,256,224,28.97,35.09,128.06
-dpn107,566.19,452.128,256,224,18.38,33.46,86.92
-eca_nfnet_l1,559.71,457.367,256,320,14.92,34.42,41.41
-gluon_senet154,559.51,457.529,256,224,20.77,38.69,115.09
-legacy_senet154,558.72,458.178,256,224,20.77,38.69,115.09
-xcit_tiny_24_p16_384_dist,556.09,460.342,256,384,6.87,34.29,12.12
-xcit_small_12_p16_384_dist,546.87,468.099,256,384,14.14,36.51,26.25
-resnet152d,546.29,468.597,256,320,24.08,47.67,60.21
-regnety_320,522.26,490.163,256,224,32.34,30.26,145.05
-jx_nest_base,511.39,500.584,256,224,17.96,53.39,67.72
-regnety_160,509.43,502.501,256,288,26.37,38.07,83.59
-tnt_s_patch16_224,497.77,514.282,256,224,5.24,24.37,23.76
-seresnet152d,489.63,522.829,256,320,24.09,47.72,66.84
-resnetrs152,483.98,528.928,256,320,24.34,48.14,86.62
-regnetx_320,460.34,556.1,256,224,31.81,36.3,107.81
-vit_large_patch32_384,457.67,559.348,256,384,45.31,43.86,306.63
-mixer_l16_224,429.82,595.582,256,224,44.6,41.69,208.2
-xcit_tiny_24_p8_224,417.75,612.797,256,224,9.21,45.39,12.11
-xcit_tiny_24_p8_224_dist,417.56,613.068,256,224,9.21,45.39,12.11
-xcit_small_12_p8_224,416.17,615.111,256,224,18.69,47.21,26.21
-xcit_small_12_p8_224_dist,415.93,615.473,256,224,18.69,47.21,26.21
-efficientnetv2_rw_m,404.25,633.263,256,416,21.49,79.62,53.24
-resnet200d,391.43,653.991,256,320,31.25,67.33,64.69
-resnetv2_50x1_bitm,388.67,658.636,256,448,16.62,44.46,25.55
-xcit_nano_12_p8_384_dist,387.79,660.138,256,384,6.34,46.08,3.05
-swin_large_patch4_window7_224,385.11,664.722,256,224,34.53,54.94,196.53
-swin_large_patch4_window7_224_in22k,384.77,665.318,256,224,34.56,54.96,228.57
-xcit_large_24_p16_224,374.96,682.722,256,224,35.86,47.27,189.1
-xcit_large_24_p16_224_dist,374.65,683.281,256,224,35.86,47.27,189.1
-ssl_resnext101_32x16d,366.28,698.897,256,224,36.27,51.18,194.03
-ig_resnext101_32x16d,365.74,699.945,256,224,36.27,51.18,194.03
-swsl_resnext101_32x16d,365.46,700.478,256,224,36.27,51.18,194.03
-tresnet_l_448,350.13,731.146,256,448,43.5,47.56,55.99
-resnetrs200,346.06,739.735,256,320,31.51,67.81,93.21
-tf_efficientnetv2_m,339.58,753.868,256,480,24.76,89.84,54.14
-tf_efficientnetv2_m_in21ft1k,339.57,753.879,256,480,24.76,89.84,54.14
-vit_large_patch16_224,332.37,770.202,256,224,61.6,63.52,304.33
-dm_nfnet_f1,329.67,776.526,256,320,35.97,46.77,132.63
-tf_efficientnet_b5,319.74,800.623,256,456,10.46,98.86,30.39
-tf_efficientnet_b5_ap,319.51,801.207,256,456,10.46,98.86,30.39
-tf_efficientnet_b5_ns,319.48,801.287,256,456,10.46,98.86,30.39
-crossvit_15_dagger_408,308.22,830.557,256,408,21.45,95.05,28.5
-beit_large_patch16_224,304.41,840.964,256,224,61.6,63.52,304.43
-beit_large_patch16_224_in22k,303.89,842.402,256,224,61.62,63.54,325.79
-xcit_small_24_p16_384_dist,291.33,878.717,256,384,26.72,68.58,47.67
-convmixer_768_32,284.41,900.105,256,224,19.55,25.95,21.11
-eca_nfnet_l2,277.7,921.837,256,384,30.05,68.28,56.72
-resnetv2_152x2_bit_teacher,272.08,940.888,256,224,46.95,45.11,236.34
-xcit_tiny_12_p8_384_dist,271.29,943.618,256,384,14.13,69.14,6.71
-tresnet_xl_448,265.72,963.421,256,448,60.65,61.31,78.44
-vit_base_patch16_384,260.82,981.51,256,384,55.54,101.56,86.86
-deit_base_patch16_384,260.27,983.563,256,384,55.54,101.56,86.86
-resnest200e,257.06,995.856,256,320,35.69,82.78,70.2
-deit_base_distilled_patch16_384,256.78,996.936,256,384,55.65,101.82,87.63
-resnetv2_101x1_bitm,243.19,1052.661,256,448,31.65,64.93,44.54
-cait_xxs24_384,238.44,1073.626,256,384,9.63,122.66,12.03
-ecaresnet269d,229.27,1116.568,256,352,50.25,101.25,102.09
-crossvit_18_dagger_408,229.17,558.535,128,408,32.47,124.87,44.61
-vit_large_r50_s32_384,229.1,1117.419,256,384,57.43,76.52,329.09
-nasnetalarge,226.47,1130.365,256,331,23.89,90.56,88.75
-pnasnet5large,225.65,1134.478,256,331,25.04,92.89,86.06
-beit_base_patch16_384,223.91,1143.322,256,384,55.54,101.56,86.74
-resnetrs270,222.05,1152.857,256,352,51.13,105.48,129.86
-xcit_small_24_p8_224,218.32,1172.571,256,224,35.81,90.78,47.63
-xcit_small_24_p8_224_dist,218.2,1173.193,256,224,35.81,90.78,47.63
-xcit_medium_24_p16_384_dist,212.67,1203.725,256,384,47.39,91.64,84.4
-resmlp_big_24_224,199.99,1280.033,256,224,100.23,87.31,129.14
-resmlp_big_24_224_in22ft1k,199.55,1282.862,256,224,100.23,87.31,129.14
-resmlp_big_24_distilled_224,199.5,1283.208,256,224,100.23,87.31,129.14
-tf_efficientnetv2_l,196.16,1305.054,256,480,56.4,157.99,118.52
-tf_efficientnetv2_l_in21ft1k,195.97,1306.333,256,480,56.4,157.99,118.52
-dm_nfnet_f2,187.32,1366.603,256,352,63.22,79.06,193.78
-tf_efficientnet_b6_ns,184.64,693.239,128,528,19.4,167.39,43.04
-tf_efficientnet_b6_ap,184.63,693.254,128,528,19.4,167.39,43.04
-tf_efficientnet_b6,184.56,693.543,128,528,19.4,167.39,43.04
-swin_base_patch4_window12_384,172.29,742.9,128,384,47.19,134.78,87.9
-swin_base_patch4_window12_384_in22k,172.09,743.782,128,384,47.21,134.8,109.27
-cait_xs24_384,168.76,1516.911,256,384,19.28,183.98,26.67
-vit_base_r50_s16_384,164.61,1555.149,256,384,67.43,135.03,98.95
-convmixer_1536_20,163.58,1564.936,256,224,48.68,33.03,51.63
-cait_xxs36_384,159.38,1606.206,256,384,14.35,183.7,17.37
-xcit_medium_24_p8_224,158.47,1615.434,256,224,63.53,121.23,84.32
-xcit_medium_24_p8_224_dist,158.47,1615.404,256,224,63.53,121.23,84.32
-resnetrs350,147.45,1736.143,256,384,77.59,154.74,163.96
-ig_resnext101_32x32d,144.94,1766.221,256,224,87.29,91.12,468.53
-xcit_tiny_24_p8_384_dist,140.02,1828.249,256,384,27.05,132.95,12.11
-xcit_small_12_p8_384_dist,138.32,1850.769,256,384,54.92,138.29,26.21
-cait_s24_384,130.37,1963.692,256,384,32.17,245.31,47.06
-xcit_large_24_p16_384_dist,128.67,1989.551,256,384,105.35,137.17,189.1
-tf_efficientnetv2_xl_in21ft1k,125.28,2043.358,256,512,93.85,247.32,208.12
-resnest269e,119.95,2134.143,256,416,77.69,171.98,110.93
-swin_large_patch4_window12_384,108.71,1177.454,128,384,104.08,202.16,196.74
-swin_large_patch4_window12_384_in22k,108.63,1178.305,128,384,104.11,202.18,228.77
-resnetrs420,107.0,2392.495,256,416,108.45,213.79,191.89
-tf_efficientnet_b7_ns,106.07,603.375,64,600,38.33,289.94,66.35
-tf_efficientnet_b7_ap,105.98,603.889,64,600,38.33,289.94,66.35
-tf_efficientnet_b7,105.95,604.03,64,600,38.33,289.94,66.35
-dm_nfnet_f3,102.63,2494.386,256,416,115.58,141.78,254.92
-xcit_large_24_p8_224,95.99,2666.968,256,224,141.23,181.56,188.93
-xcit_large_24_p8_224_dist,95.77,2673.153,256,224,141.23,181.56,188.93
-resnetv2_152x2_bit_teacher_384,94.95,2696.08,256,384,136.16,132.56,236.34
-resnetv2_50x3_bitm,93.71,1365.901,128,448,145.7,133.37,217.32
-vit_large_patch16_384,89.71,2853.476,256,384,191.21,270.24,304.72
-cait_s36_384,87.22,2935.166,256,384,47.99,367.4,68.37
-ig_resnext101_32x48d,85.91,1489.933,128,224,153.57,131.06,828.41
-beit_large_patch16_384,77.75,3292.734,256,384,191.21,270.24,305.0
-xcit_small_24_p8_384_dist,72.45,3533.673,256,384,105.24,265.91,47.63
-resnetv2_152x2_bitm,69.73,1835.714,128,448,184.99,180.43,236.34
-tf_efficientnet_b8,66.92,956.333,64,672,63.48,442.89,87.41
-tf_efficientnet_b8_ap,66.91,956.434,64,672,63.48,442.89,87.41
-dm_nfnet_f4,57.01,4490.519,256,512,216.26,262.26,316.07
-resnetv2_101x3_bitm,56.17,2278.772,128,448,280.33,194.78,387.93
-xcit_medium_24_p8_384_dist,53.81,4757.31,256,384,186.67,354.73,84.32
-dm_nfnet_f5,41.46,6174.452,256,544,290.97,349.71,377.21
-tf_efficientnet_l2_ns_475,38.52,1661.585,64,475,172.11,609.89,480.31
-xcit_large_24_p8_384_dist,32.44,7891.525,256,384,415.0,531.82,188.93
-beit_large_patch16_512,31.38,2039.285,64,512,362.24,656.39,305.67
-cait_m36_384,30.16,8488.214,256,384,173.11,734.81,271.22
-dm_nfnet_f6,30.03,8525.766,256,576,378.69,452.2,438.36
-resnetv2_152x4_bitm,18.21,3515.177,64,480,844.84,414.26,936.53
-tf_efficientnet_l2_ns,13.56,1770.059,24,800,479.12,1707.39,480.31
-cait_m48_448,13.08,9784.533,128,448,329.41,1708.23,356.46

results/model_metadata-in1k.csv

@@ -452,10 +452,10 @@ visformer_small,in1k
 vit_base_patch16_224,in21k
 vit_base_patch16_224_miil,in21k
 vit_base_patch16_384,in21k
-vit_base_patch16_sam_224,in1k
+vit_base_patch16_224_sam,in1k
 vit_base_patch32_224,in21k
 vit_base_patch32_384,in21k
-vit_base_patch32_sam_224,in1k
+vit_base_patch32_224_sam,in1k
 vit_base_r50_s16_384,in21k
 vit_large_patch16_224,in21k
 vit_large_patch16_384,in21k

results/results-imagenet-a-clean.csv

@@ -1,522 +1,553 @@
 model,top1,top1_err,top5,top5_err,param_count,img_size,cropt_pct,interpolation
 beit_large_patch16_512,98.560,1.440,99.840,0.160,305.67,512,1.000,bicubic
-tf_efficientnet_l2_ns,98.540,1.460,99.820,0.180,480.31,800,0.960,bicubic
-beit_large_patch16_384,98.500,1.500,99.820,0.180,305.00,384,1.000,bicubic
-tf_efficientnet_l2_ns_475,98.490,1.510,99.830,0.170,480.31,475,0.936,bicubic
+tf_efficientnet_l2_ns,98.550,1.450,99.820,0.180,480.31,800,0.960,bicubic
+beit_large_patch16_384,98.520,1.480,99.820,0.180,305.00,384,1.000,bicubic
+tf_efficientnet_l2_ns_475,98.500,1.500,99.830,0.170,480.31,475,0.936,bicubic
+convnext_xlarge_384_in22ft1k,98.350,1.650,99.800,0.200,350.20,384,1.000,bicubic
+convnext_large_384_in22ft1k,98.220,1.780,99.730,0.270,197.77,384,1.000,bicubic
 vit_large_patch16_384,98.210,1.790,99.800,0.200,304.72,384,1.000,bicubic
-beit_large_patch16_224,98.170,1.830,99.760,0.240,304.43,224,0.900,bicubic
-swin_large_patch4_window12_384,98.020,1.980,99.690,0.310,196.74,384,1.000,bicubic
-tf_efficientnet_b7_ns,97.880,2.120,99.720,0.280,66.35,600,0.949,bicubic
-swin_base_patch4_window12_384,97.870,2.130,99.710,0.290,87.90,384,1.000,bicubic
+beit_large_patch16_224,98.180,1.820,99.760,0.240,304.43,224,0.900,bicubic
+swin_large_patch4_window12_384,98.040,1.960,99.690,0.310,196.74,384,1.000,bicubic
+convnext_base_384_in22ft1k,97.950,2.050,99.650,0.350,88.59,384,1.000,bicubic
+convnext_xlarge_in22ft1k,97.920,2.080,99.680,0.320,350.20,224,0.875,bicubic
+tf_efficientnet_b7_ns,97.910,2.090,99.720,0.280,66.35,600,0.949,bicubic
+swin_base_patch4_window12_384,97.890,2.110,99.710,0.290,87.90,384,1.000,bicubic
 vit_large_r50_s32_384,97.860,2.140,99.670,0.330,329.09,384,1.000,bicubic
-beit_base_patch16_384,97.830,2.170,99.700,0.300,86.74,384,1.000,bicubic
-vit_base_patch16_384,97.830,2.170,99.670,0.330,86.86,384,1.000,bicubic
-tf_efficientnetv2_l_in21ft1k,97.680,2.320,99.670,0.330,118.52,480,1.000,bicubic
-tf_efficientnetv2_xl_in21ft1k,97.670,2.330,99.490,0.510,208.12,512,1.000,bicubic
-swin_large_patch4_window7_224,97.660,2.340,99.580,0.420,196.53,224,0.900,bicubic
+vit_base_patch16_384,97.840,2.160,99.670,0.330,86.86,384,1.000,bicubic
+convnext_large_in22ft1k,97.830,2.170,99.690,0.310,197.77,224,0.875,bicubic
+beit_base_patch16_384,97.810,2.190,99.700,0.300,86.74,384,1.000,bicubic
+tf_efficientnetv2_l_in21ft1k,97.700,2.300,99.670,0.330,118.52,480,1.000,bicubic
+swin_large_patch4_window7_224,97.650,2.350,99.580,0.420,196.53,224,0.900,bicubic
+tf_efficientnetv2_xl_in21ft1k,97.650,2.350,99.490,0.510,208.12,512,1.000,bicubic
 vit_large_patch16_224,97.640,2.360,99.590,0.410,304.33,224,0.900,bicubic
-ig_resnext101_32x48d,97.640,2.360,99.710,0.290,828.41,224,0.875,bilinear
 tf_efficientnet_b6_ns,97.630,2.370,99.580,0.420,43.04,528,0.942,bicubic
-dm_nfnet_f6,97.610,2.390,99.550,0.450,438.36,576,0.956,bicubic
-dm_nfnet_f4,97.570,2.430,99.510,0.490,316.07,512,0.951,bicubic
+ig_resnext101_32x48d,97.620,2.380,99.710,0.290,828.41,224,0.875,bilinear
+dm_nfnet_f6,97.600,2.400,99.550,0.450,438.36,576,0.956,bicubic
+dm_nfnet_f4,97.580,2.420,99.510,0.490,316.07,512,0.951,bicubic
+vit_base_patch8_224,97.580,2.420,99.670,0.330,86.58,224,0.900,bicubic
 dm_nfnet_f5,97.540,2.460,99.570,0.430,377.21,544,0.954,bicubic
-xcit_large_24_p8_384_dist,97.530,2.470,99.540,0.460,188.93,384,1.000,bicubic
-xcit_large_24_p16_384_dist,97.530,2.470,99.480,0.520,189.10,384,1.000,bicubic
-resnetv2_152x4_bitm,97.490,2.510,99.620,0.380,936.53,480,1.000,bilinear
-tf_efficientnet_b5_ns,97.490,2.510,99.630,0.370,30.39,456,0.934,bicubic
+xcit_large_24_p8_384_dist,97.520,2.480,99.540,0.460,188.93,384,1.000,bicubic
+xcit_large_24_p16_384_dist,97.520,2.480,99.480,0.520,189.10,384,1.000,bicubic
+tf_efficientnet_b5_ns,97.500,2.500,99.630,0.370,30.39,456,0.934,bicubic
+resnetv2_152x4_bitm,97.490,2.510,99.610,0.390,936.53,480,1.000,bilinear
 cait_m48_448,97.480,2.520,99.550,0.450,356.46,448,1.000,bicubic
 tf_efficientnetv2_m_in21ft1k,97.480,2.520,99.530,0.470,54.14,480,1.000,bicubic
+convnext_base_in22ft1k,97.470,2.530,99.610,0.390,88.59,224,0.875,bicubic
 cait_m36_384,97.400,2.600,99.510,0.490,271.22,384,1.000,bicubic
-ig_resnext101_32x32d,97.360,2.640,99.680,0.320,468.53,224,0.875,bilinear
+ig_resnext101_32x32d,97.370,2.630,99.680,0.320,468.53,224,0.875,bilinear
 dm_nfnet_f3,97.350,2.650,99.560,0.440,254.92,416,0.940,bicubic
-cait_s36_384,97.340,2.660,99.530,0.470,68.37,384,1.000,bicubic
-xcit_medium_24_p8_384_dist,97.300,2.700,99.510,0.490,84.32,384,1.000,bicubic
+cait_s36_384,97.330,2.670,99.530,0.470,68.37,384,1.000,bicubic
+xcit_medium_24_p8_384_dist,97.290,2.710,99.510,0.490,84.32,384,1.000,bicubic
+tf_efficientnetv2_l,97.280,2.720,99.550,0.450,118.52,480,1.000,bicubic
 xcit_medium_24_p16_384_dist,97.280,2.720,99.460,0.540,84.40,384,1.000,bicubic
-tf_efficientnetv2_l,97.270,2.730,99.550,0.450,118.52,480,1.000,bicubic
-swin_base_patch4_window7_224,97.260,2.740,99.530,0.470,87.77,224,0.900,bicubic
-xcit_small_24_p8_384_dist,97.250,2.750,99.610,0.390,47.63,384,1.000,bicubic
+swin_base_patch4_window7_224,97.250,2.750,99.530,0.470,87.77,224,0.900,bicubic
+xcit_small_24_p8_384_dist,97.240,2.760,99.610,0.390,47.63,384,1.000,bicubic
 xcit_small_12_p8_384_dist,97.230,2.770,99.480,0.520,26.21,384,1.000,bicubic
-tf_efficientnet_b8,97.210,2.790,99.500,0.500,87.41,672,0.954,bicubic
 swsl_resnext101_32x8d,97.200,2.800,99.570,0.430,88.79,224,0.875,bilinear
 tf_efficientnet_b7_ap,97.200,2.800,99.540,0.460,66.35,600,0.949,bicubic
+tf_efficientnet_b8,97.200,2.800,99.500,0.500,87.41,672,0.954,bicubic
+regnetz_e8,97.200,2.800,99.500,0.500,57.70,320,1.000,bicubic
 vit_base_r50_s16_384,97.180,2.820,99.560,0.440,98.95,384,1.000,bicubic
 tf_efficientnetv2_m,97.140,2.860,99.410,0.590,54.14,480,1.000,bicubic
-xcit_small_24_p16_384_dist,97.130,2.870,99.450,0.550,47.67,384,1.000,bicubic
+xcit_small_24_p16_384_dist,97.120,2.880,99.460,0.540,47.67,384,1.000,bicubic
 tf_efficientnet_b8_ap,97.110,2.890,99.660,0.340,87.41,672,0.954,bicubic
 eca_nfnet_l2,97.090,2.910,99.510,0.490,56.72,384,1.000,bicubic
-ecaresnet269d,97.090,2.910,99.470,0.530,102.09,352,1.000,bicubic
-cait_s24_384,97.080,2.920,99.430,0.570,47.06,384,1.000,bicubic
+beit_base_patch16_224,97.090,2.910,99.610,0.390,86.53,224,0.900,bicubic
+ecaresnet269d,97.080,2.920,99.470,0.530,102.09,352,1.000,bicubic
 tf_efficientnet_b6_ap,97.080,2.920,99.620,0.380,43.04,528,0.942,bicubic
-beit_base_patch16_224,97.080,2.920,99.610,0.390,86.53,224,0.900,bicubic
-xcit_large_24_p8_224_dist,97.060,2.940,99.420,0.580,188.93,224,1.000,bicubic
-resnetv2_152x2_bitm,97.030,2.970,99.590,0.410,236.34,448,1.000,bilinear
-dm_nfnet_f2,97.030,2.970,99.440,0.560,193.78,352,0.920,bicubic
-resnetv2_101x3_bitm,97.020,2.980,99.490,0.510,387.93,448,1.000,bilinear
+cait_s24_384,97.070,2.930,99.430,0.570,47.06,384,1.000,bicubic
+xcit_large_24_p8_224_dist,97.070,2.930,99.420,0.580,188.93,224,1.000,bicubic
+dm_nfnet_f2,97.020,2.980,99.440,0.560,193.78,352,0.920,bicubic
+resnetv2_152x2_bitm,97.010,2.990,99.590,0.410,236.34,448,1.000,bilinear
 tf_efficientnet_b7,97.010,2.990,99.520,0.480,66.35,600,0.949,bicubic
-efficientnetv2_rw_m,96.990,3.010,99.530,0.470,53.24,416,1.000,bicubic
-deit_base_distilled_patch16_384,96.970,3.030,99.490,0.510,87.63,384,1.000,bicubic
-xcit_small_12_p16_384_dist,96.940,3.060,99.400,0.600,26.25,384,1.000,bicubic
-tf_efficientnet_b4_ns,96.930,3.070,99.580,0.420,19.34,380,0.922,bicubic
-xcit_medium_24_p8_224_dist,96.930,3.070,99.390,0.610,84.32,224,1.000,bicubic
-resnetrs420,96.910,3.090,99.460,0.540,191.89,416,1.000,bicubic
-dm_nfnet_f1,96.890,3.110,99.410,0.590,132.63,320,0.910,bicubic
-vit_base_patch16_224,96.870,3.130,99.530,0.470,86.57,224,0.900,bicubic
+resnetv2_101x3_bitm,96.990,3.010,99.490,0.510,387.93,448,1.000,bilinear
+efficientnetv2_rw_m,96.980,3.020,99.530,0.470,53.24,416,1.000,bicubic
+deit_base_distilled_patch16_384,96.960,3.040,99.480,0.520,87.63,384,1.000,bicubic
+tf_efficientnet_b4_ns,96.950,3.050,99.580,0.420,19.34,380,0.922,bicubic
+xcit_small_12_p16_384_dist,96.930,3.070,99.400,0.600,26.25,384,1.000,bicubic
+resnetrs420,96.920,3.080,99.460,0.540,191.89,416,1.000,bicubic
+xcit_medium_24_p8_224_dist,96.920,3.080,99.390,0.610,84.32,224,1.000,bicubic
+dm_nfnet_f1,96.910,3.090,99.410,0.590,132.63,320,0.910,bicubic
+vit_base_patch16_224,96.880,3.120,99.530,0.470,86.57,224,0.900,bicubic
 xcit_small_24_p8_224_dist,96.870,3.130,99.480,0.520,47.63,224,1.000,bicubic
-ig_resnext101_32x16d,96.820,3.180,99.600,0.400,194.03,224,0.875,bilinear
-resnetv2_152x2_bit_teacher_384,96.810,3.190,99.450,0.550,236.34,384,1.000,bicubic
+resnetv2_152x2_bit_teacher_384,96.830,3.170,99.450,0.550,236.34,384,1.000,bicubic
+ig_resnext101_32x16d,96.810,3.190,99.600,0.400,194.03,224,0.875,bilinear
 xcit_large_24_p16_224_dist,96.800,3.200,99.350,0.650,189.10,224,1.000,bicubic
 vit_large_r50_s32_224,96.790,3.210,99.350,0.650,328.99,224,0.900,bicubic
+convnext_large,96.770,3.230,99.310,0.690,197.77,224,0.875,bicubic
 seresnet152d,96.770,3.230,99.450,0.550,66.84,320,1.000,bicubic
-resnetrs350,96.760,3.240,99.360,0.640,163.96,384,1.000,bicubic
-resnet200d,96.740,3.260,99.340,0.660,64.69,320,1.000,bicubic
-tf_efficientnetv2_s_in21ft1k,96.740,3.260,99.420,0.580,21.46,384,1.000,bicubic
-resnetv2_50x3_bitm,96.730,3.270,99.540,0.460,217.32,448,1.000,bilinear
+resnetrs350,96.760,3.240,99.370,0.630,163.96,384,1.000,bicubic
+resnet200d,96.720,3.280,99.330,0.670,64.69,320,1.000,bicubic
+tf_efficientnetv2_s_in21ft1k,96.720,3.280,99.420,0.580,21.46,384,1.000,bicubic
+resnetv2_50x3_bitm,96.710,3.290,99.550,0.450,217.32,448,1.000,bilinear
 eca_nfnet_l1,96.700,3.300,99.290,0.710,41.41,320,1.000,bicubic
 xcit_small_12_p8_224_dist,96.700,3.300,99.390,0.610,26.21,224,1.000,bicubic
 resnetrs270,96.690,3.310,99.350,0.650,129.86,352,1.000,bicubic
 vit_small_patch16_384,96.690,3.310,99.480,0.520,22.20,384,1.000,bicubic
-tf_efficientnet_b5_ap,96.680,3.320,99.460,0.540,30.39,456,0.934,bicubic
 vit_small_r26_s32_384,96.680,3.320,99.570,0.430,36.47,384,1.000,bicubic
-pit_b_distilled_224,96.670,3.330,99.350,0.650,74.79,224,0.900,bicubic
 tf_efficientnet_b6,96.670,3.330,99.370,0.630,43.04,528,0.942,bicubic
-resnest200e,96.620,3.380,99.350,0.650,70.20,320,0.909,bicubic
+tf_efficientnet_b5_ap,96.670,3.330,99.460,0.540,30.39,456,0.934,bicubic
+pit_b_distilled_224,96.670,3.330,99.350,0.650,74.79,224,0.900,bicubic
+regnetz_d8,96.620,3.380,99.450,0.550,23.37,320,1.000,bicubic
 resmlp_big_24_224_in22ft1k,96.620,3.380,99.510,0.490,129.14,224,0.875,bicubic
-xcit_medium_24_p16_224_dist,96.600,3.400,99.270,0.730,84.40,224,1.000,bicubic
+resnest200e,96.610,3.390,99.350,0.650,70.20,320,0.909,bicubic
+regnetz_d32,96.600,3.400,99.380,0.620,27.58,320,0.950,bicubic
 swsl_resnext101_32x16d,96.600,3.400,99.530,0.470,194.03,224,0.875,bilinear
-regnetz_d,96.590,3.410,99.380,0.620,27.58,320,0.950,bicubic
-resnetrs152,96.570,3.430,99.240,0.760,86.62,320,1.000,bicubic
-cait_xs24_384,96.560,3.440,99.420,0.580,26.67,384,1.000,bicubic
-xcit_tiny_24_p8_384_dist,96.550,3.450,99.320,0.680,12.11,384,1.000,bicubic
+xcit_medium_24_p16_224_dist,96.590,3.410,99.270,0.730,84.40,224,1.000,bicubic
+resnetrs152,96.580,3.420,99.240,0.760,86.62,320,1.000,bicubic
+cait_xs24_384,96.550,3.450,99.420,0.580,26.67,384,1.000,bicubic
 efficientnetv2_rw_s,96.540,3.460,99.360,0.640,23.94,384,1.000,bicubic
+xcit_tiny_24_p8_384_dist,96.540,3.460,99.320,0.680,12.11,384,1.000,bicubic
 crossvit_18_dagger_408,96.530,3.470,99.260,0.740,44.61,408,1.000,bicubic
+resnetrs200,96.530,3.470,99.350,0.650,93.21,320,1.000,bicubic
 resnest269e,96.520,3.480,99.350,0.650,110.93,416,0.928,bicubic
-resnetrs200,96.520,3.480,99.350,0.650,93.21,320,1.000,bicubic
-vit_base_patch32_384,96.500,3.500,99.410,0.590,88.30,384,1.000,bicubic
-resmlp_big_24_distilled_224,96.470,3.530,99.310,0.690,129.14,224,0.875,bicubic
-vit_base_patch16_224_miil,96.440,3.560,99.310,0.690,86.54,224,0.875,bilinear
-swsl_resnext101_32x4d,96.420,3.580,99.470,0.530,44.18,224,0.875,bilinear
-xcit_small_24_p8_224,96.400,3.600,99.140,0.860,47.63,224,1.000,bicubic
-xcit_large_24_p8_224,96.400,3.600,98.980,1.020,188.93,224,1.000,bicubic
-cait_s24_224,96.390,3.610,99.150,0.850,46.92,224,1.000,bicubic
+vit_base_patch32_384,96.490,3.510,99.410,0.590,88.30,384,1.000,bicubic
+resmlp_big_24_distilled_224,96.450,3.550,99.310,0.690,129.14,224,0.875,bicubic
+vit_base_patch16_224_miil,96.440,3.560,99.300,0.700,86.54,224,0.875,bilinear
+convnext_base,96.440,3.560,99.230,0.770,88.59,224,0.875,bicubic
+swsl_resnext101_32x4d,96.440,3.560,99.470,0.530,44.18,224,0.875,bilinear
+xcit_large_24_p8_224,96.410,3.590,98.980,1.020,188.93,224,1.000,bicubic
+xcit_small_24_p8_224,96.400,3.600,99.150,0.850,47.63,224,1.000,bicubic
 crossvit_15_dagger_408,96.390,3.610,99.160,0.840,28.50,408,1.000,bicubic
-tf_efficientnet_b3_ns,96.370,3.630,99.350,0.650,12.23,300,0.904,bicubic
-resnet152d,96.350,3.650,99.390,0.610,60.21,320,1.000,bicubic
-regnety_160,96.340,3.660,99.330,0.670,83.59,288,1.000,bicubic
-tf_efficientnet_b5,96.340,3.660,99.310,0.690,30.39,456,0.934,bicubic
+tf_efficientnet_b3_ns,96.390,3.610,99.350,0.650,12.23,300,0.904,bicubic
+cait_s24_224,96.380,3.620,99.150,0.850,46.92,224,1.000,bicubic
+resnet152d,96.360,3.640,99.390,0.610,60.21,320,1.000,bicubic
+regnety_160,96.350,3.650,99.330,0.670,83.59,288,1.000,bicubic
+tf_efficientnet_b5,96.350,3.650,99.310,0.690,30.39,456,0.934,bicubic
 tf_efficientnetv2_s,96.340,3.660,99.200,0.800,21.46,384,1.000,bicubic
 ig_resnext101_32x8d,96.310,3.690,99.430,0.570,88.79,224,0.875,bilinear
 resnet101d,96.300,3.700,99.230,0.770,44.57,320,1.000,bicubic
-twins_svt_large,96.250,3.750,99.170,0.830,99.27,224,0.900,bicubic
+twins_svt_large,96.270,3.730,99.170,0.830,99.27,224,0.900,bicubic
 jx_nest_base,96.240,3.760,99.210,0.790,67.72,224,0.875,bicubic
 xcit_small_24_p16_224_dist,96.210,3.790,99.210,0.790,47.67,224,1.000,bicubic
 tf_efficientnet_b4_ap,96.170,3.830,99.280,0.720,19.34,380,0.922,bicubic
-efficientnet_b4,96.160,3.840,99.200,0.800,19.34,384,1.000,bicubic
+convnext_small,96.170,3.830,99.110,0.890,50.22,224,0.875,bicubic
 twins_svt_base,96.160,3.840,99.060,0.940,56.07,224,0.900,bicubic
-twins_pcpvt_large,96.150,3.850,99.190,0.810,60.99,224,0.900,bicubic
 deit_base_patch16_384,96.150,3.850,99.140,0.860,86.86,384,1.000,bicubic
-dm_nfnet_f0,96.140,3.860,99.250,0.750,71.49,256,0.900,bicubic
-nfnet_l0,96.130,3.870,99.240,0.760,35.07,288,1.000,bicubic
-resnetv2_50x1_bit_distilled,96.110,3.890,99.280,0.720,25.55,224,0.875,bicubic
+dm_nfnet_f0,96.150,3.850,99.250,0.750,71.49,256,0.900,bicubic
+efficientnet_b4,96.150,3.850,99.190,0.810,19.34,384,1.000,bicubic
+twins_pcpvt_large,96.150,3.850,99.180,0.820,60.99,224,0.900,bicubic
+nfnet_l0,96.120,3.880,99.240,0.760,35.07,288,1.000,bicubic
+resnetv2_50x1_bit_distilled,96.120,3.880,99.280,0.720,25.55,224,0.875,bicubic
 xcit_medium_24_p8_224,96.110,3.890,98.890,1.110,84.32,224,1.000,bicubic
 xcit_small_12_p8_224,96.110,3.890,99.160,0.840,26.21,224,1.000,bicubic
-deit_base_distilled_patch16_224,96.100,3.900,99.190,0.810,87.34,224,0.900,bicubic
 resnetv2_101x1_bitm,96.100,3.900,99.280,0.720,44.54,448,1.000,bilinear
-resnetv2_152x2_bit_teacher,96.090,3.910,99.270,0.730,236.34,224,0.875,bicubic
-xcit_tiny_12_p8_384_dist,96.060,3.940,99.140,0.860,6.71,384,1.000,bicubic
-xcit_small_12_p16_224_dist,96.030,3.970,99.130,0.870,26.25,224,1.000,bicubic
-regnety_032,95.980,4.020,99.190,0.810,19.44,288,1.000,bicubic
-jx_nest_small,95.970,4.030,99.040,0.960,38.35,224,0.875,bicubic
-tresnet_xl_448,95.970,4.030,99.120,0.880,78.44,448,0.875,bilinear
-eca_nfnet_l0,95.960,4.040,99.210,0.790,24.14,288,1.000,bicubic
-xcit_tiny_24_p16_384_dist,95.920,4.080,99.220,0.780,12.12,384,1.000,bicubic
-tf_efficientnet_b4,95.890,4.110,99.170,0.830,19.34,380,0.922,bicubic
-resnet51q,95.880,4.120,99.120,0.880,35.70,288,1.000,bilinear
-swsl_resnext50_32x4d,95.880,4.120,99.250,0.750,25.03,224,0.875,bilinear
-swin_small_patch4_window7_224,95.880,4.120,99.020,0.980,49.61,224,0.900,bicubic
-tresnet_l_448,95.870,4.130,99.120,0.880,55.99,448,0.875,bilinear
-cait_xxs36_384,95.860,4.140,99.090,0.910,17.37,384,1.000,bicubic
+resnetv2_152x2_bit_teacher,96.100,3.900,99.270,0.730,236.34,224,0.875,bicubic
+deit_base_distilled_patch16_224,96.090,3.910,99.190,0.810,87.34,224,0.900,bicubic
+xcit_tiny_12_p8_384_dist,96.080,3.920,99.140,0.860,6.71,384,1.000,bicubic
+xcit_small_12_p16_224_dist,96.030,3.970,99.140,0.860,26.25,224,1.000,bicubic
+tresnet_xl_448,95.970,4.030,99.130,0.870,78.44,448,0.875,bilinear
+regnety_032,95.970,4.030,99.190,0.810,19.44,288,1.000,bicubic
+jx_nest_small,95.970,4.030,99.030,0.970,38.35,224,0.875,bicubic
+eca_nfnet_l0,95.950,4.050,99.210,0.790,24.14,288,1.000,bicubic
+xcit_tiny_24_p16_384_dist,95.930,4.070,99.220,0.780,12.12,384,1.000,bicubic
+swin_small_patch4_window7_224,95.910,4.090,99.020,0.980,49.61,224,0.900,bicubic
+resnet152,95.900,4.100,99.080,0.920,60.19,224,0.950,bicubic
+tf_efficientnet_b4,95.900,4.100,99.170,0.830,19.34,380,0.922,bicubic
+resnet51q,95.870,4.130,99.120,0.880,35.70,288,1.000,bilinear
+tresnet_l_448,95.860,4.140,99.120,0.880,55.99,448,0.875,bilinear
+swsl_resnext50_32x4d,95.860,4.140,99.250,0.750,25.03,224,0.875,bilinear
 resnest101e,95.860,4.140,99.210,0.790,48.28,256,0.875,bilinear
+cait_xxs36_384,95.850,4.150,99.090,0.910,17.37,384,1.000,bicubic
 vit_large_patch32_384,95.830,4.170,99.150,0.850,306.63,384,1.000,bicubic
-xcit_tiny_24_p8_224_dist,95.820,4.180,99.210,0.790,12.11,224,1.000,bicubic
-regnetz_c,95.800,4.200,99.100,0.900,13.46,320,0.940,bicubic
-ssl_resnext101_32x16d,95.800,4.200,99.180,0.820,194.03,224,0.875,bilinear
+xcit_tiny_24_p8_224_dist,95.810,4.190,99.210,0.790,12.11,224,1.000,bicubic
+regnetz_c16,95.800,4.200,99.100,0.900,13.46,320,0.940,bicubic
+ssl_resnext101_32x16d,95.790,4.210,99.180,0.820,194.03,224,0.875,bilinear
 twins_pcpvt_base,95.790,4.210,99.130,0.870,43.83,224,0.900,bicubic
-resnet61q,95.790,4.210,98.990,1.010,36.85,288,1.000,bicubic
-tf_efficientnet_b2_ns,95.750,4.250,99.120,0.880,9.11,260,0.890,bicubic
+resnet61q,95.780,4.220,98.990,1.010,36.85,288,1.000,bicubic
+tf_efficientnet_b2_ns,95.770,4.230,99.120,0.880,9.11,260,0.890,bicubic
 gc_efficientnetv2_rw_t,95.740,4.260,99.020,0.980,13.68,288,1.000,bicubic
-tresnet_m,95.730,4.270,99.030,0.970,31.39,224,0.875,bilinear
-pnasnet5large,95.720,4.280,98.920,1.080,86.06,331,0.911,bicubic
-efficientnet_b3,95.720,4.280,99.040,0.960,12.23,320,1.000,bicubic
+tresnet_m,95.720,4.280,99.030,0.970,31.39,224,0.875,bilinear
+efficientnet_b3,95.710,4.290,99.040,0.960,12.23,320,1.000,bicubic
+pnasnet5large,95.710,4.290,98.920,1.080,86.06,331,0.911,bicubic
+crossvit_15_dagger_240,95.690,4.310,98.830,1.170,28.21,240,0.875,bicubic
 nasnetalarge,95.680,4.320,98.930,1.070,88.75,331,0.911,bicubic
 xcit_tiny_24_p8_224,95.670,4.330,99.050,0.950,12.11,224,1.000,bicubic
-crossvit_15_dagger_240,95.670,4.330,98.820,1.180,28.21,240,0.875,bicubic
 vit_small_r26_s32_224,95.640,4.360,99.190,0.810,36.43,224,0.900,bicubic
-pit_b_224,95.630,4.370,98.670,1.330,73.76,224,0.900,bicubic
+pit_b_224,95.640,4.360,98.670,1.330,73.76,224,0.900,bicubic
 resnetv2_101,95.630,4.370,98.990,1.010,44.54,224,0.950,bicubic
 efficientnetv2_rw_t,95.610,4.390,99.070,0.930,13.65,288,1.000,bicubic
-crossvit_18_dagger_240,95.560,4.440,99.060,0.940,44.27,240,0.875,bicubic
-convit_base,95.550,4.450,98.890,1.110,86.54,224,0.875,bicubic
-ecaresnet101d,95.540,4.460,99.130,0.870,44.57,224,0.875,bicubic
-coat_lite_small,95.530,4.470,98.860,1.140,19.84,224,0.900,bicubic
+crossvit_18_dagger_240,95.570,4.430,99.060,0.940,44.27,240,0.875,bicubic
+convit_base,95.550,4.450,98.870,1.130,86.54,224,0.875,bicubic
+convnext_tiny,95.550,4.450,99.020,0.980,28.59,224,0.875,bicubic
+coat_lite_small,95.540,4.460,98.860,1.140,19.84,224,0.900,bicubic
+xcit_medium_24_p16_224,95.530,4.470,98.740,1.260,84.40,224,1.000,bicubic
+xcit_small_24_p16_224,95.530,4.470,98.760,1.240,47.67,224,1.000,bicubic
+ecaresnet101d,95.530,4.470,99.130,0.870,44.57,224,0.875,bicubic
 levit_384,95.530,4.470,99.050,0.950,39.13,224,0.900,bicubic
-xcit_small_24_p16_224,95.530,4.470,98.770,1.230,47.67,224,1.000,bicubic
-xcit_medium_24_p16_224,95.520,4.480,98.770,1.230,84.40,224,1.000,bicubic
 crossvit_base_240,95.520,4.480,98.820,1.180,105.03,240,0.875,bicubic
-visformer_small,95.500,4.500,98.900,1.100,40.22,224,0.900,bicubic
-ecaresnet50t,95.500,4.500,99.120,0.880,25.57,320,0.950,bicubic
-ssl_resnext101_32x8d,95.470,4.530,99.120,0.880,88.79,224,0.875,bilinear
-crossvit_18_240,95.450,4.550,98.790,1.210,43.27,240,0.875,bicubic
+ecaresnet50t,95.510,4.490,99.120,0.880,25.57,320,0.950,bicubic
+fbnetv3_g,95.510,4.490,98.990,1.010,16.62,288,0.950,bilinear
+ssl_resnext101_32x8d,95.490,4.510,99.120,0.880,88.79,224,0.875,bilinear
+visformer_small,95.480,4.520,98.900,1.100,40.22,224,0.900,bicubic
+crossvit_18_240,95.440,4.560,98.790,1.210,43.27,240,0.875,bicubic
 deit_base_patch16_224,95.440,4.560,98.840,1.160,86.57,224,0.900,bicubic
-tresnet_xl,95.440,4.560,99.060,0.940,78.44,224,0.875,bilinear
-ssl_resnext101_32x4d,95.430,4.570,99.130,0.870,44.18,224,0.875,bilinear
-resnetrs101,95.420,4.580,99.030,0.970,63.62,288,0.940,bicubic
-xcit_large_24_p16_224,95.410,4.590,98.630,1.370,189.10,224,1.000,bicubic
+ssl_resnext101_32x4d,95.440,4.560,99.130,0.870,44.18,224,0.875,bilinear
+tresnet_xl,95.440,4.560,99.050,0.950,78.44,224,0.875,bilinear
+resnetrs101,95.430,4.570,99.030,0.970,63.62,288,0.940,bicubic
+halo2botnet50ts_256,95.420,4.580,99.010,0.990,22.64,256,0.950,bicubic
+xcit_large_24_p16_224,95.420,4.580,98.620,1.380,189.10,224,1.000,bicubic
+xcit_small_12_p16_224,95.420,4.580,98.840,1.160,26.25,224,1.000,bicubic
 swsl_resnet50,95.400,4.600,99.290,0.710,25.56,224,0.875,bilinear
-xcit_small_12_p16_224,95.400,4.600,98.830,1.170,26.25,224,1.000,bicubic
 vit_small_patch16_224,95.370,4.630,99.150,0.850,22.05,224,0.900,bicubic
+resnet101,95.360,4.640,98.860,1.140,44.55,224,0.950,bicubic
 tf_efficientnet_b3_ap,95.320,4.680,98.900,1.100,12.23,300,0.904,bicubic
-mixer_b16_224_miil,95.310,4.690,98.890,1.110,59.88,224,0.875,bilinear
+mixer_b16_224_miil,95.300,4.700,98.880,1.120,59.88,224,0.875,bilinear
 tresnet_l,95.290,4.710,99.010,0.990,55.99,224,0.875,bilinear
-cait_xxs24_384,95.270,4.730,98.960,1.040,12.03,384,1.000,bicubic
+cait_xxs24_384,95.280,4.720,98.960,1.040,12.03,384,1.000,bicubic
 jx_nest_tiny,95.250,4.750,98.980,1.020,17.06,224,0.875,bicubic
 pit_s_distilled_224,95.240,4.760,99.050,0.950,24.04,224,0.900,bicubic
-twins_pcpvt_small,95.230,4.770,98.880,1.120,24.11,224,0.900,bicubic
-twins_svt_small,95.210,4.790,98.890,1.110,24.06,224,0.900,bicubic
-convit_small,95.180,4.820,98.920,1.080,27.78,224,0.875,bicubic
+twins_pcpvt_small,95.210,4.790,98.880,1.120,24.11,224,0.900,bicubic
+convit_small,95.200,4.800,98.900,1.100,27.78,224,0.875,bicubic
+twins_svt_small,95.200,4.800,98.880,1.120,24.06,224,0.900,bicubic
 tf_efficientnet_b1_ns,95.170,4.830,99.120,0.880,7.79,240,0.882,bicubic
-tf_efficientnetv2_b3,95.170,4.830,98.820,1.180,14.36,300,0.904,bicubic
+halonet50ts,95.160,4.840,98.770,1.230,22.73,256,0.940,bicubic
+tf_efficientnetv2_b3,95.160,4.840,98.820,1.180,14.36,300,0.904,bicubic
+lamhalobotnet50ts_256,95.150,4.850,98.880,1.120,22.57,256,0.950,bicubic
+crossvit_15_240,95.140,4.860,98.930,1.070,27.53,240,0.875,bicubic
+xcit_tiny_12_p16_384_dist,95.130,4.870,99.020,0.980,6.72,384,1.000,bicubic
 swin_tiny_patch4_window7_224,95.130,4.870,98.850,1.150,28.29,224,0.900,bicubic
-crossvit_15_240,95.120,4.880,98.930,1.070,27.53,240,0.875,bicubic
 efficientnet_el,95.120,4.880,98.980,1.020,10.59,300,0.904,bicubic
-gernet_l,95.110,4.890,98.900,1.100,31.08,256,0.875,bilinear
-xcit_tiny_12_p16_384_dist,95.110,4.890,99.020,0.980,6.72,384,1.000,bicubic
-convmixer_1536_20,95.080,4.920,99.030,0.970,51.63,224,0.960,bicubic
-xcit_tiny_12_p8_224_dist,95.080,4.920,98.910,1.090,6.71,224,1.000,bicubic
-ecaresnet101d_pruned,95.070,4.930,98.980,1.020,24.88,224,0.875,bicubic
-vit_small_patch32_384,95.060,4.940,98.990,1.010,22.92,384,1.000,bicubic
-legacy_senet154,95.060,4.940,98.830,1.170,115.09,224,0.875,bilinear
-regnetz_b,95.060,4.940,99.050,0.950,9.72,288,0.940,bicubic
-gluon_resnet152_v1s,95.050,4.950,98.930,1.070,60.32,224,0.875,bicubic
-halonet50ts,95.050,4.950,98.590,1.410,22.73,256,0.940,bicubic
-wide_resnet50_2,95.050,4.950,98.970,1.030,68.88,224,0.875,bicubic
+gernet_l,95.100,4.900,98.900,1.100,31.08,256,0.875,bilinear
+xcit_tiny_12_p8_224_dist,95.090,4.910,98.910,1.090,6.71,224,1.000,bicubic
+ecaresnet101d_pruned,95.080,4.920,98.980,1.020,24.88,224,0.875,bicubic
+legacy_senet154,95.070,4.930,98.830,1.170,115.09,224,0.875,bilinear
+regnetz_b16,95.070,4.930,99.050,0.950,9.72,288,0.940,bicubic
+wide_resnet50_2,95.070,4.930,98.970,1.030,68.88,224,0.875,bicubic
+convmixer_1536_20,95.060,4.940,99.030,0.970,51.63,224,0.960,bicubic
+vit_small_patch32_384,95.050,4.950,98.990,1.010,22.92,384,1.000,bicubic
+gluon_resnet152_v1s,95.040,4.960,98.930,1.070,60.32,224,0.875,bicubic
 tnt_s_patch16_224,95.040,4.960,98.840,1.160,23.76,224,0.900,bicubic
-seresnext50_32x4d,95.030,4.970,98.890,1.110,27.56,224,0.875,bicubic
-levit_256,95.030,4.970,98.890,1.110,18.89,224,0.900,bicubic
+seresnext50_32x4d,95.030,4.970,98.880,1.120,27.56,224,0.875,bicubic
+levit_256,95.020,4.980,98.890,1.110,18.89,224,0.900,bicubic
 tf_efficientnet_b3,95.020,4.980,98.910,1.090,12.23,300,0.904,bicubic
-resnetv2_50x1_bitm,95.020,4.980,99.050,0.950,25.55,448,1.000,bilinear
-vit_base_patch32_224,95.010,4.990,99.020,0.980,88.22,224,0.900,bicubic
-coat_mini,94.990,5.010,98.780,1.220,10.34,224,0.900,bicubic
+resnetv2_50x1_bitm,95.010,4.990,99.060,0.940,25.55,448,1.000,bilinear
+vit_base_patch32_224,95.000,5.000,99.030,0.970,88.22,224,0.900,bicubic
 tresnet_m_448,94.990,5.010,98.980,1.020,31.39,448,0.875,bilinear
-resnest50d_4s2x40d,94.950,5.050,99.070,0.930,30.42,224,0.875,bicubic
-rexnet_200,94.940,5.060,99.000,1.000,16.37,224,0.875,bicubic
-gluon_seresnext101_64x4d,94.940,5.060,98.820,1.180,88.23,224,0.875,bicubic
-gluon_senet154,94.930,5.070,98.770,1.230,115.09,224,0.875,bicubic
+coat_mini,94.970,5.030,98.780,1.220,10.34,224,0.900,bicubic
+resnest50d_4s2x40d,94.960,5.040,99.070,0.930,30.42,224,0.875,bicubic
+rexnet_200,94.950,5.050,99.010,0.990,16.37,224,0.875,bicubic
+gluon_senet154,94.920,5.080,98.760,1.240,115.09,224,0.875,bicubic
 gluon_seresnext101_32x4d,94.920,5.080,98.810,1.190,48.96,224,0.875,bicubic
-seresnet33ts,94.870,5.130,98.790,1.210,19.78,256,0.900,bicubic
-resmlp_36_distilled_224,94.870,5.130,98.860,1.140,44.69,224,0.875,bicubic
+gluon_seresnext101_64x4d,94.920,5.080,98.830,1.170,88.23,224,0.875,bicubic
+tf_efficientnet_lite4,94.890,5.110,99.020,0.980,13.01,380,0.920,bilinear
+resmlp_36_distilled_224,94.880,5.120,98.840,1.160,44.69,224,0.875,bicubic
+ssl_resnext50_32x4d,94.870,5.130,98.890,1.110,25.03,224,0.875,bilinear
 gcresnet50t,94.860,5.140,98.800,1.200,25.90,256,0.900,bicubic
-ssl_resnext50_32x4d,94.860,5.140,98.870,1.130,25.03,224,0.875,bilinear
-tf_efficientnet_lite4,94.860,5.140,99.020,0.980,13.01,380,0.920,bilinear
+resnest50d,94.850,5.150,98.880,1.120,27.48,224,0.875,bilinear
+seresnet33ts,94.850,5.150,98.790,1.210,19.78,256,0.900,bicubic
 crossvit_small_240,94.830,5.170,99.020,0.980,26.86,240,0.875,bicubic
-resnest50d,94.830,5.170,98.880,1.120,27.48,224,0.875,bilinear
-lamhalobotnet50ts_256,94.800,5.200,98.550,1.450,22.57,256,0.950,bicubic
-sehalonet33ts,94.780,5.220,98.570,1.430,13.69,256,0.940,bicubic
-resnest50d_1s4x24d,94.770,5.230,98.980,1.020,25.68,224,0.875,bicubic
+lambda_resnet50ts,94.790,5.210,98.460,1.540,21.54,256,0.950,bicubic
 ecaresnetlight,94.770,5.230,98.800,1.200,30.16,224,0.875,bicubic
-lambda_resnet50ts,94.770,5.230,98.470,1.530,21.54,256,0.950,bicubic
-halo2botnet50ts_256,94.760,5.240,98.660,1.340,22.64,256,0.950,bicubic
-gluon_resnet152_v1d,94.750,5.250,98.740,1.260,60.21,224,0.875,bicubic
-xcit_tiny_12_p8_224,94.710,5.290,98.830,1.170,6.71,224,1.000,bicubic
+sehalonet33ts,94.770,5.230,98.570,1.430,13.69,256,0.940,bicubic
+resnest50d_1s4x24d,94.750,5.250,98.980,1.020,25.68,224,0.875,bicubic
+gluon_resnet152_v1d,94.740,5.260,98.740,1.260,60.21,224,0.875,bicubic
+gluon_resnet101_v1s,94.720,5.280,98.820,1.180,44.67,224,0.875,bicubic
 deit_small_distilled_patch16_224,94.710,5.290,99.030,0.970,22.44,224,0.900,bicubic
-haloregnetz_b,94.710,5.290,98.660,1.340,11.68,224,0.940,bicubic
-gluon_resnet101_v1s,94.700,5.300,98.820,1.180,44.67,224,0.875,bicubic
-cspdarknet53,94.670,5.330,98.810,1.190,27.64,256,0.887,bilinear
-resmlp_big_24_224,94.650,5.350,98.490,1.510,129.14,224,0.875,bicubic
-gluon_resnext101_64x4d,94.640,5.360,98.670,1.330,83.46,224,0.875,bicubic
-efficientnet_b2,94.630,5.370,98.710,1.290,9.11,288,1.000,bicubic
-ecaresnet50d,94.620,5.380,98.890,1.110,25.58,224,0.875,bicubic
-efficientnet_b3_pruned,94.620,5.380,98.770,1.230,9.86,300,0.904,bicubic
+haloregnetz_b,94.700,5.300,98.660,1.340,11.68,224,0.940,bicubic
+xcit_tiny_12_p8_224,94.690,5.310,98.830,1.170,6.71,224,1.000,bicubic
+resmlp_big_24_224,94.670,5.330,98.480,1.520,129.14,224,0.875,bicubic
+cspdarknet53,94.660,5.340,98.800,1.200,27.64,256,0.887,bilinear
+gluon_resnext101_64x4d,94.660,5.340,98.650,1.350,83.46,224,0.875,bicubic
+efficientnet_b3_pruned,94.630,5.370,98.760,1.240,9.86,300,0.904,bicubic
+ecaresnet50d,94.630,5.370,98.890,1.110,25.58,224,0.875,bicubic
 gernet_m,94.620,5.380,98.860,1.140,21.14,224,0.875,bilinear
-pit_s_224,94.580,5.420,98.720,1.280,23.46,224,0.900,bicubic
-repvgg_b3,94.570,5.430,98.790,1.210,123.09,224,0.875,bilinear
-nf_resnet50,94.560,5.440,98.790,1.210,25.56,288,0.940,bicubic
-seresnet50,94.560,5.440,98.750,1.250,28.09,224,0.875,bicubic
-regnety_320,94.550,5.450,98.850,1.150,145.05,224,0.875,bicubic
-inception_resnet_v2,94.540,5.460,98.790,1.210,55.84,299,0.897,bicubic
+efficientnet_b2,94.610,5.390,98.710,1.290,9.11,288,1.000,bicubic
+pit_s_224,94.590,5.410,98.700,1.300,23.46,224,0.900,bicubic
+repvgg_b3,94.570,5.430,98.780,1.220,123.09,224,0.875,bilinear
+sebotnet33ts_256,94.570,5.430,98.500,1.500,13.70,256,0.940,bicubic
+nf_resnet50,94.550,5.450,98.790,1.210,25.56,288,0.940,bicubic
+seresnet50,94.550,5.450,98.750,1.250,28.09,224,0.875,bicubic
+regnety_320,94.540,5.460,98.850,1.150,145.05,224,0.875,bicubic
+resnext50_32x4d,94.540,5.460,98.610,1.390,25.03,224,0.950,bicubic
 gluon_resnext101_32x4d,94.540,5.460,98.630,1.370,44.18,224,0.875,bicubic
-repvgg_b3g4,94.530,5.470,98.960,1.040,83.83,224,0.875,bilinear
-xcit_tiny_24_p16_224_dist,94.520,5.480,98.790,1.210,12.12,224,1.000,bicubic
-convmixer_768_32,94.490,5.510,98.850,1.150,21.11,224,0.960,bicubic
+inception_resnet_v2,94.540,5.460,98.790,1.210,55.84,299,0.897,bicubic
+xcit_tiny_24_p16_224_dist,94.530,5.470,98.780,1.220,12.12,224,1.000,bicubic
+repvgg_b3g4,94.520,5.480,98.970,1.030,83.83,224,0.875,bilinear
+convmixer_768_32,94.500,5.500,98.850,1.150,21.11,224,0.960,bicubic
+gcresnext50ts,94.490,5.510,98.670,1.330,15.67,256,0.900,bicubic
 tf_efficientnet_b2_ap,94.490,5.510,98.620,1.380,9.11,260,0.890,bicubic
 regnety_120,94.480,5.520,98.810,1.190,51.82,224,0.875,bicubic
-gcresnext50ts,94.480,5.520,98.670,1.330,15.67,256,0.900,bicubic
-gcresnet33ts,94.480,5.520,98.780,1.220,19.88,256,0.900,bicubic
+rexnet_150,94.480,5.520,98.790,1.210,9.73,224,0.875,bicubic
 cspresnext50,94.470,5.530,98.680,1.320,20.57,224,0.875,bilinear
-ssl_resnet50,94.470,5.530,98.920,1.080,25.56,224,0.875,bilinear
-rexnet_150,94.470,5.530,98.790,1.210,9.73,224,0.875,bicubic
-resmlp_24_distilled_224,94.450,5.550,98.770,1.230,30.02,224,0.875,bicubic
-resnetv2_50,94.440,5.560,98.730,1.270,25.55,224,0.950,bicubic
-regnetx_320,94.440,5.560,98.730,1.270,107.81,224,0.875,bicubic
-tf_efficientnetv2_b2,94.410,5.590,98.570,1.430,10.10,260,0.890,bicubic
-deit_small_patch16_224,94.400,5.600,98.690,1.310,22.05,224,0.900,bicubic
+gcresnet33ts,94.470,5.530,98.770,1.230,19.88,256,0.900,bicubic
+regnetx_320,94.460,5.540,98.740,1.260,107.81,224,0.875,bicubic
+resmlp_24_distilled_224,94.460,5.540,98.770,1.230,30.02,224,0.875,bicubic
+ssl_resnet50,94.450,5.550,98.920,1.080,25.56,224,0.875,bilinear
+resnetv2_50,94.440,5.560,98.740,1.260,25.55,224,0.950,bicubic
+tf_efficientnetv2_b2,94.420,5.580,98.570,1.430,10.10,260,0.890,bicubic
+efficientnet_el_pruned,94.400,5.600,98.740,1.260,10.59,300,0.904,bicubic
 tf_efficientnet_el,94.400,5.600,98.710,1.290,10.59,300,0.904,bicubic
-efficientnet_el_pruned,94.390,5.610,98.750,1.250,10.59,300,0.904,bicubic
-inception_v4,94.370,5.630,98.580,1.420,42.68,299,0.875,bicubic
-tf_efficientnet_b2,94.370,5.630,98.610,1.390,9.11,260,0.890,bicubic
-legacy_seresnext101_32x4d,94.350,5.650,98.630,1.370,48.96,224,0.875,bilinear
+deit_small_patch16_224,94.390,5.610,98.690,1.310,22.05,224,0.900,bicubic
+inception_v4,94.380,5.620,98.580,1.420,42.68,299,0.875,bicubic
+legacy_seresnext101_32x4d,94.360,5.640,98.650,1.350,48.96,224,0.875,bilinear
+resnet50_gn,94.360,5.640,98.710,1.290,25.56,224,0.940,bicubic
+tf_efficientnet_b2,94.360,5.640,98.610,1.390,9.11,260,0.890,bicubic
+gluon_seresnext50_32x4d,94.330,5.670,98.610,1.390,27.56,224,0.875,bicubic
 resnet50,94.330,5.670,98.440,1.560,25.56,224,0.950,bicubic
-gluon_seresnext50_32x4d,94.330,5.670,98.620,1.380,27.56,224,0.875,bicubic
-ecaresnet26t,94.300,5.700,98.710,1.290,16.01,320,0.950,bicubic
+ecaresnet26t,94.310,5.690,98.720,1.280,16.01,320,0.950,bicubic
 dpn107,94.300,5.700,98.470,1.530,86.92,224,0.875,bicubic
-xception71,94.290,5.710,98.640,1.360,42.34,299,0.903,bicubic
-resnetrs50,94.290,5.710,98.640,1.360,35.69,224,0.910,bicubic
-resnet50d,94.270,5.730,98.720,1.280,25.58,224,0.875,bicubic
+resnetrs50,94.300,5.700,98.640,1.360,35.69,224,0.910,bicubic
+xception71,94.280,5.720,98.640,1.360,42.34,299,0.903,bicubic
+cait_xxs36_224,94.260,5.740,98.720,1.280,17.30,224,1.000,bicubic
 gluon_xception65,94.260,5.740,98.570,1.430,39.92,299,0.903,bicubic
-cait_xxs36_224,94.260,5.740,98.710,1.290,17.30,224,1.000,bicubic
-skresnext50_32x4d,94.260,5.740,98.470,1.530,27.48,224,0.875,bicubic
-regnetx_120,94.240,5.760,98.670,1.330,46.11,224,0.875,bicubic
-dpn92,94.220,5.780,98.730,1.270,37.67,224,0.875,bicubic
-ecaresnet50d_pruned,94.210,5.790,98.730,1.270,19.94,224,0.875,bicubic
-mixnet_xl,94.200,5.800,98.340,1.660,11.90,224,0.875,bicubic
-eca_resnet33ts,94.200,5.800,98.770,1.230,19.68,256,0.900,bicubic
-gluon_resnet101_v1d,94.200,5.800,98.570,1.430,44.57,224,0.875,bicubic
-resmlp_36_224,94.190,5.810,98.660,1.340,44.69,224,0.875,bicubic
-resnext50d_32x4d,94.190,5.810,98.570,1.430,25.05,224,0.875,bicubic
-tf_efficientnet_lite3,94.190,5.810,98.640,1.360,8.20,300,0.904,bilinear
-levit_192,94.180,5.820,98.560,1.440,10.95,224,0.900,bicubic
-regnety_080,94.180,5.820,98.680,1.320,39.18,224,0.875,bicubic
-ens_adv_inception_resnet_v2,94.160,5.840,98.610,1.390,55.84,299,0.897,bicubic
+resnet50d,94.260,5.740,98.720,1.280,25.58,224,0.875,bicubic
+skresnext50_32x4d,94.260,5.740,98.460,1.540,27.48,224,0.875,bicubic
+regnetx_120,94.240,5.760,98.650,1.350,46.11,224,0.875,bicubic
+gluon_resnet101_v1d,94.240,5.760,98.560,1.440,44.57,224,0.875,bicubic
+dpn92,94.230,5.770,98.730,1.270,37.67,224,0.875,bicubic
+ecaresnet50d_pruned,94.220,5.780,98.730,1.270,19.94,224,0.875,bicubic
+resmlp_36_224,94.200,5.800,98.660,1.340,44.69,224,0.875,bicubic
+tf_efficientnet_lite3,94.200,5.800,98.640,1.360,8.20,300,0.904,bilinear
+eca_resnet33ts,94.190,5.810,98.770,1.230,19.68,256,0.900,bicubic
+mixnet_xl,94.190,5.810,98.340,1.660,11.90,224,0.875,bicubic
+resnext50d_32x4d,94.180,5.820,98.570,1.430,25.05,224,0.875,bicubic
+regnety_080,94.170,5.830,98.680,1.320,39.18,224,0.875,bicubic
+levit_192,94.170,5.830,98.550,1.450,10.95,224,0.900,bicubic
+ens_adv_inception_resnet_v2,94.160,5.840,98.600,1.400,55.84,299,0.897,bicubic
 gluon_resnet152_v1c,94.160,5.840,98.640,1.360,60.21,224,0.875,bicubic
 gmlp_s16_224,94.150,5.850,98.500,1.500,19.42,224,0.875,bicubic
-regnety_064,94.150,5.850,98.740,1.260,30.58,224,0.875,bicubic
-vit_base_patch16_sam_224,94.150,5.850,98.670,1.330,86.57,224,0.900,bicubic
-efficientnet_b2_pruned,94.140,5.860,98.520,1.480,8.31,260,0.890,bicubic
-dpn98,94.120,5.880,98.580,1.420,61.57,224,0.875,bicubic
+efficientnet_b2_pruned,94.140,5.860,98.530,1.470,8.31,260,0.890,bicubic
+regnety_064,94.140,5.860,98.730,1.270,30.58,224,0.875,bicubic
+vit_base_patch16_224_sam,94.140,5.860,98.670,1.330,86.57,224,0.900,bicubic
 regnetx_160,94.120,5.880,98.740,1.260,54.28,224,0.875,bicubic
-nf_regnet_b1,94.120,5.880,98.620,1.380,10.22,288,0.900,bicubic
-resnext50_32x4d,94.110,5.890,98.350,1.650,25.03,224,0.875,bicubic
+nf_regnet_b1,94.120,5.880,98.630,1.370,10.22,288,0.900,bicubic
+dpn98,94.110,5.890,98.580,1.420,61.57,224,0.875,bicubic
 ese_vovnet39b,94.090,5.910,98.660,1.340,24.57,224,0.875,bicubic
 gluon_resnet152_v1b,94.080,5.920,98.460,1.540,60.19,224,0.875,bicubic
-xcit_tiny_24_p16_224,94.080,5.920,98.510,1.490,12.12,224,1.000,bicubic
-coat_lite_mini,94.050,5.950,98.540,1.460,11.01,224,0.900,bicubic
+xcit_tiny_24_p16_224,94.070,5.930,98.510,1.490,12.12,224,1.000,bicubic
+coat_lite_mini,94.060,5.940,98.550,1.450,11.01,224,0.900,bicubic
+eca_halonext26ts,94.050,5.950,98.500,1.500,10.76,256,0.940,bicubic
+halonet26t,94.020,5.980,98.500,1.500,12.48,256,0.950,bicubic
 resmlp_24_224,94.020,5.980,98.330,1.670,30.02,224,0.875,bicubic
+hrnet_w64,94.010,5.990,98.620,1.380,128.06,224,0.875,bilinear
 dpn131,93.990,6.010,98.720,1.280,79.25,224,0.875,bicubic
-hrnet_w64,93.990,6.010,98.620,1.380,128.06,224,0.875,bilinear
-halonet26t,93.980,6.020,98.490,1.510,12.48,256,0.950,bicubic
 dla102x2,93.960,6.040,98.480,1.520,41.28,224,0.875,bilinear
-hrnet_w48,93.940,6.060,98.610,1.390,77.47,224,0.875,bilinear
+fbnetv3_b,93.960,6.040,98.630,1.370,8.60,256,0.950,bilinear
+resnetblur50,93.950,6.050,98.580,1.420,25.56,224,0.875,bicubic
 tf_efficientnetv2_b1,93.940,6.060,98.620,1.380,8.14,240,0.882,bicubic
-resnetblur50,93.930,6.070,98.580,1.420,25.56,224,0.875,bicubic
-tf_efficientnet_cc_b1_8e,93.920,6.080,98.250,1.750,39.72,240,0.882,bicubic
+fbnetv3_d,93.930,6.070,98.740,1.260,10.31,256,0.950,bilinear
+hrnet_w48,93.920,6.080,98.610,1.390,77.47,224,0.875,bilinear
+tf_efficientnet_cc_b1_8e,93.910,6.090,98.260,1.740,39.72,240,0.882,bicubic
 rexnet_130,93.900,6.100,98.400,1.600,7.56,224,0.875,bicubic
 regnetx_064,93.890,6.110,98.630,1.370,26.21,224,0.875,bicubic
 regnetx_080,93.870,6.130,98.520,1.480,39.57,224,0.875,bicubic
-regnety_040,93.860,6.140,98.640,1.360,20.65,224,0.875,bicubic
-repvgg_b2g4,93.840,6.160,98.600,1.400,61.76,224,0.875,bilinear
-efficientnet_em,93.840,6.160,98.810,1.190,6.90,240,0.882,bicubic
-gluon_resnext50_32x4d,93.820,6.180,98.410,1.590,25.03,224,0.875,bicubic
-lambda_resnet26t,93.820,6.180,98.650,1.350,10.96,256,0.940,bicubic
+regnety_040,93.860,6.140,98.650,1.350,20.65,224,0.875,bicubic
+repvgg_b2g4,93.840,6.160,98.590,1.410,61.76,224,0.875,bilinear
+efficientnet_em,93.830,6.170,98.810,1.190,6.90,240,0.882,bicubic
+lambda_resnet26t,93.830,6.170,98.650,1.350,10.96,256,0.940,bicubic
 pit_xs_distilled_224,93.820,6.180,98.670,1.330,11.00,224,0.900,bicubic
-eca_botnext26ts_256,93.790,6.210,98.500,1.500,10.59,256,0.950,bicubic
-resnext101_32x8d,93.790,6.210,98.580,1.420,88.79,224,0.875,bilinear
-gluon_resnet50_v1d,93.780,6.220,98.400,1.600,25.58,224,0.875,bicubic
-xception65,93.770,6.230,98.360,1.640,39.92,299,0.903,bicubic
-cspresnet50,93.750,6.250,98.630,1.370,21.62,256,0.887,bilinear
-gluon_resnet101_v1b,93.730,6.270,98.400,1.600,44.55,224,0.875,bicubic
+resnext101_32x8d,93.820,6.180,98.580,1.420,88.79,224,0.875,bilinear
+gluon_resnext50_32x4d,93.810,6.190,98.410,1.590,25.03,224,0.875,bicubic
+eca_botnext26ts_256,93.780,6.220,98.500,1.500,10.59,256,0.950,bicubic
+gluon_resnet50_v1d,93.770,6.230,98.390,1.610,25.58,224,0.875,bicubic
+xception65,93.760,6.240,98.370,1.630,39.92,299,0.903,bicubic
+gluon_resnet101_v1b,93.750,6.250,98.380,1.620,44.55,224,0.875,bicubic
+res2net101_26w_4s,93.750,6.250,98.310,1.690,45.21,224,0.875,bilinear
+cspresnet50,93.740,6.260,98.640,1.360,21.62,256,0.887,bilinear
+legacy_seresnext50_32x4d,93.730,6.270,98.580,1.420,27.56,224,0.875,bilinear
 wide_resnet101_2,93.720,6.280,98.540,1.460,126.89,224,0.875,bilinear
-res2net101_26w_4s,93.720,6.280,98.320,1.680,45.21,224,0.875,bilinear
-legacy_seresnext50_32x4d,93.720,6.280,98.580,1.420,27.56,224,0.875,bilinear
-lambda_resnet26rpt_256,93.720,6.280,98.500,1.500,10.99,256,0.940,bicubic
-tf_efficientnet_b1_ap,93.710,6.290,98.360,1.640,7.79,240,0.882,bicubic
-dpn68b,93.680,6.320,98.530,1.470,12.61,224,0.875,bicubic
-gluon_resnet101_v1c,93.660,6.340,98.410,1.590,44.57,224,0.875,bicubic
+lambda_resnet26rpt_256,93.710,6.290,98.510,1.490,10.99,256,0.940,bicubic
+dpn68b,93.690,6.310,98.520,1.480,12.61,224,0.875,bicubic
+tf_efficientnet_b1_ap,93.690,6.310,98.360,1.640,7.79,240,0.882,bicubic
+gluon_resnet101_v1c,93.660,6.340,98.420,1.580,44.57,224,0.875,bicubic
 vit_tiny_patch16_384,93.650,6.350,98.600,1.400,5.79,384,1.000,bicubic
-gluon_resnet50_v1s,93.630,6.370,98.470,1.530,25.68,224,0.875,bicubic
-tf_efficientnet_b0_ns,93.620,6.380,98.640,1.360,5.29,224,0.875,bicubic
-resnet33ts,93.600,6.400,98.530,1.470,19.68,256,0.900,bicubic
-cait_xxs24_224,93.590,6.410,98.440,1.560,11.96,224,1.000,bicubic
-hrnet_w44,93.580,6.420,98.700,1.300,67.06,224,0.875,bilinear
-coat_tiny,93.580,6.420,98.410,1.590,5.50,224,0.900,bicubic
-regnetx_040,93.550,6.450,98.560,1.440,22.12,224,0.875,bicubic
-hrnet_w32,93.520,6.480,98.440,1.560,41.23,224,0.875,bilinear
-eca_halonext26ts,93.510,6.490,98.280,1.720,10.76,256,0.940,bicubic
-tf_efficientnet_b1,93.510,6.490,98.360,1.640,7.79,240,0.882,bicubic
-dla102x,93.510,6.490,98.500,1.500,26.31,224,0.875,bilinear
+tf_efficientnet_b0_ns,93.630,6.370,98.640,1.360,5.29,224,0.875,bicubic
+gluon_resnet50_v1s,93.620,6.380,98.460,1.540,25.68,224,0.875,bicubic
+cait_xxs24_224,93.600,6.400,98.440,1.560,11.96,224,1.000,bicubic
+resnet33ts,93.600,6.400,98.540,1.460,19.68,256,0.900,bicubic
+coat_tiny,93.590,6.410,98.420,1.580,5.50,224,0.900,bicubic
+hrnet_w44,93.550,6.450,98.700,1.300,67.06,224,0.875,bilinear
+regnetx_040,93.550,6.450,98.550,1.450,22.12,224,0.875,bicubic
+hrnet_w32,93.530,6.470,98.460,1.540,41.23,224,0.875,bilinear
+dla102x,93.520,6.480,98.510,1.490,26.31,224,0.875,bilinear
+xcit_nano_12_p8_384_dist,93.520,6.480,98.540,1.460,3.05,384,1.000,bicubic
+tf_efficientnet_b1,93.500,6.500,98.360,1.640,7.79,240,0.882,bicubic
 botnet26t_256,93.500,6.500,98.300,1.700,12.49,256,0.950,bicubic
 repvgg_b2,93.490,6.510,98.730,1.270,89.02,224,0.875,bilinear
-hrnet_w40,93.490,6.510,98.590,1.410,57.56,224,0.875,bilinear
-xcit_nano_12_p8_384_dist,93.480,6.520,98.520,1.480,3.05,384,1.000,bicubic
-xception,93.480,6.520,98.530,1.470,22.86,299,0.897,bicubic
+hrnet_w40,93.490,6.510,98.580,1.420,57.56,224,0.875,bilinear
 resnet32ts,93.470,6.530,98.490,1.510,17.96,256,0.900,bicubic
-gluon_inception_v3,93.460,6.540,98.560,1.440,23.83,299,0.875,bicubic
-res2net50_26w_8s,93.430,6.570,98.180,1.820,48.40,224,0.875,bilinear
-mixnet_l,93.430,6.570,98.220,1.780,7.33,224,0.875,bicubic
-legacy_seresnet152,93.420,6.580,98.340,1.660,66.82,224,0.875,bilinear
-xception41,93.410,6.590,98.420,1.580,26.97,299,0.903,bicubic
+xception,93.470,6.530,98.530,1.470,22.86,299,0.897,bicubic
+gluon_inception_v3,93.460,6.540,98.570,1.430,23.83,299,0.875,bicubic
+mixnet_l,93.450,6.550,98.220,1.780,7.33,224,0.875,bicubic
+xception41,93.430,6.570,98.430,1.570,26.97,299,0.903,bicubic
+res2net50_26w_8s,93.410,6.590,98.180,1.820,48.40,224,0.875,bilinear
+res2net50_26w_6s,93.410,6.590,98.280,1.720,37.05,224,0.875,bilinear
+legacy_seresnet152,93.400,6.600,98.340,1.660,66.82,224,0.875,bilinear
 xcit_tiny_12_p16_224_dist,93.400,6.600,98.490,1.510,6.72,224,1.000,bicubic
-res2net50_26w_6s,93.400,6.600,98.280,1.720,37.05,224,0.875,bilinear
-resnest26d,93.360,6.640,98.640,1.360,17.07,224,0.875,bilinear
-levit_128,93.340,6.660,98.380,1.620,9.21,224,0.900,bicubic
-dla169,93.340,6.660,98.600,1.400,53.39,224,0.875,bilinear
+dla169,93.340,6.660,98.590,1.410,53.39,224,0.875,bilinear
+bat_resnext26ts,93.330,6.670,98.350,1.650,10.73,256,0.900,bicubic
+levit_128,93.330,6.670,98.380,1.620,9.21,224,0.900,bicubic
 repvgg_b1,93.330,6.670,98.510,1.490,57.42,224,0.875,bilinear
-tf_inception_v3,93.330,6.670,98.040,1.960,23.83,299,0.875,bicubic
-tv_resnet152,93.330,6.670,98.390,1.610,60.19,224,0.875,bilinear
-tf_mixnet_l,93.310,6.690,98.030,1.970,7.33,224,0.875,bicubic
-bat_resnext26ts,93.310,6.690,98.350,1.650,10.73,256,0.900,bicubic
-legacy_seresnet101,93.300,6.700,98.500,1.500,49.33,224,0.875,bilinear
-selecsls60b,93.290,6.710,98.280,1.720,32.77,224,0.875,bicubic
-efficientnet_b1,93.240,6.760,98.290,1.710,7.79,256,1.000,bicubic
-coat_lite_tiny,93.220,6.780,98.270,1.730,5.72,224,0.900,bicubic
-efficientnet_es,93.200,6.800,98.400,1.600,5.44,224,0.875,bicubic
-hrnet_w30,93.200,6.800,98.410,1.590,37.71,224,0.875,bilinear
+tf_mixnet_l,93.320,6.680,98.030,1.970,7.33,224,0.875,bicubic
+resnest26d,93.320,6.680,98.630,1.370,17.07,224,0.875,bilinear
+tf_inception_v3,93.320,6.680,98.030,1.970,23.83,299,0.875,bicubic
+tv_resnet152,93.310,6.690,98.390,1.610,60.19,224,0.875,bilinear
+legacy_seresnet101,93.300,6.700,98.510,1.490,49.33,224,0.875,bilinear
+selecsls60b,93.300,6.700,98.280,1.720,32.77,224,0.875,bicubic
+efficientnet_b1,93.250,6.750,98.290,1.710,7.79,256,1.000,bicubic
+coat_lite_tiny,93.230,6.770,98.260,1.740,5.72,224,0.900,bicubic
+hrnet_w30,93.190,6.810,98.410,1.590,37.71,224,0.875,bilinear
+dla60_res2net,93.180,6.820,98.420,1.580,20.85,224,0.875,bilinear
 dla60_res2next,93.180,6.820,98.410,1.590,17.03,224,0.875,bilinear
-dla60_res2net,93.160,6.840,98.410,1.590,20.85,224,0.875,bilinear
+efficientnet_es,93.140,6.860,98.420,1.580,5.44,224,0.875,bicubic
+dla60x,93.120,6.880,98.510,1.490,17.35,224,0.875,bilinear
 regnetx_032,93.120,6.880,98.390,1.610,15.30,224,0.875,bicubic
-pit_xs_224,93.120,6.880,98.320,1.680,10.62,224,0.900,bicubic
+pit_xs_224,93.110,6.890,98.320,1.680,10.62,224,0.900,bicubic
 tf_efficientnetv2_b0,93.110,6.890,98.390,1.610,7.14,224,0.875,bicubic
-dla60x,93.090,6.910,98.490,1.510,17.35,224,0.875,bilinear
-dla102,93.080,6.920,98.540,1.460,33.27,224,0.875,bilinear
-gluon_resnet50_v1c,93.030,6.970,98.370,1.630,25.58,224,0.875,bicubic
-regnety_016,93.030,6.970,98.360,1.640,11.20,224,0.875,bicubic
+dla102,93.060,6.940,98.550,1.450,33.27,224,0.875,bilinear
 rexnet_100,93.030,6.970,98.190,1.810,4.80,224,0.875,bicubic
+regnety_016,93.030,6.970,98.360,1.640,11.20,224,0.875,bicubic
+gluon_resnet50_v1c,93.030,6.970,98.390,1.610,25.58,224,0.875,bicubic
 selecsls60,93.020,6.980,98.310,1.690,30.67,224,0.875,bicubic
-repvgg_b1g4,93.000,7.000,98.430,1.570,39.97,224,0.875,bilinear
-legacy_seresnet50,92.950,7.050,98.190,1.810,28.09,224,0.875,bilinear
+repvgg_b1g4,92.980,7.020,98.430,1.570,39.97,224,0.875,bilinear
+legacy_seresnet50,92.960,7.040,98.190,1.810,28.09,224,0.875,bilinear
 hardcorenas_f,92.950,7.050,98.160,1.840,8.20,224,0.875,bilinear
-tf_efficientnet_em,92.950,7.050,98.210,1.790,6.90,240,0.882,bicubic
-crossvit_9_dagger_240,92.920,7.080,98.250,1.750,8.78,240,0.875,bicubic
-adv_inception_v3,92.890,7.110,98.130,1.870,23.83,299,0.875,bicubic
-res2next50,92.850,7.150,98.180,1.820,24.67,224,0.875,bilinear
-gmixer_24_224,92.840,7.160,97.880,2.120,24.72,224,0.875,bicubic
+tf_efficientnet_em,92.930,7.070,98.200,1.800,6.90,240,0.882,bicubic
+adv_inception_v3,92.890,7.110,98.140,1.860,23.83,299,0.875,bicubic
+crossvit_9_dagger_240,92.890,7.110,98.230,1.770,8.78,240,0.875,bicubic
+res2next50,92.860,7.140,98.190,1.810,24.67,224,0.875,bilinear
+gmixer_24_224,92.830,7.170,97.880,2.120,24.72,224,0.875,bicubic
 resmlp_12_distilled_224,92.830,7.170,98.140,1.860,15.35,224,0.875,bicubic
-tf_efficientnet_cc_b0_8e,92.820,7.180,98.180,1.820,24.01,224,0.875,bicubic
-seresnext26t_32x4d,92.810,7.190,98.370,1.630,16.81,224,0.875,bicubic
-tv_resnet101,92.810,7.190,98.230,1.770,44.55,224,0.875,bilinear
+tf_efficientnet_cc_b0_8e,92.830,7.170,98.180,1.820,24.01,224,0.875,bicubic
+seresnext26t_32x4d,92.820,7.180,98.370,1.630,16.81,224,0.875,bicubic
+tv_resnet101,92.820,7.180,98.250,1.750,44.55,224,0.875,bilinear
+gcresnext26ts,92.780,7.220,98.260,1.740,10.48,256,0.900,bicubic
 efficientnet_b1_pruned,92.770,7.230,98.040,1.960,6.33,240,0.882,bicubic
-gcresnext26ts,92.770,7.230,98.270,1.730,10.48,256,0.900,bicubic
-tv_resnext50_32x4d,92.760,7.240,98.280,1.720,25.03,224,0.875,bilinear
-densenet201,92.750,7.250,98.240,1.760,20.01,224,0.875,bicubic
-resnet26t,92.750,7.250,98.240,1.760,16.01,256,0.940,bicubic
-seresnext26d_32x4d,92.740,7.260,98.150,1.850,16.81,224,0.875,bicubic
-inception_v3,92.720,7.280,97.960,2.040,23.83,299,0.875,bicubic
-res2net50_14w_8s,92.720,7.280,98.190,1.810,25.06,224,0.875,bilinear
-resnet34d,92.700,7.300,98.300,1.700,21.82,224,0.875,bicubic
-seresnext26ts,92.680,7.320,98.300,1.700,10.39,256,0.900,bicubic
-efficientnet_b0,92.670,7.330,98.080,1.920,5.29,224,0.875,bicubic
-eca_resnext26ts,92.660,7.340,98.260,1.740,10.30,256,0.900,bicubic
-tf_efficientnet_lite2,92.660,7.340,98.230,1.770,6.09,260,0.890,bicubic
-legacy_seresnext26_32x4d,92.630,7.370,98.120,1.880,16.79,224,0.875,bicubic
-tf_efficientnet_cc_b0_4e,92.620,7.380,98.080,1.920,13.31,224,0.875,bicubic
-tf_efficientnet_lite1,92.620,7.380,98.080,1.920,5.42,240,0.882,bicubic
-hardcorenas_e,92.580,7.420,98.110,1.890,8.07,224,0.875,bilinear
-res2net50_48w_2s,92.550,7.450,98.080,1.920,25.29,224,0.875,bilinear
-gluon_resnet50_v1b,92.540,7.460,98.190,1.810,25.56,224,0.875,bicubic
-densenet161,92.500,7.500,98.290,1.710,28.68,224,0.875,bicubic
-xcit_tiny_12_p16_224,92.490,7.510,98.250,1.750,6.72,224,1.000,bicubic
-res2net50_26w_4s,92.480,7.520,98.070,1.930,25.70,224,0.875,bilinear
-mixnet_m,92.440,7.560,97.870,2.130,5.01,224,0.875,bicubic
-convmixer_1024_20_ks9_p14,92.430,7.570,98.270,1.730,24.38,224,0.960,bicubic
-hardcorenas_d,92.420,7.580,98.070,1.930,7.50,224,0.875,bilinear
+densenet201,92.750,7.250,98.230,1.770,20.01,224,0.875,bicubic
+resnet26t,92.750,7.250,98.230,1.770,16.01,256,0.940,bicubic
+tv_resnext50_32x4d,92.750,7.250,98.270,1.730,25.03,224,0.875,bilinear
+res2net50_14w_8s,92.740,7.260,98.180,1.820,25.06,224,0.875,bilinear
+inception_v3,92.720,7.280,97.970,2.030,23.83,299,0.875,bicubic
+efficientnet_b0,92.690,7.310,98.070,1.930,5.29,224,0.875,bicubic
+seresnext26d_32x4d,92.690,7.310,98.150,1.850,16.81,224,0.875,bicubic
+seresnext26ts,92.690,7.310,98.290,1.710,10.39,256,0.900,bicubic
+resnet34d,92.680,7.320,98.310,1.690,21.82,224,0.875,bicubic
+tf_efficientnet_lite2,92.650,7.350,98.230,1.770,6.09,260,0.890,bicubic
+legacy_seresnext26_32x4d,92.640,7.360,98.130,1.870,16.79,224,0.875,bicubic
+tf_efficientnet_lite1,92.620,7.380,98.070,1.930,5.42,240,0.882,bicubic
+eca_resnext26ts,92.610,7.390,98.260,1.740,10.30,256,0.900,bicubic
+tf_efficientnet_cc_b0_4e,92.600,7.400,98.080,1.920,13.31,224,0.875,bicubic
+hardcorenas_e,92.570,7.430,98.110,1.890,8.07,224,0.875,bilinear
+gluon_resnet50_v1b,92.540,7.460,98.170,1.830,25.56,224,0.875,bicubic
+res2net50_48w_2s,92.540,7.460,98.090,1.910,25.29,224,0.875,bilinear
+xcit_tiny_12_p16_224,92.500,7.500,98.240,1.760,6.72,224,1.000,bicubic
+densenet161,92.490,7.510,98.290,1.710,28.68,224,0.875,bicubic
+res2net50_26w_4s,92.490,7.510,98.060,1.940,25.70,224,0.875,bilinear
+tinynet_a,92.440,7.560,98.080,1.920,6.19,192,0.875,bicubic
+mixnet_m,92.430,7.570,97.870,2.130,5.01,224,0.875,bicubic
+convmixer_1024_20_ks9_p14,92.420,7.580,98.270,1.730,24.38,224,0.960,bicubic
+hardcorenas_d,92.400,7.600,98.070,1.930,7.50,224,0.875,bilinear
 mobilenetv2_120d,92.400,7.600,98.050,1.950,5.83,224,0.875,bicubic
-skresnet34,92.380,7.620,98.140,1.860,22.28,224,0.875,bicubic
+skresnet34,92.390,7.610,98.150,1.850,22.28,224,0.875,bicubic
 tf_mixnet_m,92.330,7.670,97.890,2.110,5.01,224,0.875,bicubic
-hrnet_w18,92.310,7.690,98.250,1.750,21.30,224,0.875,bilinear
-selecsls42b,92.300,7.700,98.140,1.860,32.46,224,0.875,bicubic
-mobilenetv3_large_100_miil,92.270,7.730,97.640,2.360,5.48,224,0.875,bilinear
-ese_vovnet19b_dw,92.270,7.730,98.100,1.900,6.54,224,0.875,bicubic
-tf_efficientnet_b0,92.230,7.770,98.000,2.000,5.29,224,0.875,bicubic
-tf_efficientnet_b0_ap,92.220,7.780,98.020,1.980,5.29,224,0.875,bicubic
-dla60,92.220,7.780,98.110,1.890,22.04,224,0.875,bilinear
-resmlp_12_224,92.190,7.810,98.160,1.840,15.35,224,0.875,bicubic
+hrnet_w18,92.320,7.680,98.250,1.750,21.30,224,0.875,bilinear
+ese_vovnet19b_dw,92.280,7.720,98.090,1.910,6.54,224,0.875,bicubic
+selecsls42b,92.280,7.720,98.140,1.860,32.46,224,0.875,bicubic
+mobilenetv3_large_100_miil,92.260,7.740,97.640,2.360,5.48,224,0.875,bilinear
+tf_efficientnet_b0,92.260,7.740,98.000,2.000,5.29,224,0.875,bicubic
+dla60,92.230,7.770,98.110,1.890,22.04,224,0.875,bilinear
+resmlp_12_224,92.210,7.790,98.160,1.840,15.35,224,0.875,bicubic
+tf_efficientnet_b0_ap,92.200,7.800,98.020,1.980,5.29,224,0.875,bicubic
 regnetx_016,92.160,7.840,98.210,1.790,9.19,224,0.875,bicubic
 gernet_s,92.140,7.860,98.190,1.810,8.17,224,0.875,bilinear
 xcit_nano_12_p8_224_dist,92.100,7.900,98.160,1.840,3.05,224,1.000,bicubic
-resnet26d,92.050,7.950,97.960,2.040,16.01,224,0.875,bicubic
-vit_small_patch32_224,92.030,7.970,98.230,1.770,22.88,224,0.900,bicubic
-vit_tiny_r_s16_p8_384,92.030,7.970,98.290,1.710,6.36,384,1.000,bicubic
-dpn68,92.020,7.980,98.040,1.960,12.61,224,0.875,bicubic
+resnet26d,92.070,7.930,97.970,2.030,16.01,224,0.875,bicubic
+vit_small_patch32_224,92.040,7.960,98.230,1.770,22.88,224,0.900,bicubic
+vit_tiny_r_s16_p8_384,92.040,7.960,98.290,1.710,6.36,384,1.000,bicubic
+dpn68,92.030,7.970,98.050,1.950,12.61,224,0.875,bicubic
 hardcorenas_c,92.020,7.980,97.840,2.160,5.52,224,0.875,bilinear
-tf_efficientnet_es,91.990,8.010,97.870,2.130,5.44,224,0.875,bicubic
-levit_128s,91.950,8.050,98.060,1.940,7.78,224,0.900,bicubic
+tf_efficientnet_es,91.980,8.020,97.870,2.130,5.44,224,0.875,bicubic
+levit_128s,91.960,8.040,98.060,1.940,7.78,224,0.900,bicubic
 repvgg_a2,91.940,8.060,98.150,1.850,28.21,224,0.875,bilinear
-densenet169,91.910,8.090,98.100,1.900,14.15,224,0.875,bicubic
-densenetblur121d,91.910,8.090,98.090,1.910,8.00,224,0.875,bicubic
-tv_resnet50,91.890,8.110,98.040,1.960,25.56,224,0.875,bilinear
-mixer_b16_224,91.860,8.140,97.230,2.770,59.88,224,0.875,bicubic
-resnext26ts,91.860,8.140,97.930,2.070,10.30,256,0.900,bicubic
-xcit_nano_12_p16_384_dist,91.830,8.170,98.010,1.990,3.05,384,1.000,bicubic
-mobilenetv2_140,91.830,8.170,97.860,2.140,6.11,224,0.875,bicubic
-mixnet_s,91.820,8.180,97.690,2.310,4.13,224,0.875,bicubic
+densenet169,91.920,8.080,98.100,1.900,14.15,224,0.875,bicubic
+densenetblur121d,91.910,8.090,98.070,1.930,8.00,224,0.875,bicubic
+tv_resnet50,91.880,8.120,98.040,1.960,25.56,224,0.875,bilinear
+resnext26ts,91.870,8.130,97.920,2.080,10.30,256,0.900,bicubic
+mixer_b16_224,91.860,8.140,97.250,2.750,59.88,224,0.875,bicubic
+mobilenetv2_140,91.840,8.160,97.850,2.150,6.11,224,0.875,bicubic
+mixnet_s,91.830,8.170,97.690,2.310,4.13,224,0.875,bicubic
+xcit_nano_12_p16_384_dist,91.830,8.170,98.020,1.980,3.05,384,1.000,bicubic
+hardcorenas_b,91.780,8.220,97.780,2.220,5.18,224,0.875,bilinear
 vit_tiny_patch16_224,91.760,8.240,98.040,1.960,5.72,224,0.900,bicubic
-hardcorenas_b,91.740,8.260,97.780,2.220,5.18,224,0.875,bilinear
-resnest14d,91.730,8.270,97.870,2.130,10.61,224,0.875,bilinear
-regnety_008,91.710,8.290,98.180,1.820,6.26,224,0.875,bicubic
-densenet121,91.570,8.430,98.030,1.970,7.98,224,0.875,bicubic
-tf_mixnet_s,91.510,8.490,97.610,2.390,4.13,224,0.875,bicubic
-repvgg_b0,91.450,8.550,97.980,2.020,15.82,224,0.875,bilinear
+regnety_008,91.720,8.280,98.180,1.820,6.26,224,0.875,bicubic
+resnest14d,91.720,8.280,97.870,2.130,10.61,224,0.875,bilinear
+densenet121,91.580,8.420,98.030,1.970,7.98,224,0.875,bicubic
+tf_mixnet_s,91.510,8.490,97.620,2.380,4.13,224,0.875,bicubic
+repvgg_b0,91.420,8.580,97.990,2.010,15.82,224,0.875,bilinear
 regnety_006,91.380,8.620,97.710,2.290,6.06,224,0.875,bicubic
-mobilenetv3_large_100,91.340,8.660,97.710,2.290,5.48,224,0.875,bicubic
-hardcorenas_a,91.340,8.660,97.860,2.140,5.26,224,0.875,bilinear
-semnasnet_100,91.280,8.720,97.570,2.430,3.89,224,0.875,bicubic
-tf_mobilenetv3_large_100,91.230,8.770,97.660,2.340,5.48,224,0.875,bilinear
+hardcorenas_a,91.350,8.650,97.860,2.140,5.26,224,0.875,bilinear
+mobilenetv3_large_100,91.330,8.670,97.710,2.290,5.48,224,0.875,bicubic
+semnasnet_100,91.270,8.730,97.560,2.440,3.89,224,0.875,bicubic
+tf_mobilenetv3_large_100,91.240,8.760,97.660,2.340,5.48,224,0.875,bilinear
 mobilenetv3_rw,91.210,8.790,97.660,2.340,5.48,224,0.875,bicubic
-efficientnet_es_pruned,91.190,8.810,97.740,2.260,5.44,224,0.875,bicubic
 hrnet_w18_small_v2,91.190,8.810,97.900,2.100,15.60,224,0.875,bilinear
-efficientnet_lite0,91.140,8.860,97.630,2.370,4.65,224,0.875,bicubic
-resnet26,91.130,8.870,97.740,2.260,16.00,224,0.875,bicubic
-resnet34,91.120,8.880,97.630,2.370,21.80,224,0.875,bilinear
+efficientnet_es_pruned,91.180,8.820,97.750,2.250,5.44,224,0.875,bicubic
+efficientnet_lite0,91.130,8.870,97.620,2.380,4.65,224,0.875,bicubic
+resnet34,91.130,8.870,97.620,2.380,21.80,224,0.875,bilinear
+resnet26,91.120,8.880,97.750,2.250,16.00,224,0.875,bicubic
 regnetx_008,91.050,8.950,97.710,2.290,7.26,224,0.875,bicubic
-tf_efficientnet_lite0,91.050,8.950,97.570,2.430,4.65,224,0.875,bicubic
-xcit_nano_12_p8_224,90.990,9.010,97.800,2.200,3.05,224,1.000,bicubic
-gluon_resnet34_v1b,90.990,9.010,97.650,2.350,21.80,224,0.875,bicubic
-mobilenetv2_110d,90.970,9.030,97.560,2.440,4.52,224,0.875,bicubic
+tf_efficientnet_lite0,91.050,8.950,97.590,2.410,4.65,224,0.875,bicubic
+xcit_nano_12_p8_224,91.010,8.990,97.800,2.200,3.05,224,1.000,bicubic
+gluon_resnet34_v1b,90.960,9.040,97.640,2.360,21.80,224,0.875,bicubic
+mobilenetv2_110d,90.960,9.040,97.560,2.440,4.52,224,0.875,bicubic
+tinynet_b,90.930,9.070,97.670,2.330,3.73,188,0.875,bicubic
 legacy_seresnet34,90.900,9.100,97.580,2.420,21.96,224,0.875,bilinear
-tv_densenet121,90.900,9.100,97.700,2.300,7.98,224,0.875,bicubic
-pit_ti_distilled_224,90.880,9.120,97.720,2.280,5.10,224,0.900,bicubic
-dla34,90.770,9.230,97.650,2.350,15.74,224,0.875,bilinear
-deit_tiny_distilled_patch16_224,90.730,9.270,97.580,2.420,5.91,224,0.900,bicubic
-fbnetc_100,90.720,9.280,97.210,2.790,5.57,224,0.875,bilinear
-swsl_resnet18,90.680,9.320,97.710,2.290,11.69,224,0.875,bilinear
-crossvit_9_240,90.660,9.340,97.740,2.260,8.55,240,0.875,bicubic
-convit_tiny,90.610,9.390,97.730,2.270,5.71,224,0.875,bicubic
+pit_ti_distilled_224,90.900,9.100,97.710,2.290,5.10,224,0.900,bicubic
+tv_densenet121,90.890,9.110,97.710,2.290,7.98,224,0.875,bicubic
+dla34,90.770,9.230,97.660,2.340,15.74,224,0.875,bilinear
+deit_tiny_distilled_patch16_224,90.710,9.290,97.570,2.430,5.91,224,0.900,bicubic
+fbnetc_100,90.710,9.290,97.210,2.790,5.57,224,0.875,bilinear
+swsl_resnet18,90.690,9.310,97.700,2.300,11.69,224,0.875,bilinear
+crossvit_9_240,90.640,9.360,97.740,2.260,8.55,240,0.875,bicubic
+convit_tiny,90.640,9.360,97.740,2.260,5.71,224,0.875,bicubic
 mnasnet_100,90.510,9.490,97.470,2.530,4.38,224,0.875,bicubic
-regnety_004,90.480,9.520,97.560,2.440,4.34,224,0.875,bicubic
+regnety_004,90.490,9.510,97.540,2.460,4.34,224,0.875,bicubic
+regnetx_006,90.350,9.650,97.430,2.570,6.20,224,0.875,bicubic
 spnasnet_100,90.350,9.650,97.190,2.810,4.42,224,0.875,bilinear
-regnetx_006,90.320,9.680,97.430,2.570,6.20,224,0.875,bicubic
-crossvit_tiny_240,90.250,9.750,97.610,2.390,7.01,240,0.875,bicubic
-ssl_resnet18,90.230,9.770,97.560,2.440,11.69,224,0.875,bilinear
-vgg19_bn,90.100,9.900,97.580,2.420,143.68,224,0.875,bilinear
+crossvit_tiny_240,90.250,9.750,97.590,2.410,7.01,240,0.875,bicubic
+ssl_resnet18,90.220,9.780,97.550,2.450,11.69,224,0.875,bilinear
 vgg16_bn,90.090,9.910,97.370,2.630,138.37,224,0.875,bilinear
+vgg19_bn,90.080,9.920,97.580,2.420,143.68,224,0.875,bilinear
+semnasnet_075,90.080,9.920,97.430,2.570,2.91,224,0.875,bicubic
 ghostnet_100,90.030,9.970,97.370,2.630,5.18,224,0.875,bilinear
-pit_ti_224,89.930,10.070,97.440,2.560,4.85,224,0.900,bicubic
-tv_resnet34,89.920,10.080,97.340,2.660,21.80,224,0.875,bilinear
-vit_base_patch32_sam_224,89.750,10.250,97.000,3.000,88.22,224,0.900,bicubic
-tf_mobilenetv3_large_075,89.710,10.290,97.220,2.780,3.99,224,0.875,bilinear
+pit_ti_224,89.950,10.050,97.450,2.550,4.85,224,0.900,bicubic
+tv_resnet34,89.930,10.070,97.340,2.660,21.80,224,0.875,bilinear
+vit_base_patch32_224_sam,89.750,10.250,97.000,3.000,88.22,224,0.900,bicubic
+tf_mobilenetv3_large_075,89.680,10.320,97.210,2.790,3.99,224,0.875,bilinear
 xcit_nano_12_p16_224_dist,89.680,10.320,97.090,2.910,3.05,224,1.000,bicubic
 deit_tiny_patch16_224,89.670,10.330,97.440,2.560,5.72,224,0.900,bicubic
-skresnet18,89.660,10.340,97.220,2.780,11.96,224,0.875,bicubic
+skresnet18,89.660,10.340,97.230,2.770,11.96,224,0.875,bicubic
 mobilenetv2_100,89.610,10.390,97.150,2.850,3.50,224,0.875,bicubic
-resnet18d,89.290,10.710,97.140,2.860,11.71,224,0.875,bicubic
-vit_tiny_r_s16_p8_224,89.190,10.810,97.230,2.770,6.34,224,0.900,bicubic
-vgg19,89.060,10.940,96.870,3.130,143.67,224,0.875,bilinear
-hrnet_w18_small,89.060,10.940,97.100,2.900,13.19,224,0.875,bilinear
-tf_mobilenetv3_large_minimal_100,88.950,11.050,96.870,3.130,3.92,224,0.875,bilinear
-legacy_seresnet18,88.880,11.120,96.970,3.030,11.78,224,0.875,bicubic
-regnetx_004,88.880,11.120,97.120,2.880,5.16,224,0.875,bicubic
-vgg13_bn,88.780,11.220,96.970,3.030,133.05,224,0.875,bilinear
-xcit_nano_12_p16_224,88.590,11.410,96.790,3.210,3.05,224,1.000,bicubic
+resnet18d,89.270,10.730,97.140,2.860,11.71,224,0.875,bicubic
+vit_tiny_r_s16_p8_224,89.170,10.830,97.230,2.770,6.34,224,0.900,bicubic
+hrnet_w18_small,89.050,10.950,97.110,2.890,13.19,224,0.875,bilinear
+vgg19,89.040,10.960,96.870,3.130,143.67,224,0.875,bilinear
+tf_mobilenetv3_large_minimal_100,88.960,11.040,96.860,3.140,3.92,224,0.875,bilinear
+regnetx_004,88.900,11.100,97.120,2.880,5.16,224,0.875,bicubic
+legacy_seresnet18,88.880,11.120,96.980,3.020,11.78,224,0.875,bicubic
+lcnet_100,88.790,11.210,96.730,3.270,2.95,224,0.875,bicubic
+vgg13_bn,88.760,11.240,96.970,3.030,133.05,224,0.875,bilinear
+xcit_nano_12_p16_224,88.610,11.390,96.790,3.210,3.05,224,1.000,bicubic
 vgg16,88.550,11.450,96.790,3.210,138.36,224,0.875,bilinear
-gluon_resnet18_v1b,88.380,11.620,96.700,3.300,11.69,224,0.875,bicubic
-vgg11_bn,87.520,12.480,96.810,3.190,132.87,224,0.875,bilinear
-resnet18,87.370,12.630,96.260,3.740,11.69,224,0.875,bilinear
-regnety_002,87.360,12.640,96.570,3.430,3.16,224,0.875,bicubic
-mixer_l16_224,87.160,12.840,93.530,6.470,208.20,224,0.875,bicubic
-vgg13,87.030,12.970,96.310,3.690,133.05,224,0.875,bilinear
-vgg11,86.580,13.420,96.290,3.710,132.86,224,0.875,bilinear
-dla60x_c,86.280,13.720,96.160,3.840,1.32,224,0.875,bilinear
-regnetx_002,86.200,13.800,95.970,4.030,2.68,224,0.875,bicubic
-tf_mobilenetv3_small_100,85.210,14.790,95.780,4.220,2.54,224,0.875,bilinear
-dla46x_c,84.250,15.750,95.280,4.720,1.07,224,0.875,bilinear
-dla46_c,83.640,16.360,94.910,5.090,1.30,224,0.875,bilinear
-tf_mobilenetv3_small_075,83.480,16.520,94.800,5.200,2.04,224,0.875,bilinear
-tf_mobilenetv3_small_minimal_100,81.390,18.610,93.670,6.330,2.04,224,0.875,bilinear
+gluon_resnet18_v1b,88.400,11.600,96.680,3.320,11.69,224,0.875,bicubic
+tinynet_c,87.770,12.230,96.370,3.630,2.46,184,0.875,bicubic
+vgg11_bn,87.500,12.500,96.820,3.180,132.87,224,0.875,bilinear
+resnet18,87.390,12.610,96.290,3.710,11.69,224,0.875,bilinear
+regnety_002,87.380,12.620,96.590,3.410,3.16,224,0.875,bicubic
+mixer_l16_224,87.150,12.850,93.510,6.490,208.20,224,0.875,bicubic
+vgg13,87.050,12.950,96.320,3.680,133.05,224,0.875,bilinear
+vgg11,86.550,13.450,96.280,3.720,132.86,224,0.875,bilinear
+dla60x_c,86.290,13.710,96.160,3.840,1.32,224,0.875,bilinear
+regnetx_002,86.190,13.810,95.980,4.020,2.68,224,0.875,bicubic
+lcnet_075,85.990,14.010,95.680,4.320,2.36,224,0.875,bicubic
+tf_mobilenetv3_small_100,85.210,14.790,95.770,4.230,2.54,224,0.875,bilinear
+tinynet_d,84.750,15.250,95.180,4.820,2.34,152,0.875,bicubic
+dla46x_c,84.250,15.750,95.270,4.730,1.07,224,0.875,bilinear
+mnasnet_small,83.990,16.010,94.920,5.080,2.03,224,0.875,bicubic
+mobilenetv2_050,83.890,16.110,94.710,5.290,1.97,224,0.875,bicubic
+dla46_c,83.650,16.350,94.920,5.080,1.30,224,0.875,bilinear
+tf_mobilenetv3_small_075,83.510,16.490,94.800,5.200,2.04,224,0.875,bilinear
+lcnet_050,81.780,18.220,93.710,6.290,1.88,224,0.875,bicubic
+tf_mobilenetv3_small_minimal_100,81.380,18.620,93.670,6.330,2.04,224,0.875,bilinear
+tinynet_e,78.900,21.100,92.560,7.440,2.04,106,0.875,bicubic

results/results-imagenet-r-clean.csv

@@ -1,522 +1,553 @@
 model,top1,top1_err,top5,top5_err,param_count,img_size,cropt_pct,interpolation
-beit_large_patch16_384,97.820,2.180,99.790,0.210,305.00,384,1.000,bicubic
-beit_large_patch16_512,97.770,2.230,99.810,0.190,305.67,512,1.000,bicubic
-tf_efficientnet_l2_ns,97.770,2.230,99.890,0.110,480.31,800,0.960,bicubic
+beit_large_patch16_384,97.810,2.190,99.790,0.210,305.00,384,1.000,bicubic
+beit_large_patch16_512,97.780,2.220,99.820,0.180,305.67,512,1.000,bicubic
+tf_efficientnet_l2_ns,97.780,2.220,99.890,0.110,480.31,800,0.960,bicubic
 tf_efficientnet_l2_ns_475,97.750,2.250,99.820,0.180,480.31,475,0.936,bicubic
-beit_large_patch16_224,97.470,2.530,99.690,0.310,304.43,224,0.900,bicubic
+convnext_xlarge_384_in22ft1k,97.550,2.450,99.800,0.200,350.20,384,1.000,bicubic
+beit_large_patch16_224,97.480,2.520,99.690,0.310,304.43,224,0.900,bicubic
+convnext_large_384_in22ft1k,97.440,2.560,99.780,0.220,197.77,384,1.000,bicubic
 vit_large_patch16_384,97.420,2.580,99.780,0.220,304.72,384,1.000,bicubic
-beit_base_patch16_384,97.350,2.650,99.710,0.290,86.74,384,1.000,bicubic
-tf_efficientnet_b7_ns,97.210,2.790,99.700,0.300,66.35,600,0.949,bicubic
-swin_large_patch4_window12_384,97.180,2.820,99.690,0.310,196.74,384,1.000,bicubic
+beit_base_patch16_384,97.330,2.670,99.720,0.280,86.74,384,1.000,bicubic
+convnext_base_384_in22ft1k,97.280,2.720,99.770,0.230,88.59,384,1.000,bicubic
+convnext_large_in22ft1k,97.260,2.740,99.650,0.350,197.77,224,0.875,bicubic
+convnext_xlarge_in22ft1k,97.240,2.760,99.730,0.270,350.20,224,0.875,bicubic
+tf_efficientnet_b7_ns,97.200,2.800,99.700,0.300,66.35,600,0.949,bicubic
+swin_large_patch4_window12_384,97.180,2.820,99.680,0.320,196.74,384,1.000,bicubic
 tf_efficientnetv2_xl_in21ft1k,97.150,2.850,99.620,0.380,208.12,512,1.000,bicubic
-tf_efficientnetv2_l_in21ft1k,97.110,2.890,99.700,0.300,118.52,480,1.000,bicubic
-swin_base_patch4_window12_384,97.070,2.930,99.770,0.230,87.90,384,1.000,bicubic
-tf_efficientnet_b6_ns,97.030,2.970,99.710,0.290,43.04,528,0.942,bicubic
-vit_base_patch16_384,97.020,2.980,99.700,0.300,86.86,384,1.000,bicubic
-ig_resnext101_32x48d,96.960,3.040,99.670,0.330,828.41,224,0.875,bilinear
+swin_base_patch4_window12_384,97.120,2.880,99.780,0.220,87.90,384,1.000,bicubic
+tf_efficientnetv2_l_in21ft1k,97.110,2.890,99.710,0.290,118.52,480,1.000,bicubic
+vit_base_patch8_224,97.080,2.920,99.620,0.380,86.58,224,0.900,bicubic
+tf_efficientnet_b6_ns,97.020,2.980,99.710,0.290,43.04,528,0.942,bicubic
+vit_base_patch16_384,97.020,2.980,99.710,0.290,86.86,384,1.000,bicubic
+ig_resnext101_32x48d,96.970,3.030,99.670,0.330,828.41,224,0.875,bilinear
+tf_efficientnetv2_m_in21ft1k,96.970,3.030,99.610,0.390,54.14,480,1.000,bicubic
 swin_large_patch4_window7_224,96.950,3.050,99.660,0.340,196.53,224,0.900,bicubic
-tf_efficientnetv2_m_in21ft1k,96.950,3.050,99.610,0.390,54.14,480,1.000,bicubic
-vit_large_r50_s32_384,96.950,3.050,99.720,0.280,329.09,384,1.000,bicubic
-xcit_large_24_p16_384_dist,96.930,3.070,99.510,0.490,189.10,384,1.000,bicubic
-dm_nfnet_f6,96.910,3.090,99.720,0.280,438.36,576,0.956,bicubic
+vit_large_r50_s32_384,96.950,3.050,99.710,0.290,329.09,384,1.000,bicubic
+xcit_large_24_p16_384_dist,96.940,3.060,99.510,0.490,189.10,384,1.000,bicubic
+dm_nfnet_f6,96.920,3.080,99.720,0.280,438.36,576,0.956,bicubic
+cait_m48_448,96.880,3.120,99.620,0.380,356.46,448,1.000,bicubic
 resnetv2_152x4_bitm,96.880,3.120,99.660,0.340,936.53,480,1.000,bilinear
-tf_efficientnet_b5_ns,96.880,3.120,99.640,0.360,30.39,456,0.934,bicubic
-cait_m48_448,96.870,3.130,99.620,0.380,356.46,448,1.000,bicubic
-xcit_small_24_p8_384_dist,96.830,3.170,99.630,0.370,47.63,384,1.000,bicubic
-cait_m36_384,96.820,3.180,99.660,0.340,271.22,384,1.000,bicubic
-dm_nfnet_f5,96.800,3.200,99.670,0.330,377.21,544,0.954,bicubic
-dm_nfnet_f4,96.780,3.220,99.620,0.380,316.07,512,0.951,bicubic
+tf_efficientnet_b5_ns,96.870,3.130,99.640,0.360,30.39,456,0.934,bicubic
+convnext_base_in22ft1k,96.860,3.140,99.650,0.350,88.59,224,0.875,bicubic
+cait_m36_384,96.830,3.170,99.660,0.340,271.22,384,1.000,bicubic
+dm_nfnet_f5,96.810,3.190,99.670,0.330,377.21,544,0.954,bicubic
+xcit_small_24_p8_384_dist,96.810,3.190,99.630,0.370,47.63,384,1.000,bicubic
+xcit_medium_24_p8_384_dist,96.780,3.220,99.610,0.390,84.32,384,1.000,bicubic
 ig_resnext101_32x32d,96.780,3.220,99.530,0.470,468.53,224,0.875,bilinear
-xcit_medium_24_p8_384_dist,96.780,3.220,99.620,0.380,84.32,384,1.000,bicubic
+dm_nfnet_f4,96.780,3.220,99.620,0.380,316.07,512,0.951,bicubic
 xcit_large_24_p8_384_dist,96.760,3.240,99.560,0.440,188.93,384,1.000,bicubic
-dm_nfnet_f3,96.720,3.280,99.630,0.370,254.92,416,0.940,bicubic
+dm_nfnet_f3,96.730,3.270,99.630,0.370,254.92,416,0.940,bicubic
 tf_efficientnet_b4_ns,96.710,3.290,99.640,0.360,19.34,380,0.922,bicubic
 vit_large_patch16_224,96.710,3.290,99.650,0.350,304.33,224,0.900,bicubic
-tf_efficientnet_b8,96.700,3.300,99.550,0.450,87.41,672,0.954,bicubic
-xcit_medium_24_p16_384_dist,96.690,3.310,99.600,0.400,84.40,384,1.000,bicubic
-swin_base_patch4_window7_224,96.670,3.330,99.670,0.330,87.77,224,0.900,bicubic
-beit_base_patch16_224,96.650,3.350,99.660,0.340,86.53,224,0.900,bicubic
-tf_efficientnetv2_l,96.650,3.350,99.570,0.430,118.52,480,1.000,bicubic
-cait_s36_384,96.630,3.370,99.590,0.410,68.37,384,1.000,bicubic
-xcit_large_24_p8_224_dist,96.620,3.380,99.460,0.540,188.93,224,1.000,bicubic
-cait_s24_384,96.580,3.420,99.550,0.450,47.06,384,1.000,bicubic
+tf_efficientnet_b8,96.700,3.300,99.530,0.470,87.41,672,0.954,bicubic
+xcit_medium_24_p16_384_dist,96.700,3.300,99.600,0.400,84.40,384,1.000,bicubic
+swin_base_patch4_window7_224,96.680,3.320,99.660,0.340,87.77,224,0.900,bicubic
+beit_base_patch16_224,96.660,3.340,99.660,0.340,86.53,224,0.900,bicubic
+tf_efficientnetv2_l,96.650,3.350,99.560,0.440,118.52,480,1.000,bicubic
+xcit_large_24_p8_224_dist,96.640,3.360,99.460,0.540,188.93,224,1.000,bicubic
+cait_s36_384,96.630,3.370,99.600,0.400,68.37,384,1.000,bicubic
+regnetz_e8,96.600,3.400,99.610,0.390,57.70,320,1.000,bicubic
 tf_efficientnet_b7,96.580,3.420,99.510,0.490,66.35,600,0.949,bicubic
-tf_efficientnet_b8_ap,96.560,3.440,99.550,0.450,87.41,672,0.954,bicubic
-tf_efficientnetv2_m,96.550,3.450,99.570,0.430,54.14,480,1.000,bicubic
-xcit_small_24_p8_224_dist,96.550,3.450,99.560,0.440,47.63,224,1.000,bicubic
+cait_s24_384,96.570,3.430,99.550,0.450,47.06,384,1.000,bicubic
+tf_efficientnet_b8_ap,96.550,3.450,99.540,0.460,87.41,672,0.954,bicubic
+xcit_small_24_p8_224_dist,96.550,3.450,99.570,0.430,47.63,224,1.000,bicubic
+tf_efficientnetv2_m,96.540,3.460,99.570,0.430,54.14,480,1.000,bicubic
 resnetv2_152x2_bitm,96.520,3.480,99.590,0.410,236.34,448,1.000,bilinear
-xcit_medium_24_p8_224_dist,96.500,3.500,99.500,0.500,84.32,224,1.000,bicubic
-deit_base_distilled_patch16_384,96.490,3.510,99.590,0.410,87.63,384,1.000,bicubic
-xcit_small_12_p8_384_dist,96.480,3.520,99.480,0.520,26.21,384,1.000,bicubic
-vit_base_r50_s16_384,96.470,3.530,99.660,0.340,98.95,384,1.000,bicubic
+xcit_medium_24_p8_224_dist,96.520,3.480,99.510,0.490,84.32,224,1.000,bicubic
+deit_base_distilled_patch16_384,96.510,3.490,99.590,0.410,87.63,384,1.000,bicubic
+xcit_small_12_p8_384_dist,96.480,3.520,99.490,0.510,26.21,384,1.000,bicubic
+tf_efficientnetv2_s_in21ft1k,96.470,3.530,99.570,0.430,21.46,384,1.000,bicubic
+dm_nfnet_f2,96.460,3.540,99.540,0.460,193.78,352,0.920,bicubic
 ecaresnet269d,96.460,3.540,99.610,0.390,102.09,352,1.000,bicubic
-tf_efficientnetv2_s_in21ft1k,96.460,3.540,99.570,0.430,21.46,384,1.000,bicubic
-eca_nfnet_l2,96.460,3.540,99.630,0.370,56.72,384,1.000,bicubic
-dm_nfnet_f2,96.450,3.550,99.540,0.460,193.78,352,0.920,bicubic
-ig_resnext101_32x16d,96.430,3.570,99.540,0.460,194.03,224,0.875,bilinear
+eca_nfnet_l2,96.450,3.550,99.630,0.370,56.72,384,1.000,bicubic
+vit_base_r50_s16_384,96.450,3.550,99.660,0.340,98.95,384,1.000,bicubic
+ig_resnext101_32x16d,96.440,3.560,99.540,0.460,194.03,224,0.875,bilinear
 resnetrs420,96.400,3.600,99.540,0.460,191.89,416,1.000,bicubic
-dm_nfnet_f1,96.370,3.630,99.480,0.520,132.63,320,0.910,bicubic
-tf_efficientnet_b6_ap,96.360,3.640,99.550,0.450,43.04,528,0.942,bicubic
-xcit_small_24_p16_384_dist,96.360,3.640,99.590,0.410,47.67,384,1.000,bicubic
-tf_efficientnet_b7_ap,96.350,3.650,99.600,0.400,66.35,600,0.949,bicubic
-resmlp_big_24_224_in22ft1k,96.340,3.660,99.510,0.490,129.14,224,0.875,bicubic
-seresnet152d,96.330,3.670,99.510,0.490,66.84,320,1.000,bicubic
+dm_nfnet_f1,96.380,3.620,99.470,0.530,132.63,320,0.910,bicubic
+tf_efficientnet_b6_ap,96.370,3.630,99.550,0.450,43.04,528,0.942,bicubic
+resmlp_big_24_224_in22ft1k,96.350,3.650,99.520,0.480,129.14,224,0.875,bicubic
+tf_efficientnet_b7_ap,96.350,3.650,99.590,0.410,66.35,600,0.949,bicubic
+xcit_small_24_p16_384_dist,96.340,3.660,99.580,0.420,47.67,384,1.000,bicubic
 xcit_small_12_p16_384_dist,96.330,3.670,99.490,0.510,26.25,384,1.000,bicubic
-xcit_large_24_p16_224_dist,96.310,3.690,99.500,0.500,189.10,224,1.000,bicubic
+xcit_large_24_p16_224_dist,96.320,3.680,99.500,0.500,189.10,224,1.000,bicubic
+seresnet152d,96.310,3.690,99.510,0.490,66.84,320,1.000,bicubic
 vit_base_patch16_224,96.300,3.700,99.560,0.440,86.57,224,0.900,bicubic
-resnetv2_50x3_bitm,96.290,3.710,99.630,0.370,217.32,448,1.000,bilinear
-resnetv2_101x3_bitm,96.290,3.710,99.580,0.420,387.93,448,1.000,bilinear
-tf_efficientnet_b6,96.280,3.720,99.520,0.480,43.04,528,0.942,bicubic
+tf_efficientnet_b6,96.290,3.710,99.520,0.480,43.04,528,0.942,bicubic
 swsl_resnext101_32x16d,96.280,3.720,99.500,0.500,194.03,224,0.875,bilinear
 efficientnetv2_rw_m,96.270,3.730,99.560,0.440,53.24,416,1.000,bicubic
+resnetv2_50x3_bitm,96.270,3.730,99.630,0.370,217.32,448,1.000,bilinear
 xcit_medium_24_p16_224_dist,96.260,3.740,99.400,0.600,84.40,224,1.000,bicubic
-xcit_tiny_24_p8_384_dist,96.250,3.750,99.440,0.560,12.11,384,1.000,bicubic
+resnetv2_101x3_bitm,96.250,3.750,99.580,0.420,387.93,448,1.000,bilinear
 resnetrs350,96.240,3.760,99.470,0.530,163.96,384,1.000,bicubic
-swsl_resnext101_32x8d,96.230,3.770,99.590,0.410,88.79,224,0.875,bilinear
-vit_large_r50_s32_224,96.190,3.810,99.530,0.470,328.99,224,0.900,bicubic
-resnetv2_152x2_bit_teacher_384,96.170,3.830,99.510,0.490,236.34,384,1.000,bicubic
-resnest269e,96.120,3.880,99.520,0.480,110.93,416,0.928,bicubic
-crossvit_18_dagger_408,96.110,3.890,99.470,0.530,44.61,408,1.000,bicubic
+swsl_resnext101_32x8d,96.240,3.760,99.590,0.410,88.79,224,0.875,bilinear
+xcit_tiny_24_p8_384_dist,96.240,3.760,99.440,0.560,12.11,384,1.000,bicubic
+resnetv2_152x2_bit_teacher_384,96.190,3.810,99.500,0.500,236.34,384,1.000,bicubic
+vit_large_r50_s32_224,96.180,3.820,99.530,0.470,328.99,224,0.900,bicubic
+crossvit_18_dagger_408,96.130,3.870,99.470,0.530,44.61,408,1.000,bicubic
+resnest269e,96.130,3.870,99.520,0.480,110.93,416,0.928,bicubic
 resnet200d,96.110,3.890,99.460,0.540,64.69,320,1.000,bicubic
-tf_efficientnet_b3_ns,96.110,3.890,99.470,0.530,12.23,300,0.904,bicubic
+tf_efficientnet_b3_ns,96.100,3.900,99.480,0.520,12.23,300,0.904,bicubic
 tf_efficientnet_b5_ap,96.080,3.920,99.540,0.460,30.39,456,0.934,bicubic
-resnest200e,96.080,3.920,99.470,0.530,70.20,320,0.909,bicubic
-pit_b_distilled_224,96.080,3.920,99.380,0.620,74.79,224,0.900,bicubic
-resnetrs270,96.070,3.930,99.480,0.520,129.86,352,1.000,bicubic
-xcit_large_24_p8_224,96.060,3.940,99.150,0.850,188.93,224,1.000,bicubic
+xcit_large_24_p8_224,96.080,3.920,99.150,0.850,188.93,224,1.000,bicubic
+pit_b_distilled_224,96.070,3.930,99.380,0.620,74.79,224,0.900,bicubic
+resnest200e,96.070,3.930,99.480,0.520,70.20,320,0.909,bicubic
+resnetrs270,96.060,3.940,99.480,0.520,129.86,352,1.000,bicubic
 vit_small_r26_s32_384,96.060,3.940,99.550,0.450,36.47,384,1.000,bicubic
-swsl_resnext101_32x4d,96.050,3.950,99.540,0.460,44.18,224,0.875,bilinear
-vit_base_patch16_224_miil,96.040,3.960,99.350,0.650,86.54,224,0.875,bilinear
+swsl_resnext101_32x4d,96.040,3.960,99.530,0.470,44.18,224,0.875,bilinear
+vit_base_patch16_224_miil,96.030,3.970,99.350,0.650,86.54,224,0.875,bilinear
+convnext_large,96.020,3.980,99.470,0.530,197.77,224,0.875,bicubic
 cait_xs24_384,96.010,3.990,99.430,0.570,26.67,384,1.000,bicubic
-resnetrs200,96.000,4.000,99.440,0.560,93.21,320,1.000,bicubic
+regnetz_d8,96.010,3.990,99.520,0.480,23.37,320,1.000,bicubic
+resnetrs200,95.990,4.010,99.440,0.560,93.21,320,1.000,bicubic
 tf_efficientnet_b5,95.980,4.020,99.450,0.550,30.39,456,0.934,bicubic
-vit_small_patch16_384,95.980,4.020,99.600,0.400,22.20,384,1.000,bicubic
+vit_small_patch16_384,95.980,4.020,99.590,0.410,22.20,384,1.000,bicubic
 resnetrs152,95.960,4.040,99.380,0.620,86.62,320,1.000,bicubic
 xcit_small_12_p8_224_dist,95.960,4.040,99.420,0.580,26.21,224,1.000,bicubic
-ig_resnext101_32x8d,95.950,4.050,99.390,0.610,88.79,224,0.875,bilinear
-eca_nfnet_l1,95.920,4.080,99.500,0.500,41.41,320,1.000,bicubic
+convnext_base,95.950,4.050,99.380,0.620,88.59,224,0.875,bicubic
+ig_resnext101_32x8d,95.940,4.060,99.380,0.620,88.79,224,0.875,bilinear
+eca_nfnet_l1,95.930,4.070,99.490,0.510,41.41,320,1.000,bicubic
 xcit_small_24_p8_224,95.910,4.090,99.180,0.820,47.63,224,1.000,bicubic
-vit_base_patch32_384,95.910,4.090,99.440,0.560,88.30,384,1.000,bicubic
-regnety_160,95.900,4.100,99.560,0.440,83.59,288,1.000,bicubic
-xcit_medium_24_p8_224,95.870,4.130,99.090,0.910,84.32,224,1.000,bicubic
+vit_base_patch32_384,95.900,4.100,99.440,0.560,88.30,384,1.000,bicubic
+regnety_160,95.880,4.120,99.560,0.440,83.59,288,1.000,bicubic
 resmlp_big_24_distilled_224,95.870,4.130,99.440,0.560,129.14,224,0.875,bicubic
-regnetz_d,95.860,4.140,99.440,0.560,27.58,320,0.950,bicubic
-resnet152d,95.850,4.150,99.430,0.570,60.21,320,1.000,bicubic
-crossvit_15_dagger_408,95.820,4.180,99.300,0.700,28.50,408,1.000,bicubic
-xcit_small_24_p16_224_dist,95.810,4.190,99.340,0.660,47.67,224,1.000,bicubic
-deit_base_distilled_patch16_224,95.780,4.220,99.280,0.720,87.34,224,0.900,bicubic
+resnet152d,95.870,4.130,99.430,0.570,60.21,320,1.000,bicubic
+xcit_medium_24_p8_224,95.870,4.130,99.080,0.920,84.32,224,1.000,bicubic
+regnetz_d32,95.860,4.140,99.430,0.570,27.58,320,0.950,bicubic
+crossvit_15_dagger_408,95.820,4.180,99.310,0.690,28.50,408,1.000,bicubic
+xcit_small_24_p16_224_dist,95.800,4.200,99.340,0.660,47.67,224,1.000,bicubic
+deit_base_distilled_patch16_224,95.750,4.250,99.280,0.720,87.34,224,0.900,bicubic
 resnet101d,95.750,4.250,99.440,0.560,44.57,320,1.000,bicubic
-xcit_small_12_p16_224_dist,95.750,4.250,99.290,0.710,26.25,224,1.000,bicubic
-resnetv2_152x2_bit_teacher,95.730,4.270,99.430,0.570,236.34,224,0.875,bicubic
-twins_pcpvt_large,95.720,4.280,99.490,0.510,60.99,224,0.900,bicubic
+resnetv2_152x2_bit_teacher,95.750,4.250,99.430,0.570,236.34,224,0.875,bicubic
+xcit_small_12_p16_224_dist,95.740,4.260,99.300,0.700,26.25,224,1.000,bicubic
 twins_svt_large,95.720,4.280,99.370,0.630,99.27,224,0.900,bicubic
+twins_pcpvt_large,95.720,4.280,99.490,0.510,60.99,224,0.900,bicubic
 swin_small_patch4_window7_224,95.720,4.280,99.290,0.710,49.61,224,0.900,bicubic
 tf_efficientnetv2_s,95.710,4.290,99.400,0.600,21.46,384,1.000,bicubic
-dm_nfnet_f0,95.710,4.290,99.330,0.670,71.49,256,0.900,bicubic
 efficientnetv2_rw_s,95.700,4.300,99.380,0.620,23.94,384,1.000,bicubic
-deit_base_patch16_384,95.660,4.340,99.240,0.760,86.86,384,1.000,bicubic
-cait_s24_224,95.640,4.360,99.390,0.610,46.92,224,1.000,bicubic
-tf_efficientnet_b4,95.590,4.410,99.320,0.680,19.34,380,0.922,bicubic
-swsl_resnext50_32x4d,95.590,4.410,99.440,0.560,25.03,224,0.875,bilinear
-resnest101e,95.580,4.420,99.270,0.730,48.28,256,0.875,bilinear
+dm_nfnet_f0,95.690,4.310,99.330,0.670,71.49,256,0.900,bicubic
+cait_s24_224,95.650,4.350,99.390,0.610,46.92,224,1.000,bicubic
+deit_base_patch16_384,95.650,4.350,99.240,0.760,86.86,384,1.000,bicubic
+convnext_small,95.630,4.370,99.260,0.740,50.22,224,0.875,bicubic
+swsl_resnext50_32x4d,95.600,4.400,99.440,0.560,25.03,224,0.875,bilinear
+tf_efficientnet_b4,95.590,4.410,99.330,0.670,19.34,380,0.922,bicubic
 twins_svt_base,95.570,4.430,99.230,0.770,56.07,224,0.900,bicubic
-efficientnet_b4,95.540,4.460,99.400,0.600,19.34,384,1.000,bicubic
-jx_nest_small,95.540,4.460,99.230,0.770,38.35,224,0.875,bicubic
-jx_nest_base,95.530,4.470,99.300,0.700,67.72,224,0.875,bicubic
-tf_efficientnet_b2_ns,95.530,4.470,99.340,0.660,9.11,260,0.890,bicubic
-tresnet_xl_448,95.520,4.480,99.340,0.660,78.44,448,0.875,bilinear
-tf_efficientnet_b4_ap,95.500,4.500,99.390,0.610,19.34,380,0.922,bicubic
-regnety_032,95.470,4.530,99.320,0.680,19.44,288,1.000,bicubic
-twins_pcpvt_base,95.470,4.530,99.380,0.620,43.83,224,0.900,bicubic
-xcit_tiny_24_p16_384_dist,95.460,4.540,99.350,0.650,12.12,384,1.000,bicubic
-xcit_tiny_24_p8_224_dist,95.460,4.540,99.370,0.630,12.11,224,1.000,bicubic
+resnest101e,95.560,4.440,99.270,0.730,48.28,256,0.875,bilinear
+resnet152,95.560,4.440,99.270,0.730,60.19,224,0.950,bicubic
+jx_nest_base,95.540,4.460,99.300,0.700,67.72,224,0.875,bicubic
+jx_nest_small,95.540,4.460,99.220,0.780,38.35,224,0.875,bicubic
+efficientnet_b4,95.520,4.480,99.400,0.600,19.34,384,1.000,bicubic
+tf_efficientnet_b2_ns,95.520,4.480,99.340,0.660,9.11,260,0.890,bicubic
+tresnet_xl_448,95.510,4.490,99.340,0.660,78.44,448,0.875,bilinear
+tf_efficientnet_b4_ap,95.490,4.510,99.390,0.610,19.34,380,0.922,bicubic
+regnety_032,95.480,4.520,99.320,0.680,19.44,288,1.000,bicubic
+xcit_tiny_24_p16_384_dist,95.480,4.520,99.360,0.640,12.12,384,1.000,bicubic
+twins_pcpvt_base,95.460,4.540,99.390,0.610,43.83,224,0.900,bicubic
+xcit_tiny_24_p8_224_dist,95.460,4.540,99.360,0.640,12.11,224,1.000,bicubic
 eca_nfnet_l0,95.450,4.550,99.390,0.610,24.14,288,1.000,bicubic
-nfnet_l0,95.420,4.580,99.430,0.570,35.07,288,1.000,bicubic
 xcit_small_12_p8_224,95.420,4.580,99.200,0.800,26.21,224,1.000,bicubic
-regnetz_c,95.410,4.590,99.310,0.690,13.46,320,0.940,bicubic
-ssl_resnext101_32x16d,95.400,4.600,99.410,0.590,194.03,224,0.875,bilinear
-tresnet_m,95.400,4.600,99.150,0.850,31.39,224,0.875,bilinear
+ssl_resnext101_32x16d,95.410,4.590,99.410,0.590,194.03,224,0.875,bilinear
+tresnet_l_448,95.400,4.600,99.300,0.700,55.99,448,0.875,bilinear
+nfnet_l0,95.390,4.610,99.420,0.580,35.07,288,1.000,bicubic
+regnetz_c16,95.390,4.610,99.310,0.690,13.46,320,0.940,bicubic
 resnetv2_50x1_bit_distilled,95.390,4.610,99.430,0.570,25.55,224,0.875,bicubic
-tresnet_l_448,95.390,4.610,99.280,0.720,55.99,448,0.875,bilinear
-pnasnet5large,95.370,4.630,99.130,0.870,86.06,331,0.911,bicubic
+tresnet_m,95.380,4.620,99.150,0.850,31.39,224,0.875,bilinear
+pnasnet5large,95.360,4.640,99.130,0.870,86.06,331,0.911,bicubic
 xcit_tiny_12_p8_384_dist,95.340,4.660,99.340,0.660,6.71,384,1.000,bicubic
-resnetv2_101x1_bitm,95.330,4.670,99.380,0.620,44.54,448,1.000,bilinear
-ssl_resnext101_32x8d,95.320,4.680,99.310,0.690,88.79,224,0.875,bilinear
+ssl_resnext101_32x8d,95.330,4.670,99.310,0.690,88.79,224,0.875,bilinear
+resnetv2_101x1_bitm,95.320,4.680,99.370,0.630,44.54,448,1.000,bilinear
 gc_efficientnetv2_rw_t,95.280,4.720,99.220,0.780,13.68,288,1.000,bicubic
-vit_large_patch32_384,95.250,4.750,99.320,0.680,306.63,384,1.000,bicubic
-cait_xxs36_384,95.240,4.760,99.330,0.670,17.37,384,1.000,bicubic
-resnetrs101,95.230,4.770,99.210,0.790,63.62,288,0.940,bicubic
-swsl_resnet50,95.220,4.780,99.400,0.600,25.56,224,0.875,bilinear
-levit_384,95.200,4.800,99.160,0.840,39.13,224,0.900,bicubic
-resnet51q,95.190,4.810,99.280,0.720,35.70,288,1.000,bilinear
-crossvit_18_dagger_240,95.180,4.820,99.120,0.880,44.27,240,0.875,bicubic
-nasnetalarge,95.170,4.830,99.130,0.870,88.75,331,0.911,bicubic
+resnetrs101,95.250,4.750,99.210,0.790,63.62,288,0.940,bicubic
+vit_large_patch32_384,95.240,4.760,99.320,0.680,306.63,384,1.000,bicubic
+cait_xxs36_384,95.230,4.770,99.320,0.680,17.37,384,1.000,bicubic
+levit_384,95.210,4.790,99.160,0.840,39.13,224,0.900,bicubic
+resnet51q,95.200,4.800,99.280,0.720,35.70,288,1.000,bilinear
+swsl_resnet50,95.200,4.800,99.390,0.610,25.56,224,0.875,bilinear
+crossvit_18_dagger_240,95.190,4.810,99.120,0.880,44.27,240,0.875,bicubic
 ecaresnet101d,95.160,4.840,99.230,0.770,44.57,224,0.875,bicubic
-efficientnet_b3,95.150,4.850,99.210,0.790,12.23,320,1.000,bicubic
-ssl_resnext101_32x4d,95.140,4.860,99.310,0.690,44.18,224,0.875,bilinear
-xcit_medium_24_p16_224,95.130,4.870,98.930,1.070,84.40,224,1.000,bicubic
-tf_efficientnetv2_b3,95.120,4.880,99.190,0.810,14.36,300,0.904,bicubic
-vit_small_r26_s32_224,95.120,4.880,99.220,0.780,36.43,224,0.900,bicubic
-coat_lite_small,95.110,4.890,99.030,0.970,19.84,224,0.900,bicubic
-ecaresnet50t,95.110,4.890,99.290,0.710,25.57,320,0.950,bicubic
+ssl_resnext101_32x4d,95.160,4.840,99.310,0.690,44.18,224,0.875,bilinear
+nasnetalarge,95.150,4.850,99.130,0.870,88.75,331,0.911,bicubic
+efficientnet_b3,95.130,4.870,99.210,0.790,12.23,320,1.000,bicubic
+fbnetv3_g,95.130,4.870,99.200,0.800,16.62,288,0.950,bilinear
+vit_small_r26_s32_224,95.130,4.870,99.220,0.780,36.43,224,0.900,bicubic
+xcit_medium_24_p16_224,95.130,4.870,98.920,1.080,84.40,224,1.000,bicubic
+tf_efficientnetv2_b3,95.120,4.880,99.200,0.800,14.36,300,0.904,bicubic
 resnet61q,95.110,4.890,99.080,0.920,36.85,288,1.000,bicubic
-convit_base,95.100,4.900,99.130,0.870,86.54,224,0.875,bicubic
-tresnet_xl,95.080,4.920,99.250,0.750,78.44,224,0.875,bilinear
-efficientnetv2_rw_t,95.080,4.920,99.220,0.780,13.65,288,1.000,bicubic
-xcit_small_24_p16_224,95.070,4.930,99.060,0.940,47.67,224,1.000,bicubic
-crossvit_base_240,95.060,4.940,98.980,1.020,105.03,240,0.875,bicubic
-crossvit_18_240,95.050,4.950,99.120,0.880,43.27,240,0.875,bicubic
-deit_base_patch16_224,95.020,4.980,98.970,1.030,86.57,224,0.900,bicubic
-visformer_small,94.970,5.030,99.210,0.790,40.22,224,0.900,bicubic
-crossvit_15_dagger_240,94.970,5.030,99.150,0.850,28.21,240,0.875,bicubic
-tf_efficientnet_b3_ap,94.960,5.040,99.110,0.890,12.23,300,0.904,bicubic
+convit_base,95.100,4.900,99.140,0.860,86.54,224,0.875,bicubic
+xcit_small_24_p16_224,95.080,4.920,99.060,0.940,47.67,224,1.000,bicubic
+coat_lite_small,95.080,4.920,99.030,0.970,19.84,224,0.900,bicubic
+crossvit_18_240,95.070,4.930,99.120,0.880,43.27,240,0.875,bicubic
+crossvit_base_240,95.070,4.930,98.980,1.020,105.03,240,0.875,bicubic
+ecaresnet50t,95.070,4.930,99.290,0.710,25.57,320,0.950,bicubic
+efficientnetv2_rw_t,95.070,4.930,99.220,0.780,13.65,288,1.000,bicubic
+tresnet_xl,95.060,4.940,99.260,0.740,78.44,224,0.875,bilinear
+halo2botnet50ts_256,95.010,4.990,99.040,0.960,22.64,256,0.950,bicubic
+deit_base_patch16_224,95.010,4.990,98.980,1.020,86.57,224,0.900,bicubic
+convnext_tiny,94.990,5.010,99.200,0.800,28.59,224,0.875,bicubic
+crossvit_15_dagger_240,94.980,5.020,99.160,0.840,28.21,240,0.875,bicubic
+convmixer_1536_20,94.970,5.030,99.170,0.830,51.63,224,0.960,bicubic
+resnet101,94.970,5.030,99.080,0.920,44.55,224,0.950,bicubic
+tf_efficientnet_b3_ap,94.970,5.030,99.110,0.890,12.23,300,0.904,bicubic
+visformer_small,94.960,5.040,99.210,0.790,40.22,224,0.900,bicubic
 xcit_large_24_p16_224,94.950,5.050,98.830,1.170,189.10,224,1.000,bicubic
-convmixer_1536_20,94.950,5.050,99.170,0.830,51.63,224,0.960,bicubic
+jx_nest_tiny,94.950,5.050,99.100,0.900,17.06,224,0.875,bicubic
 cait_xxs24_384,94.940,5.060,99.130,0.870,12.03,384,1.000,bicubic
-jx_nest_tiny,94.940,5.060,99.100,0.900,17.06,224,0.875,bicubic
-resnetv2_101,94.930,5.070,99.110,0.890,44.54,224,0.950,bicubic
-convit_small,94.920,5.080,99.120,0.880,27.78,224,0.875,bicubic
-gernet_l,94.920,5.080,99.200,0.800,31.08,256,0.875,bilinear
-tf_efficientnet_b3,94.910,5.090,99.100,0.900,12.23,300,0.904,bicubic
-vit_small_patch16_224,94.900,5.100,99.280,0.720,22.05,224,0.900,bicubic
+gernet_l,94.930,5.070,99.200,0.800,31.08,256,0.875,bilinear
+resnetv2_101,94.930,5.070,99.120,0.880,44.54,224,0.950,bicubic
+convit_small,94.920,5.080,99.100,0.900,27.78,224,0.875,bicubic
+tf_efficientnet_b3,94.910,5.090,99.110,0.890,12.23,300,0.904,bicubic
 tresnet_l,94.900,5.100,99.030,0.970,55.99,224,0.875,bilinear
-mixer_b16_224_miil,94.890,5.110,99.080,0.920,59.88,224,0.875,bilinear
-tf_efficientnet_b1_ns,94.880,5.120,99.250,0.750,7.79,240,0.882,bicubic
-xcit_tiny_24_p8_224,94.870,5.130,99.190,0.810,12.11,224,1.000,bicubic
+mixer_b16_224_miil,94.880,5.120,99.080,0.920,59.88,224,0.875,bilinear
+vit_small_patch16_224,94.880,5.120,99.270,0.730,22.05,224,0.900,bicubic
+xcit_tiny_24_p8_224,94.880,5.120,99.190,0.810,12.11,224,1.000,bicubic
 tf_efficientnet_lite4,94.870,5.130,99.090,0.910,13.01,380,0.920,bilinear
-xcit_small_12_p16_224,94.810,5.190,99.060,0.940,26.25,224,1.000,bicubic
-seresnext50_32x4d,94.800,5.200,99.130,0.870,27.56,224,0.875,bicubic
-pit_b_224,94.790,5.210,98.810,1.190,73.76,224,0.900,bicubic
+tf_efficientnet_b1_ns,94.860,5.140,99.250,0.750,7.79,240,0.882,bicubic
+xcit_small_12_p16_224,94.820,5.180,99.060,0.940,26.25,224,1.000,bicubic
+seresnext50_32x4d,94.810,5.190,99.130,0.870,27.56,224,0.875,bicubic
+pit_b_224,94.790,5.210,98.820,1.180,73.76,224,0.900,bicubic
+lamhalobotnet50ts_256,94.780,5.220,98.980,1.020,22.57,256,0.950,bicubic
 twins_svt_small,94.770,5.230,99.080,0.920,24.06,224,0.900,bicubic
-coat_mini,94.760,5.240,98.950,1.050,10.34,224,0.900,bicubic
-resnetv2_50x1_bitm,94.760,5.240,99.180,0.820,25.55,448,1.000,bilinear
+coat_mini,94.770,5.230,98.950,1.050,10.34,224,0.900,bicubic
+resnetv2_50x1_bitm,94.750,5.250,99.180,0.820,25.55,448,1.000,bilinear
 pit_s_distilled_224,94.750,5.250,99.180,0.820,24.04,224,0.900,bicubic
-xcit_tiny_12_p8_224_dist,94.740,5.260,99.180,0.820,6.71,224,1.000,bicubic
 legacy_senet154,94.730,5.270,99.100,0.900,115.09,224,0.875,bilinear
+crossvit_15_240,94.720,5.280,99.080,0.920,27.53,240,0.875,bicubic
+gluon_senet154,94.710,5.290,98.970,1.030,115.09,224,0.875,bicubic
+halonet50ts,94.710,5.290,98.820,1.180,22.73,256,0.940,bicubic
 resnest50d_4s2x40d,94.710,5.290,99.130,0.870,30.42,224,0.875,bicubic
+xcit_tiny_12_p8_224_dist,94.710,5.290,99.180,0.820,6.71,224,1.000,bicubic
+gluon_resnet152_v1s,94.700,5.300,99.060,0.940,60.32,224,0.875,bicubic
 ssl_resnext50_32x4d,94.700,5.300,99.240,0.760,25.03,224,0.875,bilinear
-gluon_senet154,94.700,5.300,98.970,1.030,115.09,224,0.875,bicubic
-gluon_resnet152_v1s,94.690,5.310,99.050,0.950,60.32,224,0.875,bicubic
-regnetz_b,94.690,5.310,99.160,0.840,9.72,288,0.940,bicubic
-crossvit_15_240,94.680,5.320,99.070,0.930,27.53,240,0.875,bicubic
+regnetz_b16,94.680,5.320,99.160,0.840,9.72,288,0.940,bicubic
 efficientnet_el,94.670,5.330,99.130,0.870,10.59,300,0.904,bicubic
-rexnet_200,94.670,5.330,99.090,0.910,16.37,224,0.875,bicubic
-tresnet_m_448,94.670,5.330,99.170,0.830,31.39,448,0.875,bilinear
-gluon_seresnext101_64x4d,94.660,5.340,98.970,1.030,88.23,224,0.875,bicubic
-wide_resnet50_2,94.650,5.350,99.050,0.950,68.88,224,0.875,bicubic
-swin_tiny_patch4_window7_224,94.640,5.360,99.120,0.880,28.29,224,0.900,bicubic
+gluon_seresnext101_64x4d,94.660,5.340,98.980,1.020,88.23,224,0.875,bicubic
+rexnet_200,94.660,5.340,99.090,0.910,16.37,224,0.875,bicubic
+wide_resnet50_2,94.660,5.340,99.050,0.950,68.88,224,0.875,bicubic
+tresnet_m_448,94.650,5.350,99.150,0.850,31.39,448,0.875,bilinear
 resnest50d,94.620,5.380,99.030,0.970,27.48,224,0.875,bilinear
-gcresnet50t,94.610,5.390,98.990,1.010,25.90,256,0.900,bicubic
-lamhalobotnet50ts_256,94.610,5.390,98.610,1.390,22.57,256,0.950,bicubic
-twins_pcpvt_small,94.600,5.400,99.140,0.860,24.11,224,0.900,bicubic
-deit_small_distilled_patch16_224,94.590,5.410,99.090,0.910,22.44,224,0.900,bicubic
-pit_s_224,94.590,5.410,98.930,1.070,23.46,224,0.900,bicubic
-crossvit_small_240,94.590,5.410,99.120,0.880,26.86,240,0.875,bicubic
+swin_tiny_patch4_window7_224,94.620,5.380,99.120,0.880,28.29,224,0.900,bicubic
+gcresnet50t,94.620,5.380,98.980,1.020,25.90,256,0.900,bicubic
+deit_small_distilled_patch16_224,94.600,5.400,99.100,0.900,22.44,224,0.900,bicubic
+twins_pcpvt_small,94.600,5.400,99.150,0.850,24.11,224,0.900,bicubic
+vit_small_patch32_384,94.600,5.400,99.140,0.860,22.92,384,1.000,bicubic
+pit_s_224,94.590,5.410,98.920,1.080,23.46,224,0.900,bicubic
+crossvit_small_240,94.580,5.420,99.120,0.880,26.86,240,0.875,bicubic
 efficientnet_b3_pruned,94.580,5.420,99.070,0.930,9.86,300,0.904,bicubic
+lambda_resnet50ts,94.570,5.430,98.650,1.350,21.54,256,0.950,bicubic
+resnext50_32x4d,94.570,5.430,98.800,1.200,25.03,224,0.950,bicubic
 tnt_s_patch16_224,94.570,5.430,99.180,0.820,23.76,224,0.900,bicubic
-vit_small_patch32_384,94.570,5.430,99.140,0.860,22.92,384,1.000,bicubic
-lambda_resnet50ts,94.560,5.440,98.650,1.350,21.54,256,0.950,bicubic
 repvgg_b3,94.560,5.440,98.910,1.090,123.09,224,0.875,bilinear
-gernet_m,94.550,5.450,98.920,1.080,21.14,224,0.875,bilinear
-halo2botnet50ts_256,94.550,5.450,98.760,1.240,22.64,256,0.950,bicubic
+gernet_m,94.550,5.450,98.930,1.070,21.14,224,0.875,bilinear
 resmlp_36_distilled_224,94.550,5.450,99.160,0.840,44.69,224,0.875,bicubic
-xcit_tiny_12_p16_384_dist,94.540,5.460,99.170,0.830,6.72,384,1.000,bicubic
-sehalonet33ts,94.530,5.470,98.780,1.220,13.69,256,0.940,bicubic
-regnety_320,94.500,5.500,99.170,0.830,145.05,224,0.875,bicubic
-haloregnetz_b,94.500,5.500,98.960,1.040,11.68,224,0.940,bicubic
-repvgg_b3g4,94.490,5.510,99.020,0.980,83.83,224,0.875,bilinear
-gluon_resnet152_v1d,94.460,5.540,99.000,1.000,60.21,224,0.875,bicubic
-ecaresnet101d_pruned,94.440,5.560,99.100,0.900,24.88,224,0.875,bicubic
-gluon_seresnext101_32x4d,94.430,5.570,99.090,0.910,48.96,224,0.875,bicubic
+sehalonet33ts,94.530,5.470,98.760,1.240,13.69,256,0.940,bicubic
+xcit_tiny_12_p16_384_dist,94.530,5.470,99.170,0.830,6.72,384,1.000,bicubic
+regnety_320,94.520,5.480,99.170,0.830,145.05,224,0.875,bicubic
+haloregnetz_b,94.520,5.480,98.960,1.040,11.68,224,0.940,bicubic
+repvgg_b3g4,94.500,5.500,99.020,0.980,83.83,224,0.875,bilinear
+ecaresnet101d_pruned,94.450,5.550,99.100,0.900,24.88,224,0.875,bicubic
+gluon_seresnext101_32x4d,94.450,5.550,99.090,0.910,48.96,224,0.875,bicubic
+gluon_resnet152_v1d,94.440,5.560,99.010,0.990,60.21,224,0.875,bicubic
 convmixer_768_32,94.430,5.570,99.110,0.890,21.11,224,0.960,bicubic
-halonet50ts,94.420,5.580,98.760,1.240,22.73,256,0.940,bicubic
 gcresnext50ts,94.410,5.590,98.990,1.010,15.67,256,0.900,bicubic
-levit_256,94.410,5.590,99.060,0.940,18.89,224,0.900,bicubic
+levit_256,94.400,5.600,99.060,0.940,18.89,224,0.900,bicubic
+resnest50d_1s4x24d,94.390,5.610,99.070,0.930,25.68,224,0.875,bicubic
+vit_base_patch32_224,94.390,5.610,99.060,0.940,88.22,224,0.900,bicubic
+inception_v4,94.380,5.620,98.820,1.180,42.68,299,0.875,bicubic
 nf_resnet50,94.380,5.620,99.070,0.930,25.56,288,0.940,bicubic
-resnest50d_1s4x24d,94.380,5.620,99.060,0.940,25.68,224,0.875,bicubic
-vit_base_patch32_224,94.380,5.620,99.060,0.940,88.22,224,0.900,bicubic
 efficientnet_b2,94.370,5.630,99.050,0.950,9.11,288,1.000,bicubic
-inception_v4,94.370,5.630,98.820,1.180,42.68,299,0.875,bicubic
-xcit_tiny_12_p8_224,94.370,5.630,99.070,0.930,6.71,224,1.000,bicubic
-tf_efficientnet_el,94.350,5.650,99.090,0.910,10.59,300,0.904,bicubic
+tf_efficientnet_el,94.360,5.640,99.100,0.900,10.59,300,0.904,bicubic
+xcit_tiny_12_p8_224,94.360,5.640,99.070,0.930,6.71,224,1.000,bicubic
+gluon_resnext101_64x4d,94.350,5.650,98.880,1.120,83.46,224,0.875,bicubic
 resmlp_24_distilled_224,94.340,5.660,99.090,0.910,30.02,224,0.875,bicubic
-gluon_resnext101_64x4d,94.330,5.670,98.880,1.120,83.46,224,0.875,bicubic
-inception_resnet_v2,94.320,5.680,98.800,1.200,55.84,299,0.897,bicubic
-ssl_resnet50,94.320,5.680,99.160,0.840,25.56,224,0.875,bilinear
-resnetv2_50,94.290,5.710,98.930,1.070,25.55,224,0.950,bicubic
-regnetx_120,94.290,5.710,99.200,0.800,46.11,224,0.875,bicubic
-tf_efficientnet_b2_ap,94.280,5.720,98.950,1.050,9.11,260,0.890,bicubic
-rexnet_150,94.270,5.730,99.090,0.910,9.73,224,0.875,bicubic
+inception_resnet_v2,94.340,5.660,98.800,1.200,55.84,299,0.897,bicubic
+sebotnet33ts_256,94.310,5.690,98.600,1.400,13.70,256,0.940,bicubic
+ssl_resnet50,94.310,5.690,99.150,0.850,25.56,224,0.875,bilinear
 resmlp_big_24_224,94.270,5.730,98.820,1.180,129.14,224,0.875,bicubic
-seresnet33ts,94.260,5.740,98.780,1.220,19.78,256,0.900,bicubic
-mixnet_xl,94.220,5.780,98.810,1.190,11.90,224,0.875,bicubic
-xcit_tiny_24_p16_224_dist,94.220,5.780,98.960,1.040,12.12,224,1.000,bicubic
-ecaresnet50d,94.210,5.790,99.010,0.990,25.58,224,0.875,bicubic
-regnetx_320,94.210,5.790,99.050,0.950,107.81,224,0.875,bicubic
-tf_efficientnet_b2,94.200,5.800,99.030,0.970,9.11,260,0.890,bicubic
-gluon_resnet101_v1s,94.180,5.820,99.020,0.980,44.67,224,0.875,bicubic
-gluon_resnet101_v1d,94.180,5.820,98.950,1.050,44.57,224,0.875,bicubic
+resnetv2_50,94.270,5.730,98.930,1.070,25.55,224,0.950,bicubic
+rexnet_150,94.270,5.730,99.080,0.920,9.73,224,0.875,bicubic
+seresnet33ts,94.270,5.730,98.780,1.220,19.78,256,0.900,bicubic
+tf_efficientnet_b2_ap,94.270,5.730,98.950,1.050,9.11,260,0.890,bicubic
+regnetx_120,94.260,5.740,99.190,0.810,46.11,224,0.875,bicubic
+mixnet_xl,94.230,5.770,98.820,1.180,11.90,224,0.875,bicubic
+regnetx_320,94.220,5.780,99.050,0.950,107.81,224,0.875,bicubic
+tf_efficientnet_b2,94.210,5.790,99.030,0.970,9.11,260,0.890,bicubic
+xcit_tiny_24_p16_224_dist,94.210,5.790,98.960,1.040,12.12,224,1.000,bicubic
+ecaresnet50d,94.200,5.800,99.020,0.980,25.58,224,0.875,bicubic
 dpn92,94.180,5.820,98.930,1.070,37.67,224,0.875,bicubic
-gluon_seresnext50_32x4d,94.180,5.820,98.910,1.090,27.56,224,0.875,bicubic
-legacy_seresnext101_32x4d,94.170,5.830,98.970,1.030,48.96,224,0.875,bilinear
-regnety_064,94.140,5.860,99.030,0.970,30.58,224,0.875,bicubic
+resnet50_gn,94.180,5.820,98.920,1.080,25.56,224,0.940,bicubic
+gluon_seresnext50_32x4d,94.170,5.830,98.910,1.090,27.56,224,0.875,bicubic
+gluon_resnet101_v1s,94.170,5.830,99.010,0.990,44.67,224,0.875,bicubic
+gluon_resnet101_v1d,94.170,5.830,98.940,1.060,44.57,224,0.875,bicubic
 ecaresnetlight,94.140,5.860,98.950,1.050,30.16,224,0.875,bicubic
-ens_adv_inception_resnet_v2,94.140,5.860,98.790,1.210,55.84,299,0.897,bicubic
+ens_adv_inception_resnet_v2,94.130,5.870,98.790,1.210,55.84,299,0.897,bicubic
+regnety_064,94.130,5.870,99.030,0.970,30.58,224,0.875,bicubic
 gluon_resnext101_32x4d,94.120,5.880,98.940,1.060,44.18,224,0.875,bicubic
-tf_efficientnet_lite3,94.110,5.890,98.960,1.040,8.20,300,0.904,bilinear
-cspdarknet53,94.100,5.900,98.980,1.020,27.64,256,0.887,bilinear
-seresnet50,94.080,5.920,98.950,1.050,28.09,224,0.875,bicubic
+legacy_seresnext101_32x4d,94.120,5.880,98.970,1.030,48.96,224,0.875,bilinear
+tf_efficientnet_lite3,94.120,5.880,98.960,1.040,8.20,300,0.904,bilinear
+cspdarknet53,94.090,5.910,98.980,1.020,27.64,256,0.887,bilinear
+efficientnet_el_pruned,94.080,5.920,99.020,0.980,10.59,300,0.904,bicubic
+seresnet50,94.080,5.920,98.960,1.040,28.09,224,0.875,bicubic
 resnet50d,94.070,5.930,98.920,1.080,25.58,224,0.875,bicubic
-regnety_120,94.060,5.940,99.020,0.980,51.82,224,0.875,bicubic
-tf_efficientnetv2_b2,94.060,5.940,98.940,1.060,10.10,260,0.890,bicubic
-efficientnet_el_pruned,94.060,5.940,99.020,0.980,10.59,300,0.904,bicubic
-gluon_xception65,94.040,5.960,99.030,0.970,39.92,299,0.903,bicubic
-resnetrs50,94.030,5.970,98.830,1.170,35.69,224,0.910,bicubic
-hrnet_w48,94.030,5.970,99.030,0.970,77.47,224,0.875,bilinear
-gluon_resnet152_v1b,94.020,5.980,98.750,1.250,60.19,224,0.875,bicubic
-dla102x2,94.010,5.990,99.030,0.970,41.28,224,0.875,bilinear
+tf_efficientnetv2_b2,94.070,5.930,98.930,1.070,10.10,260,0.890,bicubic
+gluon_resnet152_v1b,94.030,5.970,98.740,1.260,60.19,224,0.875,bicubic
+hrnet_w48,94.030,5.970,99.040,0.960,77.47,224,0.875,bilinear
+gluon_xception65,94.020,5.980,99.020,0.980,39.92,299,0.903,bicubic
+resnetrs50,94.020,5.980,98.850,1.150,35.69,224,0.910,bicubic
+regnety_120,94.010,5.990,99.030,0.970,51.82,224,0.875,bicubic
+dla102x2,94.000,6.000,99.030,0.970,41.28,224,0.875,bilinear
 deit_small_patch16_224,93.990,6.010,98.960,1.040,22.05,224,0.900,bicubic
-dpn107,93.960,6.040,98.840,1.160,86.92,224,0.875,bicubic
-resnet50,93.950,6.050,98.470,1.530,25.56,224,0.950,bicubic
+dpn107,93.960,6.040,98.830,1.170,86.92,224,0.875,bicubic
+ecaresnet26t,93.950,6.050,98.920,1.080,16.01,320,0.950,bicubic
 skresnext50_32x4d,93.950,6.050,98.830,1.170,27.48,224,0.875,bicubic
-cait_xxs36_224,93.940,6.060,98.880,1.120,17.30,224,1.000,bicubic
-ecaresnet26t,93.930,6.070,98.930,1.070,16.01,320,0.950,bicubic
-dpn98,93.930,6.070,98.920,1.080,61.57,224,0.875,bicubic
-xception71,93.900,6.100,98.950,1.050,42.34,299,0.903,bicubic
-regnety_080,93.900,6.100,98.990,1.010,39.18,224,0.875,bicubic
-regnetx_160,93.900,6.100,99.080,0.920,54.28,224,0.875,bicubic
-vit_base_patch16_sam_224,93.890,6.110,98.890,1.110,86.57,224,0.900,bicubic
-nf_regnet_b1,93.890,6.110,98.750,1.250,10.22,288,0.900,bicubic
-gluon_resnet152_v1c,93.880,6.120,98.800,1.200,60.21,224,0.875,bicubic
-eca_resnet33ts,93.870,6.130,98.890,1.110,19.68,256,0.900,bicubic
-resnext50_32x4d,93.850,6.150,98.820,1.180,25.03,224,0.875,bicubic
-cspresnet50,93.850,6.150,98.870,1.130,21.62,256,0.887,bilinear
-xcit_tiny_24_p16_224,93.850,6.150,98.770,1.230,12.12,224,1.000,bicubic
-hrnet_w64,93.850,6.150,98.930,1.070,128.06,224,0.875,bilinear
+cait_xxs36_224,93.930,6.070,98.890,1.110,17.30,224,1.000,bicubic
+resnet50,93.930,6.070,98.470,1.530,25.56,224,0.950,bicubic
+dpn98,93.920,6.080,98.920,1.080,61.57,224,0.875,bicubic
+gluon_resnet152_v1c,93.890,6.110,98.800,1.200,60.21,224,0.875,bicubic
+regnetx_160,93.890,6.110,99.090,0.910,54.28,224,0.875,bicubic
+regnety_080,93.890,6.110,99.000,1.000,39.18,224,0.875,bicubic
+vit_base_patch16_224_sam,93.890,6.110,98.890,1.110,86.57,224,0.900,bicubic
+nf_regnet_b1,93.880,6.120,98.740,1.260,10.22,288,0.900,bicubic
+xception71,93.880,6.120,98.950,1.050,42.34,299,0.903,bicubic
+cspresnet50,93.860,6.140,98.860,1.140,21.62,256,0.887,bilinear
+eca_resnet33ts,93.860,6.140,98.890,1.110,19.68,256,0.900,bicubic
+xcit_tiny_24_p16_224,93.850,6.150,98.760,1.240,12.12,224,1.000,bicubic
 ese_vovnet39b,93.850,6.150,98.900,1.100,24.57,224,0.875,bicubic
-gcresnet33ts,93.820,6.180,98.930,1.070,19.88,256,0.900,bicubic
+fbnetv3_d,93.840,6.160,98.910,1.090,10.31,256,0.950,bilinear
+hrnet_w64,93.840,6.160,98.930,1.070,128.06,224,0.875,bilinear
 ecaresnet50d_pruned,93.820,6.180,99.000,1.000,19.94,224,0.875,bicubic
-repvgg_b2g4,93.820,6.180,98.920,1.080,61.76,224,0.875,bilinear
-resnext50d_32x4d,93.800,6.200,98.730,1.270,25.05,224,0.875,bicubic
+gcresnet33ts,93.820,6.180,98.910,1.090,19.88,256,0.900,bicubic
+repvgg_b2g4,93.810,6.190,98.930,1.070,61.76,224,0.875,bilinear
+resnext50d_32x4d,93.810,6.190,98.740,1.260,25.05,224,0.875,bicubic
 efficientnet_b2_pruned,93.800,6.200,98.910,1.090,8.31,260,0.890,bicubic
+dla169,93.800,6.200,98.840,1.160,53.39,224,0.875,bilinear
 regnetx_080,93.790,6.210,98.900,1.100,39.57,224,0.875,bicubic
-dla169,93.780,6.220,98.830,1.170,53.39,224,0.875,bilinear
+cspresnext50,93.780,6.220,98.840,1.160,20.57,224,0.875,bilinear
+gluon_resnet101_v1b,93.770,6.230,98.690,1.310,44.55,224,0.875,bicubic
 resnext101_32x8d,93.770,6.230,98.950,1.050,88.79,224,0.875,bilinear
-gluon_resnet101_v1b,93.770,6.230,98.720,1.280,44.55,224,0.875,bicubic
-cspresnext50,93.770,6.230,98.840,1.160,20.57,224,0.875,bilinear
-dpn131,93.760,6.240,98.850,1.150,79.25,224,0.875,bicubic
-tf_efficientnet_b0_ns,93.760,6.240,98.980,1.020,5.29,224,0.875,bicubic
-efficientnet_em,93.750,6.250,98.920,1.080,6.90,240,0.882,bicubic
-xception65,93.740,6.260,98.870,1.130,39.92,299,0.903,bicubic
+xception65,93.760,6.240,98.860,1.140,39.92,299,0.903,bicubic
+dpn131,93.750,6.250,98.840,1.160,79.25,224,0.875,bicubic
+efficientnet_em,93.740,6.260,98.930,1.070,6.90,240,0.882,bicubic
+tf_efficientnet_b0_ns,93.740,6.260,98.980,1.020,5.29,224,0.875,bicubic
 wide_resnet101_2,93.720,6.280,98.810,1.190,126.89,224,0.875,bilinear
 hrnet_w40,93.710,6.290,98.800,1.200,57.56,224,0.875,bilinear
-levit_192,93.710,6.290,98.790,1.210,10.95,224,0.900,bicubic
-resnetblur50,93.710,6.290,98.800,1.200,25.56,224,0.875,bicubic
+resnetblur50,93.710,6.290,98.810,1.190,25.56,224,0.875,bicubic
 tf_efficientnet_b1,93.710,6.290,98.800,1.200,7.79,240,0.882,bicubic
-tf_efficientnetv2_b1,93.700,6.300,98.810,1.190,8.14,240,0.882,bicubic
-rexnet_130,93.690,6.310,98.720,1.280,7.56,224,0.875,bicubic
-regnetx_040,93.670,6.330,98.950,1.050,22.12,224,0.875,bicubic
-resmlp_36_224,93.670,6.330,98.950,1.050,44.69,224,0.875,bicubic
-gluon_resnext50_32x4d,93.670,6.330,98.700,1.300,25.03,224,0.875,bicubic
-gluon_resnet101_v1c,93.660,6.340,98.760,1.240,44.57,224,0.875,bicubic
-xception,93.650,6.350,98.770,1.230,22.86,299,0.897,bicubic
-regnetx_064,93.650,6.350,99.050,0.950,26.21,224,0.875,bicubic
-tf_efficientnet_b1_ap,93.640,6.360,98.780,1.220,7.79,240,0.882,bicubic
-hrnet_w44,93.620,6.380,98.950,1.050,67.06,224,0.875,bilinear
-resnet33ts,93.620,6.380,98.770,1.230,19.68,256,0.900,bicubic
-halonet26t,93.610,6.390,98.640,1.360,12.48,256,0.950,bicubic
-regnety_040,93.610,6.390,98.960,1.040,20.65,224,0.875,bicubic
-dpn68b,93.600,6.400,98.710,1.290,12.61,224,0.875,bicubic
-gluon_resnet50_v1s,93.590,6.410,98.830,1.170,25.68,224,0.875,bicubic
-gluon_inception_v3,93.590,6.410,98.840,1.160,23.83,299,0.875,bicubic
-res2net50_26w_6s,93.580,6.420,98.740,1.260,37.05,224,0.875,bilinear
-tf_efficientnet_cc_b1_8e,93.580,6.420,98.690,1.310,39.72,240,0.882,bicubic
-repvgg_b2,93.570,6.430,99.070,0.930,89.02,224,0.875,bilinear
+tf_efficientnetv2_b1,93.710,6.290,98.820,1.180,8.14,240,0.882,bicubic
+levit_192,93.710,6.290,98.800,1.200,10.95,224,0.900,bicubic
+gluon_resnet101_v1c,93.690,6.310,98.760,1.240,44.57,224,0.875,bicubic
+regnetx_040,93.680,6.320,98.940,1.060,22.12,224,0.875,bicubic
+rexnet_130,93.680,6.320,98.710,1.290,7.56,224,0.875,bicubic
+resmlp_36_224,93.650,6.350,98.950,1.050,44.69,224,0.875,bicubic
+fbnetv3_b,93.650,6.350,98.910,1.090,8.60,256,0.950,bilinear
+gluon_resnext50_32x4d,93.650,6.350,98.690,1.310,25.03,224,0.875,bicubic
+xception,93.640,6.360,98.760,1.240,22.86,299,0.897,bicubic
+resnet33ts,93.630,6.370,98.760,1.240,19.68,256,0.900,bicubic
+tf_efficientnet_b1_ap,93.630,6.370,98.800,1.200,7.79,240,0.882,bicubic
+dpn68b,93.620,6.380,98.700,1.300,12.61,224,0.875,bicubic
+regnetx_064,93.620,6.380,99.050,0.950,26.21,224,0.875,bicubic
+res2net50_26w_6s,93.610,6.390,98.740,1.260,37.05,224,0.875,bilinear
+hrnet_w44,93.610,6.390,98.960,1.040,67.06,224,0.875,bilinear
+regnety_040,93.610,6.390,98.950,1.050,20.65,224,0.875,bicubic
+halonet26t,93.600,6.400,98.640,1.360,12.48,256,0.950,bicubic
+gluon_resnet50_v1s,93.590,6.410,98.840,1.160,25.68,224,0.875,bicubic
+repvgg_b2,93.590,6.410,99.070,0.930,89.02,224,0.875,bilinear
+dla60_res2next,93.570,6.430,98.790,1.210,17.03,224,0.875,bilinear
 resnet32ts,93.570,6.430,98.750,1.250,17.96,256,0.900,bicubic
-dla60_res2next,93.560,6.440,98.800,1.200,17.03,224,0.875,bilinear
-gluon_resnet50_v1d,93.550,6.450,98.710,1.290,25.58,224,0.875,bicubic
-res2net101_26w_4s,93.520,6.480,98.630,1.370,45.21,224,0.875,bilinear
-dla102x,93.510,6.490,98.850,1.150,26.31,224,0.875,bilinear
+tf_efficientnet_cc_b1_8e,93.570,6.430,98.690,1.310,39.72,240,0.882,bicubic
+eca_halonext26ts,93.560,6.440,98.680,1.320,10.76,256,0.940,bicubic
+gluon_inception_v3,93.540,6.460,98.830,1.170,23.83,299,0.875,bicubic
+gluon_resnet50_v1d,93.530,6.470,98.710,1.290,25.58,224,0.875,bicubic
+dla102x,93.520,6.480,98.850,1.150,26.31,224,0.875,bilinear
+res2net101_26w_4s,93.520,6.480,98.600,1.400,45.21,224,0.875,bilinear
+coat_tiny,93.510,6.490,98.690,1.310,5.50,224,0.900,bicubic
 gmlp_s16_224,93.510,6.490,98.780,1.220,19.42,224,0.875,bicubic
-coat_tiny,93.500,6.500,98.680,1.320,5.50,224,0.900,bicubic
+selecsls60b,93.500,6.500,98.840,1.160,32.77,224,0.875,bicubic
 cait_xxs24_224,93.490,6.510,98.770,1.230,11.96,224,1.000,bicubic
-coat_lite_mini,93.490,6.510,98.780,1.220,11.01,224,0.900,bicubic
-selecsls60b,93.480,6.520,98.840,1.160,32.77,224,0.875,bicubic
-xception41,93.480,6.520,98.760,1.240,26.97,299,0.903,bicubic
-legacy_seresnet152,93.460,6.540,98.850,1.150,66.82,224,0.875,bilinear
-lambda_resnet26rpt_256,93.440,6.560,98.880,1.120,10.99,256,0.940,bicubic
-botnet26t_256,93.440,6.560,98.660,1.340,12.49,256,0.950,bicubic
-res2net50_26w_8s,93.430,6.570,98.670,1.330,48.40,224,0.875,bilinear
-resmlp_24_224,93.430,6.570,98.810,1.190,30.02,224,0.875,bicubic
-vit_tiny_patch16_384,93.430,6.570,98.840,1.160,5.79,384,1.000,bicubic
-hrnet_w30,93.410,6.590,98.830,1.170,37.71,224,0.875,bilinear
-legacy_seresnext50_32x4d,93.410,6.590,98.800,1.200,27.56,224,0.875,bilinear
-repvgg_b1,93.410,6.590,98.780,1.220,57.42,224,0.875,bilinear
-lambda_resnet26t,93.400,6.600,98.760,1.240,10.96,256,0.940,bicubic
-dla60_res2net,93.380,6.620,98.830,1.170,20.85,224,0.875,bilinear
-eca_botnext26ts_256,93.360,6.640,98.690,1.310,10.59,256,0.950,bicubic
-xcit_tiny_12_p16_224_dist,93.340,6.660,98.750,1.250,6.72,224,1.000,bicubic
-dla102,93.290,6.710,98.780,1.220,33.27,224,0.875,bilinear
-legacy_seresnet101,93.290,6.710,98.750,1.250,49.33,224,0.875,bilinear
-mixnet_l,93.290,6.710,98.710,1.290,7.33,224,0.875,bicubic
-regnetx_032,93.270,6.730,98.740,1.260,15.30,224,0.875,bicubic
-resnest26d,93.260,6.740,98.840,1.160,17.07,224,0.875,bilinear
-tv_resnet152,93.260,6.740,98.750,1.250,60.19,224,0.875,bilinear
-xcit_nano_12_p8_384_dist,93.250,6.750,98.850,1.150,3.05,384,1.000,bicubic
-pit_xs_distilled_224,93.230,6.770,98.820,1.180,11.00,224,0.900,bicubic
-tf_inception_v3,93.210,6.790,98.490,1.510,23.83,299,0.875,bicubic
-dla60x,93.210,6.790,98.720,1.280,17.35,224,0.875,bilinear
-tf_efficientnet_em,93.200,6.800,98.680,1.320,6.90,240,0.882,bicubic
+xception41,93.480,6.520,98.750,1.250,26.97,299,0.903,bicubic
+coat_lite_mini,93.460,6.540,98.780,1.220,11.01,224,0.900,bicubic
+legacy_seresnet152,93.440,6.560,98.850,1.150,66.82,224,0.875,bilinear
+resmlp_24_224,93.440,6.560,98.810,1.190,30.02,224,0.875,bicubic
+botnet26t_256,93.430,6.570,98.660,1.340,12.49,256,0.950,bicubic
+lambda_resnet26rpt_256,93.430,6.570,98.880,1.120,10.99,256,0.940,bicubic
+legacy_seresnext50_32x4d,93.430,6.570,98.800,1.200,27.56,224,0.875,bilinear
+vit_tiny_patch16_384,93.430,6.570,98.830,1.170,5.79,384,1.000,bicubic
+res2net50_26w_8s,93.420,6.580,98.690,1.310,48.40,224,0.875,bilinear
+repvgg_b1,93.410,6.590,98.790,1.210,57.42,224,0.875,bilinear
+lambda_resnet26t,93.400,6.600,98.740,1.260,10.96,256,0.940,bicubic
+hrnet_w30,93.390,6.610,98.830,1.170,37.71,224,0.875,bilinear
+dla60_res2net,93.380,6.620,98.860,1.140,20.85,224,0.875,bilinear
+eca_botnext26ts_256,93.360,6.640,98.700,1.300,10.59,256,0.950,bicubic
+xcit_tiny_12_p16_224_dist,93.340,6.660,98.740,1.260,6.72,224,1.000,bicubic
+xcit_nano_12_p8_384_dist,93.280,6.720,98.850,1.150,3.05,384,1.000,bicubic
+mixnet_l,93.270,6.730,98.700,1.300,7.33,224,0.875,bicubic
+legacy_seresnet101,93.270,6.730,98.740,1.260,49.33,224,0.875,bilinear
+dla102,93.260,6.740,98.770,1.230,33.27,224,0.875,bilinear
+regnetx_032,93.250,6.750,98.730,1.270,15.30,224,0.875,bicubic
+pit_xs_distilled_224,93.240,6.760,98.830,1.170,11.00,224,0.900,bicubic
+resnest26d,93.240,6.760,98.850,1.150,17.07,224,0.875,bilinear
+tv_resnet152,93.240,6.760,98.750,1.250,60.19,224,0.875,bilinear
+tf_inception_v3,93.200,6.800,98.480,1.520,23.83,299,0.875,bicubic
+dla60x,93.190,6.810,98.710,1.290,17.35,224,0.875,bilinear
 res2net50_26w_4s,93.180,6.820,98.670,1.330,25.70,224,0.875,bilinear
-eca_halonext26ts,93.140,6.860,98.690,1.310,10.76,256,0.940,bicubic
-res2next50,93.110,6.890,98.660,1.340,24.67,224,0.875,bilinear
-bat_resnext26ts,93.100,6.900,98.720,1.280,10.73,256,0.900,bicubic
+tf_efficientnet_em,93.180,6.820,98.670,1.330,6.90,240,0.882,bicubic
+res2next50,93.140,6.860,98.650,1.350,24.67,224,0.875,bilinear
+bat_resnext26ts,93.100,6.900,98.730,1.270,10.73,256,0.900,bicubic
 tf_efficientnetv2_b0,93.060,6.940,98.700,1.300,7.14,224,0.875,bicubic
-tf_mixnet_l,93.050,6.950,98.540,1.460,7.33,224,0.875,bicubic
-levit_128,93.040,6.960,98.690,1.310,9.21,224,0.900,bicubic
-repvgg_b1g4,93.040,6.960,98.820,1.180,39.97,224,0.875,bilinear
-regnety_016,93.030,6.970,98.690,1.310,11.20,224,0.875,bicubic
-efficientnet_b1,93.020,6.980,98.710,1.290,7.79,256,1.000,bicubic
-res2net50_14w_8s,93.020,6.980,98.700,1.300,25.06,224,0.875,bilinear
-selecsls60,93.000,7.000,98.830,1.170,30.67,224,0.875,bicubic
-adv_inception_v3,92.990,7.010,98.480,1.520,23.83,299,0.875,bicubic
+levit_128,93.040,6.960,98.680,1.320,9.21,224,0.900,bicubic
+tf_mixnet_l,93.040,6.960,98.540,1.460,7.33,224,0.875,bicubic
+repvgg_b1g4,93.030,6.970,98.820,1.180,39.97,224,0.875,bilinear
+res2net50_14w_8s,93.030,6.970,98.700,1.300,25.06,224,0.875,bilinear
+efficientnet_b1,93.030,6.970,98.710,1.290,7.79,256,1.000,bicubic
+selecsls60,93.010,6.990,98.830,1.170,30.67,224,0.875,bicubic
+adv_inception_v3,93.010,6.990,98.490,1.510,23.83,299,0.875,bicubic
+regnety_016,93.000,7.000,98.680,1.320,11.20,224,0.875,bicubic
 hardcorenas_f,92.980,7.020,98.620,1.380,8.20,224,0.875,bilinear
-hardcorenas_e,92.960,7.040,98.570,1.430,8.07,224,0.875,bilinear
-efficientnet_b1_pruned,92.960,7.040,98.520,1.480,6.33,240,0.882,bicubic
+efficientnet_b1_pruned,92.970,7.030,98.520,1.480,6.33,240,0.882,bicubic
+hardcorenas_e,92.950,7.050,98.570,1.430,8.07,224,0.875,bilinear
 hrnet_w32,92.950,7.050,98.840,1.160,41.23,224,0.875,bilinear
-efficientnet_es,92.940,7.060,98.690,1.310,5.44,224,0.875,bicubic
 gluon_resnet50_v1c,92.920,7.080,98.710,1.290,25.58,224,0.875,bicubic
-tv_resnext50_32x4d,92.910,7.090,98.730,1.270,25.03,224,0.875,bilinear
-pit_xs_224,92.900,7.100,98.790,1.210,10.62,224,0.900,bicubic
-inception_v3,92.900,7.100,98.320,1.680,23.83,299,0.875,bicubic
+efficientnet_es,92.920,7.080,98.690,1.310,5.44,224,0.875,bicubic
+pit_xs_224,92.910,7.090,98.770,1.230,10.62,224,0.900,bicubic
+tv_resnext50_32x4d,92.910,7.090,98.720,1.280,25.03,224,0.875,bilinear
+inception_v3,92.900,7.100,98.330,1.670,23.83,299,0.875,bicubic
+densenet161,92.890,7.110,98.810,1.190,28.68,224,0.875,bicubic
 tv_resnet101,92.880,7.120,98.660,1.340,44.55,224,0.875,bilinear
-densenet161,92.880,7.120,98.810,1.190,28.68,224,0.875,bicubic
-resmlp_12_distilled_224,92.870,7.130,98.630,1.370,15.35,224,0.875,bicubic
-tf_efficientnet_cc_b0_8e,92.850,7.150,98.460,1.540,24.01,224,0.875,bicubic
+resmlp_12_distilled_224,92.870,7.130,98.620,1.380,15.35,224,0.875,bicubic
+tf_efficientnet_cc_b0_8e,92.870,7.130,98.460,1.540,24.01,224,0.875,bicubic
+coat_lite_tiny,92.860,7.140,98.630,1.370,5.72,224,0.900,bicubic
 rexnet_100,92.840,7.160,98.620,1.380,4.80,224,0.875,bicubic
 tf_efficientnet_cc_b0_4e,92.840,7.160,98.440,1.560,13.31,224,0.875,bicubic
-coat_lite_tiny,92.830,7.170,98.640,1.360,5.72,224,0.900,bicubic
-res2net50_48w_2s,92.800,7.200,98.470,1.530,25.29,224,0.875,bilinear
-seresnext26ts,92.790,7.210,98.600,1.400,10.39,256,0.900,bicubic
-seresnext26t_32x4d,92.770,7.230,98.550,1.450,16.81,224,0.875,bicubic
-hrnet_w18,92.750,7.250,98.650,1.350,21.30,224,0.875,bilinear
+seresnext26t_32x4d,92.820,7.180,98.560,1.440,16.81,224,0.875,bicubic
+seresnext26ts,92.810,7.190,98.600,1.400,10.39,256,0.900,bicubic
+res2net50_48w_2s,92.790,7.210,98.480,1.520,25.29,224,0.875,bilinear
+tinynet_a,92.790,7.210,98.560,1.440,6.19,192,0.875,bicubic
 crossvit_9_dagger_240,92.750,7.250,98.510,1.490,8.78,240,0.875,bicubic
-dla60,92.690,7.310,98.630,1.370,22.04,224,0.875,bilinear
-densenet201,92.690,7.310,98.660,1.340,20.01,224,0.875,bicubic
-resnet26t,92.680,7.320,98.600,1.400,16.01,256,0.940,bicubic
-gmixer_24_224,92.670,7.330,98.260,1.740,24.72,224,0.875,bicubic
-legacy_seresnet50,92.670,7.330,98.660,1.340,28.09,224,0.875,bilinear
-repvgg_a2,92.660,7.340,98.530,1.470,28.21,224,0.875,bilinear
-resnet34d,92.640,7.360,98.440,1.560,21.82,224,0.875,bicubic
-mobilenetv2_120d,92.610,7.390,98.510,1.490,5.83,224,0.875,bicubic
-tf_efficientnet_b0_ap,92.600,7.400,98.370,1.630,5.29,224,0.875,bicubic
+hrnet_w18,92.750,7.250,98.660,1.340,21.30,224,0.875,bilinear
+densenet201,92.690,7.310,98.650,1.350,20.01,224,0.875,bicubic
+gmixer_24_224,92.680,7.320,98.280,1.720,24.72,224,0.875,bicubic
+repvgg_a2,92.680,7.320,98.520,1.480,28.21,224,0.875,bilinear
+dla60,92.670,7.330,98.630,1.370,22.04,224,0.875,bilinear
+legacy_seresnet50,92.670,7.330,98.650,1.350,28.09,224,0.875,bilinear
+resnet26t,92.670,7.330,98.580,1.420,16.01,256,0.940,bicubic
+resnet34d,92.640,7.360,98.420,1.580,21.82,224,0.875,bicubic
+mobilenetv2_120d,92.610,7.390,98.500,1.500,5.83,224,0.875,bicubic
+tf_efficientnet_b0_ap,92.610,7.390,98.370,1.630,5.29,224,0.875,bicubic
 hardcorenas_d,92.600,7.400,98.430,1.570,7.50,224,0.875,bilinear
-legacy_seresnext26_32x4d,92.590,7.410,98.420,1.580,16.79,224,0.875,bicubic
-tf_efficientnet_lite2,92.570,7.430,98.550,1.450,6.09,260,0.890,bicubic
-regnetx_016,92.560,7.440,98.550,1.450,9.19,224,0.875,bicubic
-skresnet34,92.560,7.440,98.510,1.490,22.28,224,0.875,bicubic
-gluon_resnet50_v1b,92.550,7.450,98.550,1.450,25.56,224,0.875,bicubic
+legacy_seresnext26_32x4d,92.580,7.420,98.410,1.590,16.79,224,0.875,bicubic
+tf_efficientnet_lite2,92.580,7.420,98.550,1.450,6.09,260,0.890,bicubic
+skresnet34,92.570,7.430,98.520,1.480,22.28,224,0.875,bicubic
+gluon_resnet50_v1b,92.560,7.440,98.550,1.450,25.56,224,0.875,bicubic
+regnetx_016,92.540,7.460,98.550,1.450,9.19,224,0.875,bicubic
+efficientnet_b0,92.480,7.520,98.680,1.320,5.29,224,0.875,bicubic
 selecsls42b,92.480,7.520,98.440,1.560,32.46,224,0.875,bicubic
-gcresnext26ts,92.470,7.530,98.500,1.500,10.48,256,0.900,bicubic
-efficientnet_b0,92.470,7.530,98.680,1.320,5.29,224,0.875,bicubic
+gcresnext26ts,92.470,7.530,98.490,1.510,10.48,256,0.900,bicubic
 xcit_tiny_12_p16_224,92.460,7.540,98.630,1.370,6.72,224,1.000,bicubic
-gernet_s,92.440,7.560,98.490,1.510,8.17,224,0.875,bilinear
+gernet_s,92.440,7.560,98.500,1.500,8.17,224,0.875,bilinear
 seresnext26d_32x4d,92.430,7.570,98.540,1.460,16.81,224,0.875,bicubic
-eca_resnext26ts,92.420,7.580,98.610,1.390,10.30,256,0.900,bicubic
-xcit_nano_12_p8_224_dist,92.410,7.590,98.510,1.490,3.05,224,1.000,bicubic
+xcit_nano_12_p8_224_dist,92.420,7.580,98.530,1.470,3.05,224,1.000,bicubic
+eca_resnext26ts,92.410,7.590,98.620,1.380,10.30,256,0.900,bicubic
+densenetblur121d,92.400,7.600,98.410,1.590,8.00,224,0.875,bicubic
 tf_efficientnet_b0,92.400,7.600,98.470,1.530,5.29,224,0.875,bicubic
-densenetblur121d,92.400,7.600,98.420,1.580,8.00,224,0.875,bicubic
 convmixer_1024_20_ks9_p14,92.350,7.650,98.420,1.580,24.38,224,0.960,bicubic
-hardcorenas_c,92.330,7.670,98.340,1.660,5.52,224,0.875,bilinear
-tf_efficientnet_lite1,92.320,7.680,98.490,1.510,5.42,240,0.882,bicubic
+hardcorenas_c,92.330,7.670,98.350,1.650,5.52,224,0.875,bilinear
+tf_efficientnet_lite1,92.310,7.690,98.490,1.510,5.42,240,0.882,bicubic
 densenet169,92.290,7.710,98.590,1.410,14.15,224,0.875,bicubic
-mixnet_m,92.260,7.740,98.370,1.630,5.01,224,0.875,bicubic
-dpn68,92.260,7.740,98.600,1.400,12.61,224,0.875,bicubic
-resnet26d,92.250,7.750,98.470,1.530,16.01,224,0.875,bicubic
+mixnet_m,92.270,7.730,98.350,1.650,5.01,224,0.875,bicubic
+dpn68,92.260,7.740,98.610,1.390,12.61,224,0.875,bicubic
 mobilenetv3_large_100_miil,92.250,7.750,98.250,1.750,5.48,224,0.875,bilinear
-resnext26ts,92.210,7.790,98.280,1.720,10.30,256,0.900,bicubic
-tf_mixnet_m,92.170,7.830,98.420,1.580,5.01,224,0.875,bicubic
-vit_small_patch32_224,92.150,7.850,98.510,1.490,22.88,224,0.900,bicubic
-resmlp_12_224,92.130,7.870,98.570,1.430,15.35,224,0.875,bicubic
-tv_resnet50,92.110,7.890,98.420,1.580,25.56,224,0.875,bilinear
-xcit_nano_12_p16_384_dist,92.100,7.900,98.520,1.480,3.05,384,1.000,bicubic
+resnet26d,92.250,7.750,98.450,1.550,16.01,224,0.875,bicubic
+resnext26ts,92.220,7.780,98.250,1.750,10.30,256,0.900,bicubic
+tf_mixnet_m,92.200,7.800,98.420,1.580,5.01,224,0.875,bicubic
+vit_small_patch32_224,92.160,7.840,98.510,1.490,22.88,224,0.900,bicubic
+resmlp_12_224,92.120,7.880,98.570,1.430,15.35,224,0.875,bicubic
+tv_resnet50,92.120,7.880,98.420,1.580,25.56,224,0.875,bilinear
+xcit_nano_12_p16_384_dist,92.110,7.890,98.520,1.480,3.05,384,1.000,bicubic
 tf_efficientnet_es,92.100,7.900,98.430,1.570,5.44,224,0.875,bicubic
-mobilenetv2_140,92.030,7.970,98.250,1.750,6.11,224,0.875,bicubic
-ese_vovnet19b_dw,92.020,7.980,98.520,1.480,6.54,224,0.875,bicubic
-hardcorenas_b,91.970,8.030,98.400,1.600,5.18,224,0.875,bilinear
+mobilenetv2_140,92.050,7.950,98.250,1.750,6.11,224,0.875,bicubic
+ese_vovnet19b_dw,92.010,7.990,98.510,1.490,6.54,224,0.875,bicubic
 densenet121,91.940,8.060,98.280,1.720,7.98,224,0.875,bicubic
-vit_tiny_patch16_224,91.930,8.070,98.330,1.670,5.72,224,0.900,bicubic
-regnety_008,91.890,8.110,98.420,1.580,6.26,224,0.875,bicubic
+hardcorenas_b,91.930,8.070,98.400,1.600,5.18,224,0.875,bilinear
+regnety_008,91.910,8.090,98.420,1.580,6.26,224,0.875,bicubic
+vit_tiny_patch16_224,91.910,8.090,98.340,1.660,5.72,224,0.900,bicubic
 mixnet_s,91.780,8.220,98.300,1.700,4.13,224,0.875,bicubic
-vit_tiny_r_s16_p8_384,91.720,8.280,98.430,1.570,6.36,384,1.000,bicubic
-efficientnet_es_pruned,91.710,8.290,98.400,1.600,5.44,224,0.875,bicubic
-tf_mixnet_s,91.690,8.310,98.240,1.760,4.13,224,0.875,bicubic
-repvgg_b0,91.670,8.330,98.450,1.550,15.82,224,0.875,bilinear
-semnasnet_100,91.660,8.340,98.260,1.740,3.89,224,0.875,bicubic
-hardcorenas_a,91.610,8.390,98.170,1.830,5.26,224,0.875,bilinear
-regnety_006,91.580,8.420,98.430,1.570,6.06,224,0.875,bicubic
-mobilenetv3_rw,91.550,8.450,98.270,1.730,5.48,224,0.875,bicubic
+vit_tiny_r_s16_p8_384,91.730,8.270,98.430,1.570,6.36,384,1.000,bicubic
+efficientnet_es_pruned,91.710,8.290,98.410,1.590,5.44,224,0.875,bicubic
+repvgg_b0,91.680,8.320,98.450,1.550,15.82,224,0.875,bilinear
+tf_mixnet_s,91.680,8.320,98.240,1.760,4.13,224,0.875,bicubic
+semnasnet_100,91.660,8.340,98.270,1.730,3.89,224,0.875,bicubic
+hardcorenas_a,91.620,8.380,98.170,1.830,5.26,224,0.875,bilinear
+regnety_006,91.560,8.440,98.430,1.570,6.06,224,0.875,bicubic
+mobilenetv3_rw,91.540,8.460,98.270,1.730,5.48,224,0.875,bicubic
 levit_128s,91.500,8.500,98.400,1.600,7.78,224,0.900,bicubic
 legacy_seresnet34,91.480,8.520,98.200,1.800,21.96,224,0.875,bilinear
-mobilenetv3_large_100,91.480,8.520,98.330,1.670,5.48,224,0.875,bicubic
-resnet26,91.460,8.540,98.270,1.730,16.00,224,0.875,bicubic
-tf_mobilenetv3_large_100,91.410,8.590,98.250,1.750,5.48,224,0.875,bilinear
-tv_densenet121,91.400,8.600,98.250,1.750,7.98,224,0.875,bicubic
-mobilenetv2_110d,91.320,8.680,98.180,1.820,4.52,224,0.875,bicubic
-tf_efficientnet_lite0,91.280,8.720,98.090,1.910,4.65,224,0.875,bicubic
-fbnetc_100,91.260,8.740,97.820,2.180,5.57,224,0.875,bilinear
-dla34,91.250,8.750,98.180,1.820,15.74,224,0.875,bilinear
+mobilenetv3_large_100,91.480,8.520,98.320,1.680,5.48,224,0.875,bicubic
+resnet26,91.450,8.550,98.270,1.730,16.00,224,0.875,bicubic
+tf_mobilenetv3_large_100,91.420,8.580,98.260,1.740,5.48,224,0.875,bilinear
+tv_densenet121,91.410,8.590,98.250,1.750,7.98,224,0.875,bicubic
+mobilenetv2_110d,91.350,8.650,98.190,1.810,4.52,224,0.875,bicubic
+tf_efficientnet_lite0,91.300,8.700,98.090,1.910,4.65,224,0.875,bicubic
+fbnetc_100,91.260,8.740,97.830,2.170,5.57,224,0.875,bilinear
 efficientnet_lite0,91.250,8.750,98.250,1.750,4.65,224,0.875,bicubic
-mnasnet_100,91.200,8.800,98.040,1.960,4.38,224,0.875,bicubic
-resnet34,91.190,8.810,98.230,1.770,21.80,224,0.875,bilinear
-regnetx_008,91.190,8.810,98.370,1.630,7.26,224,0.875,bicubic
-hrnet_w18_small_v2,91.170,8.830,98.350,1.650,15.60,224,0.875,bilinear
-resnest14d,91.150,8.850,98.350,1.650,10.61,224,0.875,bilinear
-mixer_b16_224,91.130,8.870,97.410,2.590,59.88,224,0.875,bicubic
-xcit_nano_12_p8_224,91.100,8.900,98.240,1.760,3.05,224,1.000,bicubic
-swsl_resnet18,91.090,8.910,98.210,1.790,11.69,224,0.875,bilinear
-gluon_resnet34_v1b,91.090,8.910,98.180,1.820,21.80,224,0.875,bicubic
-deit_tiny_distilled_patch16_224,91.090,8.910,98.270,1.730,5.91,224,0.900,bicubic
-crossvit_9_240,91.070,8.930,98.310,1.690,8.55,240,0.875,bicubic
-vgg19_bn,90.990,9.010,98.120,1.880,143.68,224,0.875,bilinear
-pit_ti_distilled_224,90.900,9.100,98.230,1.770,5.10,224,0.900,bicubic
-regnetx_006,90.750,9.250,98.100,1.900,6.20,224,0.875,bicubic
-regnety_004,90.750,9.250,98.080,1.920,4.34,224,0.875,bicubic
+dla34,91.230,8.770,98.180,1.820,15.74,224,0.875,bilinear
+mnasnet_100,91.200,8.800,98.050,1.950,4.38,224,0.875,bicubic
+resnet34,91.200,8.800,98.240,1.760,21.80,224,0.875,bilinear
+hrnet_w18_small_v2,91.170,8.830,98.340,1.660,15.60,224,0.875,bilinear
+regnetx_008,91.160,8.840,98.380,1.620,7.26,224,0.875,bicubic
+mixer_b16_224,91.140,8.860,97.400,2.600,59.88,224,0.875,bicubic
+tinynet_b,91.130,8.870,98.070,1.930,3.73,188,0.875,bicubic
+resnest14d,91.120,8.880,98.330,1.670,10.61,224,0.875,bilinear
+xcit_nano_12_p8_224,91.120,8.880,98.240,1.760,3.05,224,1.000,bicubic
+deit_tiny_distilled_patch16_224,91.110,8.890,98.270,1.730,5.91,224,0.900,bicubic
+gluon_resnet34_v1b,91.100,8.900,98.180,1.820,21.80,224,0.875,bicubic
+swsl_resnet18,91.080,8.920,98.210,1.790,11.69,224,0.875,bilinear
+crossvit_9_240,91.050,8.950,98.310,1.690,8.55,240,0.875,bicubic
+vgg19_bn,90.990,9.010,98.110,1.890,143.68,224,0.875,bilinear
+pit_ti_distilled_224,90.900,9.100,98.220,1.780,5.10,224,0.900,bicubic
+regnetx_006,90.770,9.230,98.100,1.900,6.20,224,0.875,bicubic
+regnety_004,90.770,9.230,98.080,1.920,4.34,224,0.875,bicubic
 ssl_resnet18,90.700,9.300,98.030,1.970,11.69,224,0.875,bilinear
-spnasnet_100,90.610,9.390,97.950,2.050,4.42,224,0.875,bilinear
-convit_tiny,90.550,9.450,98.210,1.790,5.71,224,0.875,bicubic
-crossvit_tiny_240,90.530,9.470,97.950,2.050,7.01,240,0.875,bicubic
-vgg16_bn,90.520,9.480,97.990,2.010,138.37,224,0.875,bilinear
-ghostnet_100,90.440,9.560,97.840,2.160,5.18,224,0.875,bilinear
-pit_ti_224,90.440,9.560,98.020,1.980,4.85,224,0.900,bicubic
-tf_mobilenetv3_large_075,90.310,9.690,97.880,2.120,3.99,224,0.875,bilinear
-tv_resnet34,90.310,9.690,97.970,2.030,21.80,224,0.875,bilinear
-xcit_nano_12_p16_224_dist,90.190,9.810,97.760,2.240,3.05,224,1.000,bicubic
+spnasnet_100,90.600,9.400,97.950,2.050,4.42,224,0.875,bilinear
+convit_tiny,90.550,9.450,98.220,1.780,5.71,224,0.875,bicubic
+crossvit_tiny_240,90.540,9.460,97.950,2.050,7.01,240,0.875,bicubic
+vgg16_bn,90.540,9.460,97.990,2.010,138.37,224,0.875,bilinear
+ghostnet_100,90.440,9.560,97.830,2.170,5.18,224,0.875,bilinear
+pit_ti_224,90.440,9.560,98.010,1.990,4.85,224,0.900,bicubic
+tf_mobilenetv3_large_075,90.320,9.680,97.870,2.130,3.99,224,0.875,bilinear
+tv_resnet34,90.290,9.710,97.980,2.020,21.80,224,0.875,bilinear
+semnasnet_075,90.200,9.800,97.970,2.030,2.91,224,0.875,bicubic
 skresnet18,90.170,9.830,97.780,2.220,11.96,224,0.875,bicubic
-resnet18d,89.990,10.010,97.840,2.160,11.71,224,0.875,bicubic
-hrnet_w18_small,89.900,10.100,97.900,2.100,13.19,224,0.875,bilinear
-vit_base_patch32_sam_224,89.870,10.130,97.600,2.400,88.22,224,0.900,bicubic
-mobilenetv2_100,89.840,10.160,97.840,2.160,3.50,224,0.875,bicubic
-vgg19,89.690,10.310,97.550,2.450,143.67,224,0.875,bilinear
-deit_tiny_patch16_224,89.600,10.400,97.960,2.040,5.72,224,0.900,bicubic
-regnetx_004,89.450,10.550,97.780,2.220,5.16,224,0.875,bicubic
-vit_tiny_r_s16_p8_224,89.400,10.600,97.690,2.310,6.34,224,0.900,bicubic
-vgg16,89.380,10.620,97.520,2.480,138.36,224,0.875,bilinear
-legacy_seresnet18,89.250,10.750,97.690,2.310,11.78,224,0.875,bicubic
-vgg13_bn,89.210,10.790,97.540,2.460,133.05,224,0.875,bilinear
-tf_mobilenetv3_large_minimal_100,89.160,10.840,97.320,2.680,3.92,224,0.875,bilinear
-xcit_nano_12_p16_224,88.960,11.040,97.410,2.590,3.05,224,1.000,bicubic
-gluon_resnet18_v1b,88.670,11.330,97.110,2.890,11.69,224,0.875,bicubic
-vgg11_bn,88.390,11.610,97.280,2.720,132.87,224,0.875,bilinear
-regnety_002,88.210,11.790,97.420,2.580,3.16,224,0.875,bicubic
-resnet18,88.160,11.840,97.120,2.880,11.69,224,0.875,bilinear
-vgg13,87.550,12.450,97.120,2.880,133.05,224,0.875,bilinear
-regnetx_002,87.360,12.640,96.990,3.010,2.68,224,0.875,bicubic
-vgg11,87.330,12.670,97.110,2.890,132.86,224,0.875,bilinear
-dla60x_c,87.080,12.920,97.140,2.860,1.32,224,0.875,bilinear
-mixer_l16_224,86.960,13.040,94.030,5.970,208.20,224,0.875,bicubic
-tf_mobilenetv3_small_100,85.990,14.010,96.410,3.590,2.54,224,0.875,bilinear
-dla46x_c,85.470,14.530,96.450,3.550,1.07,224,0.875,bilinear
-dla46_c,84.710,15.290,96.210,3.790,1.30,224,0.875,bilinear
+xcit_nano_12_p16_224_dist,90.170,9.830,97.760,2.240,3.05,224,1.000,bicubic
+resnet18d,89.990,10.010,97.830,2.170,11.71,224,0.875,bicubic
+hrnet_w18_small,89.880,10.120,97.900,2.100,13.19,224,0.875,bilinear
+vit_base_patch32_224_sam,89.860,10.140,97.600,2.400,88.22,224,0.900,bicubic
+mobilenetv2_100,89.830,10.170,97.830,2.170,3.50,224,0.875,bicubic
+vgg19,89.680,10.320,97.550,2.450,143.67,224,0.875,bilinear
+deit_tiny_patch16_224,89.620,10.380,97.960,2.040,5.72,224,0.900,bicubic
+regnetx_004,89.470,10.530,97.770,2.230,5.16,224,0.875,bicubic
+vgg16,89.360,10.640,97.520,2.480,138.36,224,0.875,bilinear
+vit_tiny_r_s16_p8_224,89.350,10.650,97.700,2.300,6.34,224,0.900,bicubic
+legacy_seresnet18,89.260,10.740,97.680,2.320,11.78,224,0.875,bicubic
+vgg13_bn,89.200,10.800,97.520,2.480,133.05,224,0.875,bilinear
+tf_mobilenetv3_large_minimal_100,89.170,10.830,97.320,2.680,3.92,224,0.875,bilinear
+xcit_nano_12_p16_224,88.970,11.030,97.390,2.610,3.05,224,1.000,bicubic
+lcnet_100,88.950,11.050,97.380,2.620,2.95,224,0.875,bicubic
+gluon_resnet18_v1b,88.660,11.340,97.100,2.900,11.69,224,0.875,bicubic
+tinynet_c,88.420,11.580,97.260,2.740,2.46,184,0.875,bicubic
+vgg11_bn,88.390,11.610,97.270,2.730,132.87,224,0.875,bilinear
+regnety_002,88.190,11.810,97.420,2.580,3.16,224,0.875,bicubic
+resnet18,88.150,11.850,97.120,2.880,11.69,224,0.875,bilinear
+vgg13,87.570,12.430,97.120,2.880,133.05,224,0.875,bilinear
+regnetx_002,87.380,12.620,96.990,3.010,2.68,224,0.875,bicubic
+vgg11,87.340,12.660,97.110,2.890,132.86,224,0.875,bilinear
+dla60x_c,87.110,12.890,97.140,2.860,1.32,224,0.875,bilinear
+mixer_l16_224,86.970,13.030,94.040,5.960,208.20,224,0.875,bicubic
+lcnet_075,86.940,13.060,96.530,3.470,2.36,224,0.875,bicubic
+tf_mobilenetv3_small_100,85.970,14.030,96.410,3.590,2.54,224,0.875,bilinear
+dla46x_c,85.480,14.520,96.440,3.560,1.07,224,0.875,bilinear
+tinynet_d,85.430,14.570,96.020,3.980,2.34,152,0.875,bicubic
+mobilenetv2_050,84.990,15.010,95.620,4.380,1.97,224,0.875,bicubic
+mnasnet_small,84.920,15.080,95.930,4.070,2.03,224,0.875,bicubic
+dla46_c,84.670,15.330,96.200,3.800,1.30,224,0.875,bilinear
 tf_mobilenetv3_small_075,84.530,15.470,95.890,4.110,2.04,224,0.875,bilinear
+lcnet_050,83.000,17.000,95.020,4.980,1.88,224,0.875,bicubic
 tf_mobilenetv3_small_minimal_100,82.680,17.320,95.010,4.990,2.04,224,0.875,bilinear
+tinynet_e,79.800,20.200,93.980,6.020,2.04,106,0.875,bicubic

tests/test_models.py

@@ -28,14 +28,16 @@ NON_STD_FILTERS = [
 NUM_NON_STD = len(NON_STD_FILTERS)
 
 # exclude models that cause specific test failures
-if 'GITHUB_ACTIONS' in os.environ:  # and 'Linux' in platform.system():
+if 'GITHUB_ACTIONS' in os.environ:
     # GitHub Linux runner is slower and hits memory limits sooner than MacOS, exclude bigger models
     EXCLUDE_FILTERS = [
         '*efficientnet_l2*', '*resnext101_32x48d', '*in21k', '*152x4_bitm', '*101x3_bitm', '*50x3_bitm',
         '*nfnet_f3*', '*nfnet_f4*', '*nfnet_f5*', '*nfnet_f6*', '*nfnet_f7*', '*efficientnetv2_xl*',
-        '*resnetrs350*', '*resnetrs420*', 'xcit_large_24_p8*']
+        '*resnetrs350*', '*resnetrs420*', 'xcit_large_24_p8*', 'vit_huge*', 'vit_gi*']
+    NON_STD_EXCLUDE_FILTERS = ['vit_huge*', 'vit_gi*']
 else:
     EXCLUDE_FILTERS = []
+    NON_STD_EXCLUDE_FILTERS = ['vit_gi*']
 
 TARGET_FWD_SIZE = MAX_FWD_SIZE = 384
 TARGET_BWD_SIZE = 128
@@ -168,11 +170,12 @@ def test_model_default_cfgs(model_name, batch_size):
                 assert outputs.shape[-1] == pool_size[-1] and outputs.shape[-2] == pool_size[-2]
 
     # check classifier name matches default_cfg
-    classifier = cfg['classifier']
-    if not isinstance(classifier, (tuple, list)):
-        classifier = classifier,
-    for c in classifier:
-        assert c + ".weight" in state_dict.keys(), f'{c} not in model params'
+    if cfg.get('num_classes', None):
+        classifier = cfg['classifier']
+        if not isinstance(classifier, (tuple, list)):
+            classifier = classifier,
+        for c in classifier:
+            assert c + ".weight" in state_dict.keys(), f'{c} not in model params'
 
     # check first conv(s) names match default_cfg
     first_conv = cfg['first_conv']
@@ -184,7 +187,7 @@ def test_model_default_cfgs(model_name, batch_size):
 
 
 @pytest.mark.timeout(300)
-@pytest.mark.parametrize('model_name', list_models(filter=NON_STD_FILTERS))
+@pytest.mark.parametrize('model_name', list_models(filter=NON_STD_FILTERS, exclude_filters=NON_STD_EXCLUDE_FILTERS))
 @pytest.mark.parametrize('batch_size', [1])
 def test_model_default_cfgs_non_std(model_name, batch_size):
     """Run a single forward pass with each model"""
@@ -220,11 +223,12 @@ def test_model_default_cfgs_non_std(model_name, batch_size):
     assert outputs.shape[1] == model.num_features
 
     # check classifier name matches default_cfg
-    classifier = cfg['classifier']
-    if not isinstance(classifier, (tuple, list)):
-        classifier = classifier,
-    for c in classifier:
-        assert c + ".weight" in state_dict.keys(), f'{c} not in model params'
+    if cfg.get('num_classes', None):
+        classifier = cfg['classifier']
+        if not isinstance(classifier, (tuple, list)):
+            classifier = classifier,
+        for c in classifier:
+            assert c + ".weight" in state_dict.keys(), f'{c} not in model params'
 
     # check first conv(s) names match default_cfg
     first_conv = cfg['first_conv']
@@ -255,7 +259,7 @@ if 'GITHUB_ACTIONS' not in os.environ:
 EXCLUDE_JIT_FILTERS = [
     '*iabn*', 'tresnet*',  # models using inplace abn unlikely to ever be scriptable
     'dla*', 'hrnet*', 'ghostnet*',  # hopefully fix at some point
-    'vit_large_*', 'vit_huge_*',
+    'vit_large_*', 'vit_huge_*', 'vit_gi*',
 ]
 
 
@@ -334,7 +338,7 @@ def _create_fx_model(model, train=False):
     return fx_model
 
 
-EXCLUDE_FX_FILTERS = []
+EXCLUDE_FX_FILTERS = ['vit_gi*']
 # not enough memory to run fx on more models than other tests
 if 'GITHUB_ACTIONS' in os.environ:
     EXCLUDE_FX_FILTERS += [

timm/data/auto_augment.py

@@ -15,7 +15,7 @@ Papers:
     RandAugment: Practical automated data augmentation... - https://arxiv.org/abs/1909.13719
     AugMix: A Simple Data Processing Method to Improve Robustness and Uncertainty - https://arxiv.org/abs/1912.02781
 
-Hacked together by / Copyright 2020 Ross Wightman
+Hacked together by / Copyright 2019, Ross Wightman
 """
 import random
 import math

timm/data/dataset.py

@@ -1,6 +1,6 @@
 """ Quick n Simple Image Folder, Tarfile based DataSet
 
-Hacked together by / Copyright 2020 Ross Wightman
+Hacked together by / Copyright 2019, Ross Wightman
 """
 import torch.utils.data as data
 import os

timm/data/dataset_factory.py

@@ -1,3 +1,7 @@
+""" Dataset Factory
+
+Hacked together by / Copyright 2021, Ross Wightman
+"""
 import os
 
 from torchvision.datasets import CIFAR100, CIFAR10, MNIST, QMNIST, KMNIST, FashionMNIST, ImageNet, ImageFolder

timm/data/distributed_sampler.py

@@ -108,11 +108,11 @@ class RepeatAugSampler(Sampler):
             indices = torch.arange(start=0, end=len(self.dataset))
 
         # produce repeats e.g. [0, 0, 0, 1, 1, 1, 2, 2, 2....]
-        indices = torch.repeat_interleave(indices, repeats=self.num_repeats, dim=0)
+        indices = torch.repeat_interleave(indices, repeats=self.num_repeats, dim=0).tolist()
         # add extra samples to make it evenly divisible
         padding_size = self.total_size - len(indices)
         if padding_size > 0:
-            indices = torch.cat([indices, indices[:padding_size]], dim=0)
+            indices += indices[:padding_size]
         assert len(indices) == self.total_size
 
         # subsample per rank

timm/data/loader.py

@@ -3,7 +3,7 @@
 Prefetcher and Fast Collate inspired by NVIDIA APEX example at
 https://github.com/NVIDIA/apex/commit/d5e2bb4bdeedd27b1dfaf5bb2b24d6c000dee9be#diff-cf86c282ff7fba81fad27a559379d5bf
 
-Hacked together by / Copyright 2021 Ross Wightman
+Hacked together by / Copyright 2019, Ross Wightman
 """
 import random
 from functools import partial

timm/data/mixup.py

@@ -8,7 +8,7 @@ CutMix: Regularization Strategy to Train Strong Classifiers with Localizable Fea
 Code Reference:
 CutMix: https://github.com/clovaai/CutMix-PyTorch
 
-Hacked together by / Copyright 2020 Ross Wightman
+Hacked together by / Copyright 2019, Ross Wightman
 """
 import numpy as np
 import torch

timm/data/random_erasing.py

@@ -3,7 +3,7 @@
 Originally inspired by impl at https://github.com/zhunzhong07/Random-Erasing, Apache 2.0
 Copyright Zhun Zhong & Liang Zheng
 
-Hacked together by / Copyright 2020 Ross Wightman
+Hacked together by / Copyright 2019, Ross Wightman
 """
 import random
 import math

timm/data/transforms_factory.py

@@ -1,7 +1,7 @@
 """ Transforms Factory
 Factory methods for building image transforms for use with TIMM (PyTorch Image Models)
 
-Hacked together by / Copyright 2020 Ross Wightman
+Hacked together by / Copyright 2019, Ross Wightman
 """
 import math
 

timm/models/__init__.py

@@ -5,6 +5,7 @@ from .cait import *
 from .coat import *
 from .convit import *
 from .convmixer import *
+from .convnext import *
 from .crossvit import *
 from .cspnet import *
 from .densenet import *

timm/models/cait.py

@@ -4,6 +4,7 @@ Paper: 'Going deeper with Image Transformers' - https://arxiv.org/abs/2103.17239
 
 Original code and weights from https://github.com/facebookresearch/deit, copyright below
 
+Modifications and additions for timm hacked together by / Copyright 2021, Ross Wightman
 """
 # Copyright (c) 2015-present, Facebook, Inc.
 # All rights reserved.

timm/models/convit.py

@@ -9,6 +9,8 @@
 
 Paper link: https://arxiv.org/abs/2103.10697
 Original code: https://github.com/facebookresearch/convit, original copyright below
+
+Modifications and additions for timm hacked together by / Copyright 2021, Ross Wightman
 """
 # Copyright (c) 2015-present, Facebook, Inc.
 # All rights reserved.

timm/models/convnext.py

@@ -0,0 +1,427 @@
+""" ConvNeXt
+
+Paper: `A ConvNet for the 2020s` - https://arxiv.org/pdf/2201.03545.pdf
+
+Original code and weights from https://github.com/facebookresearch/ConvNeXt, original copyright below
+
+Modifications and additions for timm hacked together by / Copyright 2022, Ross Wightman
+"""
+# Copyright (c) Meta Platforms, Inc. and affiliates.
+# All rights reserved.
+# This source code is licensed under the MIT license
+from collections import OrderedDict
+from functools import partial
+
+import torch
+import torch.nn as nn
+import torch.nn.functional as F
+
+from timm.data import IMAGENET_DEFAULT_MEAN, IMAGENET_DEFAULT_STD
+from .fx_features import register_notrace_module
+from .helpers import named_apply, build_model_with_cfg
+from .layers import trunc_normal_, ClassifierHead, SelectAdaptivePool2d, DropPath, ConvMlp, Mlp
+from .registry import register_model
+
+
+__all__ = ['ConvNeXt']  # model_registry will add each entrypoint fn to this
+
+
+def _cfg(url='', **kwargs):
+    return {
+        'url': url,
+        'num_classes': 1000, 'input_size': (3, 224, 224), 'pool_size': (7, 7),
+        'crop_pct': 0.875, 'interpolation': 'bicubic',
+        'mean': IMAGENET_DEFAULT_MEAN, 'std': IMAGENET_DEFAULT_STD,
+        'first_conv': 'stem.0', 'classifier': 'head.fc',
+        **kwargs
+    }
+
+
+default_cfgs = dict(
+    convnext_tiny=_cfg(url="https://dl.fbaipublicfiles.com/convnext/convnext_tiny_1k_224_ema.pth"),
+    convnext_small=_cfg(url="https://dl.fbaipublicfiles.com/convnext/convnext_small_1k_224_ema.pth"),
+    convnext_base=_cfg(url="https://dl.fbaipublicfiles.com/convnext/convnext_base_1k_224_ema.pth"),
+    convnext_large=_cfg(url="https://dl.fbaipublicfiles.com/convnext/convnext_large_1k_224_ema.pth"),
+
+    convnext_tiny_hnf=_cfg(url=''),
+
+    convnext_base_in22ft1k=_cfg(
+        url='https://dl.fbaipublicfiles.com/convnext/convnext_base_22k_1k_224.pth'),
+    convnext_large_in22ft1k=_cfg(
+        url='https://dl.fbaipublicfiles.com/convnext/convnext_large_22k_1k_224.pth'),
+    convnext_xlarge_in22ft1k=_cfg(
+        url='https://dl.fbaipublicfiles.com/convnext/convnext_xlarge_22k_1k_224_ema.pth'),
+
+    convnext_base_384_in22ft1k=_cfg(
+        url='https://dl.fbaipublicfiles.com/convnext/convnext_base_22k_1k_384.pth',
+        input_size=(3, 384, 384), pool_size=(12, 12), crop_pct=1.0),
+    convnext_large_384_in22ft1k=_cfg(
+        url='https://dl.fbaipublicfiles.com/convnext/convnext_large_22k_1k_384.pth',
+        input_size=(3, 384, 384), pool_size=(12, 12), crop_pct=1.0),
+    convnext_xlarge_384_in22ft1k=_cfg(
+        url='https://dl.fbaipublicfiles.com/convnext/convnext_xlarge_22k_1k_384_ema.pth',
+        input_size=(3, 384, 384), pool_size=(12, 12), crop_pct=1.0),
+
+    convnext_base_in22k=_cfg(
+        url="https://dl.fbaipublicfiles.com/convnext/convnext_base_22k_224.pth", num_classes=21841),
+    convnext_large_in22k=_cfg(
+        url="https://dl.fbaipublicfiles.com/convnext/convnext_large_22k_224.pth", num_classes=21841),
+    convnext_xlarge_in22k=_cfg(
+        url="https://dl.fbaipublicfiles.com/convnext/convnext_xlarge_22k_224.pth", num_classes=21841),
+)
+
+
+def _is_contiguous(tensor: torch.Tensor) -> bool:
+    # jit is oh so lovely :/
+    # if torch.jit.is_tracing():
+    #     return True
+    if torch.jit.is_scripting():
+        return tensor.is_contiguous()
+    else:
+        return tensor.is_contiguous(memory_format=torch.contiguous_format)
+
+
+@register_notrace_module
+class LayerNorm2d(nn.LayerNorm):
+    r""" LayerNorm for channels_first tensors with 2d spatial dimensions (ie N, C, H, W).
+    """
+
+    def __init__(self, normalized_shape, eps=1e-6):
+        super().__init__(normalized_shape, eps=eps)
+
+    def forward(self, x) -> torch.Tensor:
+        if _is_contiguous(x):
+            return F.layer_norm(
+                x.permute(0, 2, 3, 1), self.normalized_shape, self.weight, self.bias, self.eps).permute(0, 3, 1, 2)
+        else:
+            s, u = torch.var_mean(x, dim=1, unbiased=False, keepdim=True)
+            x = (x - u) * torch.rsqrt(s + self.eps)
+            x = x * self.weight[:, None, None] + self.bias[:, None, None]
+            return x
+
+
+class ConvNeXtBlock(nn.Module):
+    """ ConvNeXt Block
+    There are two equivalent implementations:
+      (1) DwConv -> LayerNorm (channels_first) -> 1x1 Conv -> GELU -> 1x1 Conv; all in (N, C, H, W)
+      (2) DwConv -> Permute to (N, H, W, C); LayerNorm (channels_last) -> Linear -> GELU -> Linear; Permute back
+
+    Unlike the official impl, this one allows choice of 1 or 2, 1x1 conv can be faster with appropriate
+    choice of LayerNorm impl, however as model size increases the tradeoffs appear to change and nn.Linear
+    is a better choice. This was observed with PyTorch 1.10 on 3090 GPU, it could change over time & w/ different HW.
+
+    Args:
+        dim (int): Number of input channels.
+        drop_path (float): Stochastic depth rate. Default: 0.0
+        ls_init_value (float): Init value for Layer Scale. Default: 1e-6.
+    """
+
+    def __init__(self, dim, drop_path=0., ls_init_value=1e-6, conv_mlp=False, mlp_ratio=4, norm_layer=None):
+        super().__init__()
+        if not norm_layer:
+            norm_layer = partial(LayerNorm2d, eps=1e-6) if conv_mlp else partial(nn.LayerNorm, eps=1e-6)
+        mlp_layer = ConvMlp if conv_mlp else Mlp
+        self.use_conv_mlp = conv_mlp
+        self.conv_dw = nn.Conv2d(dim, dim, kernel_size=7, padding=3, groups=dim)  # depthwise conv
+        self.norm = norm_layer(dim)
+        self.mlp = mlp_layer(dim, int(mlp_ratio * dim), act_layer=nn.GELU)
+        self.gamma = nn.Parameter(ls_init_value * torch.ones(dim)) if ls_init_value > 0 else None
+        self.drop_path = DropPath(drop_path) if drop_path > 0. else nn.Identity()
+
+    def forward(self, x):
+        shortcut = x
+        x = self.conv_dw(x)
+        if self.use_conv_mlp:
+            x = self.norm(x)
+            x = self.mlp(x)
+        else:
+            x = x.permute(0, 2, 3, 1)
+            x = self.norm(x)
+            x = self.mlp(x)
+            x = x.permute(0, 3, 1, 2)
+        if self.gamma is not None:
+            x = x.mul(self.gamma.reshape(1, -1, 1, 1))
+        x = self.drop_path(x) + shortcut
+        return x
+
+
+class ConvNeXtStage(nn.Module):
+
+    def __init__(
+            self, in_chs, out_chs, stride=2, depth=2, dp_rates=None, ls_init_value=1.0, conv_mlp=False,
+            norm_layer=None, cl_norm_layer=None, cross_stage=False):
+        super().__init__()
+
+        if in_chs != out_chs or stride > 1:
+            self.downsample = nn.Sequential(
+                norm_layer(in_chs),
+                nn.Conv2d(in_chs, out_chs, kernel_size=stride, stride=stride),
+            )
+        else:
+            self.downsample = nn.Identity()
+
+        dp_rates = dp_rates or [0.] * depth
+        self.blocks = nn.Sequential(*[ConvNeXtBlock(
+            dim=out_chs, drop_path=dp_rates[j], ls_init_value=ls_init_value, conv_mlp=conv_mlp,
+            norm_layer=norm_layer if conv_mlp else cl_norm_layer)
+            for j in range(depth)]
+        )
+
+    def forward(self, x):
+        x = self.downsample(x)
+        x = self.blocks(x)
+        return x
+
+
+class ConvNeXt(nn.Module):
+    r""" ConvNeXt
+        A PyTorch impl of : `A ConvNet for the 2020s`  - https://arxiv.org/pdf/2201.03545.pdf
+
+    Args:
+        in_chans (int): Number of input image channels. Default: 3
+        num_classes (int): Number of classes for classification head. Default: 1000
+        depths (tuple(int)): Number of blocks at each stage. Default: [3, 3, 9, 3]
+        dims (tuple(int)): Feature dimension at each stage. Default: [96, 192, 384, 768]
+        drop_rate (float): Head dropout rate
+        drop_path_rate (float): Stochastic depth rate. Default: 0.
+        ls_init_value (float): Init value for Layer Scale. Default: 1e-6.
+        head_init_scale (float): Init scaling value for classifier weights and biases. Default: 1.
+    """
+
+    def __init__(
+            self, in_chans=3, num_classes=1000, global_pool='avg', output_stride=32, patch_size=4,
+            depths=(3, 3, 9, 3), dims=(96, 192, 384, 768),  ls_init_value=1e-6, conv_mlp=False,
+            head_init_scale=1., head_norm_first=False, norm_layer=None, drop_rate=0., drop_path_rate=0.,
+    ):
+        super().__init__()
+        assert output_stride == 32
+        if norm_layer is None:
+            norm_layer = partial(LayerNorm2d, eps=1e-6)
+            cl_norm_layer = norm_layer if conv_mlp else partial(nn.LayerNorm, eps=1e-6)
+        else:
+            assert conv_mlp,\
+                'If a norm_layer is specified, conv MLP must be used so all norm expect rank-4, channels-first input'
+            cl_norm_layer = norm_layer
+
+        self.num_classes = num_classes
+        self.drop_rate = drop_rate
+        self.feature_info = []
+
+        # NOTE: this stem is a minimal form of ViT PatchEmbed, as used in SwinTransformer w/ patch_size = 4
+        self.stem = nn.Sequential(
+            nn.Conv2d(in_chans, dims[0], kernel_size=patch_size, stride=patch_size),
+            norm_layer(dims[0])
+        )
+
+        self.stages = nn.Sequential()
+        dp_rates = [x.tolist() for x in torch.linspace(0, drop_path_rate, sum(depths)).split(depths)]
+        curr_stride = patch_size
+        prev_chs = dims[0]
+        stages = []
+        # 4 feature resolution stages, each consisting of multiple residual blocks
+        for i in range(4):
+            stride = 2 if i > 0 else 1
+            # FIXME support dilation / output_stride
+            curr_stride *= stride
+            out_chs = dims[i]
+            stages.append(ConvNeXtStage(
+                prev_chs, out_chs, stride=stride,
+                depth=depths[i], dp_rates=dp_rates[i], ls_init_value=ls_init_value, conv_mlp=conv_mlp,
+                norm_layer=norm_layer, cl_norm_layer=cl_norm_layer)
+            )
+            prev_chs = out_chs
+            # NOTE feature_info use currently assumes stage 0 == stride 1, rest are stride 2
+            self.feature_info += [dict(num_chs=prev_chs, reduction=curr_stride, module=f'stages.{i}')]
+        self.stages = nn.Sequential(*stages)
+
+        self.num_features = prev_chs
+        if head_norm_first:
+            # norm -> global pool -> fc ordering, like most other nets (not compat with FB weights)
+            self.norm_pre = norm_layer(self.num_features)  # final norm layer, before pooling
+            self.head = ClassifierHead(self.num_features, num_classes, pool_type=global_pool, drop_rate=drop_rate)
+        else:
+            # pool -> norm -> fc, the default ConvNeXt ordering (pretrained FB weights)
+            self.norm_pre = nn.Identity()
+            self.head = nn.Sequential(OrderedDict([
+                ('global_pool', SelectAdaptivePool2d(pool_type=global_pool)),
+                ('norm', norm_layer(self.num_features)),
+                ('flatten', nn.Flatten(1) if global_pool else nn.Identity()),
+                ('drop', nn.Dropout(self.drop_rate)),
+                ('fc', nn.Linear(self.num_features, num_classes) if num_classes > 0 else nn.Identity())
+            ]))
+
+        named_apply(partial(_init_weights, head_init_scale=head_init_scale), self)
+
+    def get_classifier(self):
+        return self.head.fc
+
+    def reset_classifier(self, num_classes=0, global_pool='avg'):
+        if isinstance(self.head, ClassifierHead):
+            # norm -> global pool -> fc
+            self.head = ClassifierHead(
+                self.num_features, num_classes, pool_type=global_pool, drop_rate=self.drop_rate)
+        else:
+            # pool -> norm -> fc
+            self.head = nn.Sequential(OrderedDict([
+                ('global_pool', SelectAdaptivePool2d(pool_type=global_pool)),
+                ('norm', self.head.norm),
+                ('flatten', nn.Flatten(1) if global_pool else nn.Identity()),
+                ('drop', nn.Dropout(self.drop_rate)),
+                ('fc', nn.Linear(self.num_features, num_classes) if num_classes > 0 else nn.Identity())
+            ]))
+
+    def forward_features(self, x):
+        x = self.stem(x)
+        x = self.stages(x)
+        x = self.norm_pre(x)
+        return x
+
+    def forward(self, x):
+        x = self.forward_features(x)
+        x = self.head(x)
+        return x
+
+
+def _init_weights(module, name=None, head_init_scale=1.0):
+    if isinstance(module, nn.Conv2d):
+        trunc_normal_(module.weight, std=.02)
+        nn.init.constant_(module.bias, 0)
+    elif isinstance(module, nn.Linear):
+        trunc_normal_(module.weight, std=.02)
+        nn.init.constant_(module.bias, 0)
+        if name and 'head.' in name:
+            module.weight.data.mul_(head_init_scale)
+            module.bias.data.mul_(head_init_scale)
+
+
+def checkpoint_filter_fn(state_dict, model):
+    """ Remap FB checkpoints -> timm """
+    if 'model' in state_dict:
+        state_dict = state_dict['model']
+    out_dict = {}
+    import re
+    for k, v in state_dict.items():
+        k = k.replace('downsample_layers.0.', 'stem.')
+        k = re.sub(r'stages.([0-9]+).([0-9]+)', r'stages.\1.blocks.\2', k)
+        k = re.sub(r'downsample_layers.([0-9]+).([0-9]+)', r'stages.\1.downsample.\2', k)
+        k = k.replace('dwconv', 'conv_dw')
+        k = k.replace('pwconv', 'mlp.fc')
+        k = k.replace('head.', 'head.fc.')
+        if k.startswith('norm.'):
+            k = k.replace('norm', 'head.norm')
+        if v.ndim == 2 and 'head' not in k:
+            model_shape = model.state_dict()[k].shape
+            v = v.reshape(model_shape)
+        out_dict[k] = v
+    return out_dict
+
+
+def _create_convnext(variant, pretrained=False, **kwargs):
+    model = build_model_with_cfg(
+        ConvNeXt, variant, pretrained,
+        default_cfg=default_cfgs[variant],
+        pretrained_filter_fn=checkpoint_filter_fn,
+        feature_cfg=dict(out_indices=(0, 1, 2, 3), flatten_sequential=True),
+        **kwargs)
+    return model
+
+
+@register_model
+def convnext_tiny(pretrained=False, **kwargs):
+    model_args = dict(depths=(3, 3, 9, 3), dims=(96, 192, 384, 768), **kwargs)
+    model = _create_convnext('convnext_tiny', pretrained=pretrained, **model_args)
+    return model
+
+
+@register_model
+def convnext_tiny_hnf(pretrained=False, **kwargs):
+    model_args = dict(depths=(3, 3, 9, 3), dims=(96, 192, 384, 768), head_norm_first=True, **kwargs)
+    model = _create_convnext('convnext_tiny_hnf', pretrained=pretrained, **model_args)
+    return model
+
+
+@register_model
+def convnext_small(pretrained=False, **kwargs):
+    model_args = dict(depths=[3, 3, 27, 3], dims=[96, 192, 384, 768], **kwargs)
+    model = _create_convnext('convnext_small', pretrained=pretrained, **model_args)
+    return model
+
+
+@register_model
+def convnext_base(pretrained=False, **kwargs):
+    model_args = dict(depths=[3, 3, 27, 3], dims=[128, 256, 512, 1024], **kwargs)
+    model = _create_convnext('convnext_base', pretrained=pretrained, **model_args)
+    return model
+
+
+@register_model
+def convnext_large(pretrained=False, **kwargs):
+    model_args = dict(depths=[3, 3, 27, 3], dims=[192, 384, 768, 1536], **kwargs)
+    model = _create_convnext('convnext_large', pretrained=pretrained, **model_args)
+    return model
+
+
+@register_model
+def convnext_base_in22ft1k(pretrained=False, **kwargs):
+    model_args = dict(depths=[3, 3, 27, 3], dims=[128, 256, 512, 1024], **kwargs)
+    model = _create_convnext('convnext_base_in22ft1k', pretrained=pretrained, **model_args)
+    return model
+
+
+@register_model
+def convnext_large_in22ft1k(pretrained=False, **kwargs):
+    model_args = dict(depths=[3, 3, 27, 3], dims=[192, 384, 768, 1536], **kwargs)
+    model = _create_convnext('convnext_large_in22ft1k', pretrained=pretrained, **model_args)
+    return model
+
+
+@register_model
+def convnext_xlarge_in22ft1k(pretrained=False, **kwargs):
+    model_args = dict(depths=[3, 3, 27, 3], dims=[256, 512, 1024, 2048], **kwargs)
+    model = _create_convnext('convnext_xlarge_in22ft1k', pretrained=pretrained, **model_args)
+    return model
+
+
+@register_model
+def convnext_base_384_in22ft1k(pretrained=False, **kwargs):
+    model_args = dict(depths=[3, 3, 27, 3], dims=[128, 256, 512, 1024], **kwargs)
+    model = _create_convnext('convnext_base_384_in22ft1k', pretrained=pretrained, **model_args)
+    return model
+
+
+@register_model
+def convnext_large_384_in22ft1k(pretrained=False, **kwargs):
+    model_args = dict(depths=[3, 3, 27, 3], dims=[192, 384, 768, 1536], **kwargs)
+    model = _create_convnext('convnext_large_384_in22ft1k', pretrained=pretrained, **model_args)
+    return model
+
+
+@register_model
+def convnext_xlarge_384_in22ft1k(pretrained=False, **kwargs):
+    model_args = dict(depths=[3, 3, 27, 3], dims=[256, 512, 1024, 2048], **kwargs)
+    model = _create_convnext('convnext_xlarge_384_in22ft1k', pretrained=pretrained, **model_args)
+    return model
+
+
+@register_model
+def convnext_base_in22k(pretrained=False, **kwargs):
+    model_args = dict(depths=[3, 3, 27, 3], dims=[128, 256, 512, 1024], **kwargs)
+    model = _create_convnext('convnext_base_in22k', pretrained=pretrained, **model_args)
+    return model
+
+
+@register_model
+def convnext_large_in22k(pretrained=False, **kwargs):
+    model_args = dict(depths=[3, 3, 27, 3], dims=[192, 384, 768, 1536], **kwargs)
+    model = _create_convnext('convnext_large_in22k', pretrained=pretrained, **model_args)
+    return model
+
+
+@register_model
+def convnext_xlarge_in22k(pretrained=False, **kwargs):
+    model_args = dict(depths=[3, 3, 27, 3], dims=[256, 512, 1024, 2048], **kwargs)
+    model = _create_convnext('convnext_xlarge_in22k', pretrained=pretrained, **model_args)
+    return model
+
+
+

timm/models/crossvit.py

@@ -12,6 +12,8 @@ Paper link: https://arxiv.org/abs/2103.14899
 Original code: https://github.com/IBM/CrossViT/blob/main/models/crossvit.py
 
 NOTE: model names have been renamed from originals to represent actual input res all *_224 -> *_240 and *_384 -> *_408
+
+Modifications and additions for timm hacked together by / Copyright 2021, Ross Wightman
 """
 
 # Copyright IBM All Rights Reserved.

timm/models/cspnet.py

@@ -411,9 +411,12 @@ class CspNet(nn.Module):
 
 def _create_cspnet(variant, pretrained=False, **kwargs):
     cfg_variant = variant.split('_')[0]
+    # NOTE: DarkNet is one of few models with stride==1 features w/ 6 out_indices [0..5]
+    out_indices = kwargs.pop('out_indices', (0, 1, 2, 3, 4, 5) if 'darknet' in variant else (0, 1, 2, 3, 4))
     return build_model_with_cfg(
         CspNet, variant, pretrained,
-        feature_cfg=dict(flatten_sequential=True), model_cfg=model_cfgs[cfg_variant],
+        model_cfg=model_cfgs[cfg_variant],
+        feature_cfg=dict(flatten_sequential=True, out_indices=out_indices),
         **kwargs)
 
 

timm/models/efficientnet.py

@@ -33,7 +33,7 @@ The majority of the above models (EfficientNet*, MixNet, MnasNet) and original w
 by Mingxing Tan, Quoc Le, and other members of their Google Brain team. Thanks for consistently releasing
 the models and weights open source!
 
-Hacked together by / Copyright 2021 Ross Wightman
+Hacked together by / Copyright 2019, Ross Wightman
 """
 from functools import partial
 from typing import List
@@ -73,16 +73,20 @@ default_cfgs = {
     'mnasnet_140': _cfg(url=''),
 
     'semnasnet_050': _cfg(url=''),
-    'semnasnet_075': _cfg(url=''),
+    'semnasnet_075': _cfg(
+        url='https://github.com/rwightman/pytorch-image-models/releases/download/v0.1-weights/semnasnet_075-18710866.pth'),
     'semnasnet_100': _cfg(
         url='https://github.com/rwightman/pytorch-image-models/releases/download/v0.1-weights/mnasnet_a1-d9418771.pth'),
     'semnasnet_140': _cfg(url=''),
-    'mnasnet_small': _cfg(url=''),
+    'mnasnet_small': _cfg(
+        url='https://github.com/rwightman/pytorch-image-models/releases/download/v0.1-weights/mnasnet_small_lamb-aff75073.pth'),
 
     'mobilenetv2_035': _cfg(
         url=''),
     'mobilenetv2_050': _cfg(
-        url=''),
+        url='https://github.com/rwightman/pytorch-image-models/releases/download/v0.1-weights/mobilenetv2_050-3d30d450.pth',
+        interpolation='bicubic',
+    ),
     'mobilenetv2_075': _cfg(
         url=''),
     'mobilenetv2_100': _cfg(
@@ -726,7 +730,7 @@ def _gen_mobilenet_v2(
     round_chs_fn = partial(round_channels, multiplier=channel_multiplier)
     model_kwargs = dict(
         block_args=decode_arch_def(arch_def, depth_multiplier=depth_multiplier, fix_first_last=fix_stem_head),
-        num_features=1280 if fix_stem_head else round_chs_fn(1280),
+        num_features=1280 if fix_stem_head else max(1280, round_chs_fn(1280)),
         stem_size=32,
         fix_stem=fix_stem_head,
         round_chs_fn=round_chs_fn,
@@ -1474,7 +1478,7 @@ def efficientnet_b0_g16_evos(pretrained=False, **kwargs):
     """ EfficientNet-B0 w/ group 16 conv + EvoNorm"""
     model = _gen_efficientnet(
         'efficientnet_b0_g16_evos', group_size=16, channel_divisor=16,
-        pretrained=pretrained, **kwargs) #norm_layer=partial(EvoNorm2dS0, group_size=16), 
+        pretrained=pretrained, **kwargs) #norm_layer=partial(EvoNorm2dS0, group_size=16),
     return model
 
 

timm/models/efficientnet_blocks.py

@@ -1,6 +1,6 @@
 """ EfficientNet, MobileNetV3, etc Blocks
 
-Hacked together by / Copyright 2020 Ross Wightman
+Hacked together by / Copyright 2019, Ross Wightman
 """
 import math
 

timm/models/efficientnet_builder.py

@@ -3,7 +3,7 @@
 Assembles EfficieNet and related network feature blocks from string definitions.
 Handles stride, dilation calculations, and selects feature extraction points.
 
-Hacked together by / Copyright 2020 Ross Wightman
+Hacked together by / Copyright 2019, Ross Wightman
 """
 
 import logging

timm/models/helpers.py

@@ -281,8 +281,8 @@ def load_pretrained(
         if num_classes != pretrained_cfg['num_classes']:
             for classifier_name in classifiers:
                 # completely discard fully connected if model num_classes doesn't match pretrained weights
-                del state_dict[classifier_name + '.weight']
-                del state_dict[classifier_name + '.bias']
+                state_dict.pop(classifier_name + '.weight', None)
+                state_dict.pop(classifier_name + '.bias', None)
             strict = False
         elif label_offset > 0:
             for classifier_name in classifiers:

timm/models/layers/__init__.py

@@ -23,7 +23,7 @@ from .helpers import to_ntuple, to_2tuple, to_3tuple, to_4tuple, make_divisible
 from .inplace_abn import InplaceAbn
 from .linear import Linear
 from .mixed_conv2d import MixedConv2d
-from .mlp import Mlp, GluMlp, GatedMlp
+from .mlp import Mlp, GluMlp, GatedMlp, ConvMlp
 from .non_local_attn import NonLocalAttn, BatNonLocalAttn
 from .norm import GroupNorm, LayerNorm2d
 from .norm_act import BatchNormAct2d, GroupNormAct

timm/models/layers/drop.py

@@ -157,7 +157,6 @@ def drop_path(x, drop_prob: float = 0., training: bool = False, scale_by_keep: b
 class DropPath(nn.Module):
     """Drop paths (Stochastic Depth) per sample  (when applied in main path of residual blocks).
     """
-
     def __init__(self, drop_prob=None, scale_by_keep=True):
         super(DropPath, self).__init__()
         self.drop_prob = drop_prob

timm/models/levit.py

@@ -14,7 +14,7 @@ Adapted from official impl at https://github.com/facebookresearch/LeViT, origina
 
 This version combines both conv/linear models and fixes torchscript compatibility.
 
-Modifications by/coyright Copyright 2021 Ross Wightman
+Modifications and additions for timm hacked together by / Copyright 2021, Ross Wightman
 """
 
 # Copyright (c) 2015-present, Facebook, Inc.

timm/models/mobilenetv3.py

@@ -1,11 +1,10 @@
-
 """ MobileNet V3
 
 A PyTorch impl of MobileNet-V3, compatible with TF weights from official impl.
 
 Paper: Searching for MobileNetV3 - https://arxiv.org/abs/1905.02244
 
-Hacked together by / Copyright 2021 Ross Wightman
+Hacked together by / Copyright 2019, Ross Wightman
 """
 from functools import partial
 from typing import List
@@ -48,9 +47,15 @@ default_cfgs = {
         interpolation='bilinear', mean=(0, 0, 0), std=(1, 1, 1),
         url='https://miil-public-eu.oss-eu-central-1.aliyuncs.com/model-zoo/ImageNet_21K_P/models/timm/mobilenetv3_large_100_in21k_miil.pth', num_classes=11221),
 
-    'mobilenetv3_small_050': _cfg(url=''),
-    'mobilenetv3_small_075': _cfg(url=''),
-    'mobilenetv3_small_100': _cfg(url=''),
+    'mobilenetv3_small_050': _cfg(
+        url='https://github.com/rwightman/pytorch-image-models/releases/download/v0.1-weights/mobilenetv3_small_050_lambc-4b7bbe87.pth',
+        interpolation='bicubic'),
+    'mobilenetv3_small_075': _cfg(
+        url='https://github.com/rwightman/pytorch-image-models/releases/download/v0.1-weights/mobilenetv3_small_075_lambc-384766db.pth',
+        interpolation='bicubic'),
+    'mobilenetv3_small_100': _cfg(
+        url='https://github.com/rwightman/pytorch-image-models/releases/download/v0.1-weights/mobilenetv3_small_100_lamb-266a294c.pth',
+        interpolation='bicubic'),
 
     'mobilenetv3_rw': _cfg(
         url='https://github.com/rwightman/pytorch-image-models/releases/download/v0.1-weights/mobilenetv3_100-35495452.pth',
@@ -75,14 +80,29 @@ default_cfgs = {
         url='https://github.com/rwightman/pytorch-image-models/releases/download/v0.1-weights/tf_mobilenetv3_small_minimal_100-922a7843.pth',
         mean=IMAGENET_INCEPTION_MEAN, std=IMAGENET_INCEPTION_STD),
 
-    'fbnetv3_b': _cfg(),
-    'fbnetv3_d': _cfg(),
-    'fbnetv3_g': _cfg(),
+    'fbnetv3_b': _cfg(
+        url='https://github.com/rwightman/pytorch-image-models/releases/download/v0.1-weights/fbnetv3_b_224-ead5d2a1.pth',
+        test_input_size=(3, 256, 256), crop_pct=0.95),
+    'fbnetv3_d': _cfg(
+        url='https://github.com/rwightman/pytorch-image-models/releases/download/v0.1-weights/fbnetv3_d_224-c98bce42.pth',
+        test_input_size=(3, 256, 256), crop_pct=0.95),
+    'fbnetv3_g': _cfg(
+        url='https://github.com/rwightman/pytorch-image-models/releases/download/v0.1-weights/fbnetv3_g_240-0b1df83b.pth',
+        input_size=(3, 240, 240), test_input_size=(3, 288, 288), crop_pct=0.95),
 
     "lcnet_035": _cfg(),
-    "lcnet_050": _cfg(),
-    "lcnet_075": _cfg(),
-    "lcnet_100": _cfg(),
+    "lcnet_050": _cfg(
+        url='https://github.com/rwightman/pytorch-image-models/releases/download/v0.1-weights/lcnet_050-f447553b.pth',
+        interpolation='bicubic',
+    ),
+    "lcnet_075": _cfg(
+        url='https://github.com/rwightman/pytorch-image-models/releases/download/v0.1-weights/lcnet_075-318cad2c.pth',
+        interpolation='bicubic',
+    ),
+    "lcnet_100": _cfg(
+        url='https://github.com/rwightman/pytorch-image-models/releases/download/v0.1-weights/lcnet_100-a929038c.pth',
+        interpolation='bicubic',
+    ),
     "lcnet_150": _cfg(),
 }
 
@@ -102,9 +122,10 @@ class MobileNetV3(nn.Module):
       * LCNet - https://arxiv.org/abs/2109.15099
     """
 
-    def __init__(self, block_args, num_classes=1000, in_chans=3, stem_size=16, num_features=1280, head_bias=True,
-                 pad_type='', act_layer=None, norm_layer=None, se_layer=None, se_from_exp=True,
-                 round_chs_fn=round_channels, drop_rate=0., drop_path_rate=0., global_pool='avg'):
+    def __init__(
+            self, block_args, num_classes=1000, in_chans=3, stem_size=16, fix_stem=False, num_features=1280,
+            head_bias=True, pad_type='', act_layer=None, norm_layer=None, se_layer=None, se_from_exp=True,
+            round_chs_fn=round_channels, drop_rate=0., drop_path_rate=0., global_pool='avg'):
         super(MobileNetV3, self).__init__()
         act_layer = act_layer or nn.ReLU
         norm_layer = norm_layer or nn.BatchNorm2d
@@ -115,7 +136,8 @@ class MobileNetV3(nn.Module):
         self.drop_rate = drop_rate
 
         # Stem
-        stem_size = round_chs_fn(stem_size)
+        if not fix_stem:
+            stem_size = round_chs_fn(stem_size)
         self.conv_stem = create_conv2d(in_chans, stem_size, 3, stride=2, padding=pad_type)
         self.bn1 = norm_act_layer(stem_size, inplace=True)
 
@@ -179,8 +201,8 @@ class MobileNetV3Features(nn.Module):
     """
 
     def __init__(self, block_args, out_indices=(0, 1, 2, 3, 4), feature_location='bottleneck', in_chans=3,
-                 stem_size=16, output_stride=32, pad_type='', round_chs_fn=round_channels, se_from_exp=True,
-                 act_layer=None, norm_layer=None, se_layer=None, drop_rate=0., drop_path_rate=0.):
+                 stem_size=16, fix_stem=False, output_stride=32, pad_type='', round_chs_fn=round_channels,
+                 se_from_exp=True, act_layer=None, norm_layer=None, se_layer=None, drop_rate=0., drop_path_rate=0.):
         super(MobileNetV3Features, self).__init__()
         act_layer = act_layer or nn.ReLU
         norm_layer = norm_layer or nn.BatchNorm2d
@@ -188,7 +210,8 @@ class MobileNetV3Features(nn.Module):
         self.drop_rate = drop_rate
 
         # Stem
-        stem_size = round_chs_fn(stem_size)
+        if not fix_stem:
+            stem_size = round_chs_fn(stem_size)
         self.conv_stem = create_conv2d(in_chans, stem_size, 3, stride=2, padding=pad_type)
         self.bn1 = norm_layer(stem_size)
         self.act1 = act_layer(inplace=True)
@@ -371,6 +394,7 @@ def _gen_mobilenet_v3(variant, channel_multiplier=1.0, pretrained=False, **kwarg
         block_args=decode_arch_def(arch_def),
         num_features=num_features,
         stem_size=16,
+        fix_stem=channel_multiplier < 0.75,
         round_chs_fn=partial(round_channels, multiplier=channel_multiplier),
         norm_layer=partial(nn.BatchNorm2d, **resolve_bn_args(kwargs)),
         act_layer=act_layer,
@@ -480,6 +504,44 @@ def _gen_lcnet(variant, channel_multiplier=1.0, pretrained=False, **kwargs):
     return model
 
 
+def _gen_lcnet(variant, channel_multiplier=1.0, pretrained=False, **kwargs):
+    """ LCNet
+    Essentially a MobileNet-V3 crossed with a MobileNet-V1
+
+    Paper: `PP-LCNet: A Lightweight CPU Convolutional Neural Network` - https://arxiv.org/abs/2109.15099
+
+    Args:
+      channel_multiplier: multiplier to number of channels per layer.
+    """
+    arch_def = [
+        # stage 0, 112x112 in
+        ['dsa_r1_k3_s1_c32'],
+        # stage 1, 112x112 in
+        ['dsa_r2_k3_s2_c64'],
+        # stage 2, 56x56 in
+        ['dsa_r2_k3_s2_c128'],
+        # stage 3, 28x28 in
+        ['dsa_r1_k3_s2_c256', 'dsa_r1_k5_s1_c256'],
+        # stage 4, 14x14in
+        ['dsa_r4_k5_s1_c256'],
+        # stage 5, 14x14in
+        ['dsa_r2_k5_s2_c512_se0.25'],
+        # 7x7
+    ]
+    model_kwargs = dict(
+        block_args=decode_arch_def(arch_def),
+        stem_size=16,
+        round_chs_fn=partial(round_channels, multiplier=channel_multiplier),
+        norm_layer=partial(nn.BatchNorm2d, **resolve_bn_args(kwargs)),
+        act_layer=resolve_act_layer(kwargs, 'hard_swish'),
+        se_layer=partial(SqueezeExcite, gate_layer='hard_sigmoid', force_act_layer=nn.ReLU),
+        num_features=1280,
+        **kwargs,
+    )
+    model = _create_mnv3(variant, pretrained, **model_kwargs)
+    return model
+
+
 @register_model
 def mobilenetv3_large_075(pretrained=False, **kwargs):
     """ MobileNet V3 """

timm/models/resnet.py

@@ -4,7 +4,8 @@ This started as a copy of https://github.com/pytorch/vision 'resnet.py' (BSD-3-C
 additional dropout and dynamic global avg/max pool.
 
 ResNeXt, SE-ResNeXt, SENet, and MXNet Gluon stem/downsample variants, tiered stems added by Ross Wightman
-Copyright 2020 Ross Wightman
+
+Copyright 2019, Ross Wightman
 """
 import math
 from functools import partial
@@ -250,6 +251,21 @@ default_cfgs = {
     'resnetblur50': _cfg(
         url='https://github.com/rwightman/pytorch-image-models/releases/download/v0.1-weights/resnetblur50-84f4748f.pth',
         interpolation='bicubic'),
+    'resnetblur50d': _cfg(
+        url='',
+        interpolation='bicubic', first_conv='conv1.0'),
+    'resnetblur101d': _cfg(
+        url='',
+        interpolation='bicubic', first_conv='conv1.0'),
+    'resnetaa50d': _cfg(
+        url='',
+        interpolation='bicubic', first_conv='conv1.0'),
+    'resnetaa101d': _cfg(
+        url='',
+        interpolation='bicubic', first_conv='conv1.0'),
+    'seresnetaa50d': _cfg(
+        url='',
+        interpolation='bicubic', first_conv='conv1.0'),
 
     # ResNet-RS models
     'resnetrs50': _cfg(
@@ -288,6 +304,12 @@ def get_padding(kernel_size, stride, dilation=1):
     return padding
 
 
+def create_aa(aa_layer, channels, stride=2, enable=True):
+    if not aa_layer or not enable:
+        return nn.Identity()
+    return aa_layer(stride) if issubclass(aa_layer, nn.AvgPool2d) else aa_layer(channels=channels, stride=stride)
+
+
 class BasicBlock(nn.Module):
     expansion = 1
 
@@ -309,7 +331,7 @@ class BasicBlock(nn.Module):
         self.bn1 = norm_layer(first_planes)
         self.drop_block = drop_block() if drop_block is not None else nn.Identity()
         self.act1 = act_layer(inplace=True)
-        self.aa = aa_layer(channels=first_planes, stride=stride) if use_aa else nn.Identity()
+        self.aa = create_aa(aa_layer, channels=first_planes, stride=stride, enable=use_aa)
 
         self.conv2 = nn.Conv2d(
             first_planes, outplanes, kernel_size=3, padding=dilation, dilation=dilation, bias=False)
@@ -376,7 +398,7 @@ class Bottleneck(nn.Module):
         self.bn2 = norm_layer(width)
         self.drop_block = drop_block() if drop_block is not None else nn.Identity()
         self.act2 = act_layer(inplace=True)
-        self.aa = aa_layer(channels=width, stride=stride) if use_aa else nn.Identity()
+        self.aa = create_aa(aa_layer, channels=width, stride=stride, enable=use_aa)
 
         self.conv3 = nn.Conv2d(width, outplanes, kernel_size=1, bias=False)
         self.bn3 = norm_layer(outplanes)
@@ -606,19 +628,22 @@ class ResNet(nn.Module):
         self.act1 = act_layer(inplace=True)
         self.feature_info = [dict(num_chs=inplanes, reduction=2, module='act1')]
 
-        # Stem Pooling
+        # Stem pooling. The name 'maxpool' remains for weight compatibility.
         if replace_stem_pool:
             self.maxpool = nn.Sequential(*filter(None, [
                 nn.Conv2d(inplanes, inplanes, 3, stride=1 if aa_layer else 2, padding=1, bias=False),
-                aa_layer(channels=inplanes, stride=2) if aa_layer else None,
+                create_aa(aa_layer, channels=inplanes, stride=2),
                 norm_layer(inplanes),
                 act_layer(inplace=True)
             ]))
         else:
             if aa_layer is not None:
-                self.maxpool = nn.Sequential(*[
-                    nn.MaxPool2d(kernel_size=3, stride=1, padding=1),
-                    aa_layer(channels=inplanes, stride=2)])
+                if issubclass(aa_layer, nn.AvgPool2d):
+                    self.maxpool = aa_layer(2)
+                else:
+                    self.maxpool = nn.Sequential(*[
+                        nn.MaxPool2d(kernel_size=3, stride=1, padding=1),
+                        aa_layer(channels=inplanes, stride=2)])
             else:
                 self.maxpool = nn.MaxPool2d(kernel_size=3, stride=2, padding=1)
 
@@ -1328,6 +1353,56 @@ def resnetblur50(pretrained=False, **kwargs):
     return _create_resnet('resnetblur50', pretrained, **model_args)
 
 
+@register_model
+def resnetblur50d(pretrained=False, **kwargs):
+    """Constructs a ResNet-50-D model with blur anti-aliasing
+    """
+    model_args = dict(
+        block=Bottleneck, layers=[3, 4, 6, 3], aa_layer=BlurPool2d,
+        stem_width=32, stem_type='deep', avg_down=True, **kwargs)
+    return _create_resnet('resnetblur50d', pretrained, **model_args)
+
+
+@register_model
+def resnetblur101d(pretrained=False, **kwargs):
+    """Constructs a ResNet-101-D model with blur anti-aliasing
+    """
+    model_args = dict(
+        block=Bottleneck, layers=[3, 4, 23, 3], aa_layer=BlurPool2d,
+        stem_width=32, stem_type='deep', avg_down=True, **kwargs)
+    return _create_resnet('resnetblur101d', pretrained, **model_args)
+
+
+@register_model
+def resnetaa50d(pretrained=False, **kwargs):
+    """Constructs a ResNet-50-D model with avgpool anti-aliasing
+    """
+    model_args = dict(
+        block=Bottleneck, layers=[3, 4, 6, 3], aa_layer=nn.AvgPool2d,
+        stem_width=32, stem_type='deep', avg_down=True, **kwargs)
+    return _create_resnet('resnetaa50d', pretrained, **model_args)
+
+
+@register_model
+def resnetaa101d(pretrained=False, **kwargs):
+    """Constructs a ResNet-101-D model with avgpool anti-aliasing
+    """
+    model_args = dict(
+        block=Bottleneck, layers=[3, 4, 23, 3], aa_layer=nn.AvgPool2d,
+        stem_width=32, stem_type='deep', avg_down=True, **kwargs)
+    return _create_resnet('resnetaa101d', pretrained, **model_args)
+
+
+@register_model
+def seresnetaa50d(pretrained=False, **kwargs):
+    """Constructs a SE=ResNet-50-D model with avgpool anti-aliasing
+    """
+    model_args = dict(
+        block=Bottleneck, layers=[3, 4, 6, 3], aa_layer=nn.AvgPool2d,
+        stem_width=32, stem_type='deep', avg_down=True, block_args=dict(attn_layer='se'), **kwargs)
+    return _create_resnet('seresnetaa50d', pretrained, **model_args)
+
+
 @register_model
 def seresnet18(pretrained=False, **kwargs):
     model_args = dict(block=BasicBlock, layers=[2, 2, 2, 2], block_args=dict(attn_layer='se'), **kwargs)

timm/models/swin_transformer.py

@@ -4,6 +4,7 @@ A PyTorch impl of : `Swin Transformer: Hierarchical Vision Transformer using Shi
 
 Code/weights from https://github.com/microsoft/Swin-Transformer, original copyright/license info below
 
+Modifications and additions for timm hacked together by / Copyright 2021, Ross Wightman
 """
 # --------------------------------------------------------
 # Swin Transformer

timm/models/vgg.py

@@ -179,8 +179,8 @@ def _filter_fn(state_dict):
 
 def _create_vgg(variant: str, pretrained: bool, **kwargs: Any) -> VGG:
     cfg = variant.split('_')[0]
-    # NOTE: VGG is one of the only models with stride==1 features, so indices are offset from other models
-    out_indices = kwargs.get('out_indices', (0, 1, 2, 3, 4, 5))
+    # NOTE: VGG is one of few models with stride==1 features w/ 6 out_indices [0..5]
+    out_indices = kwargs.pop('out_indices', (0, 1, 2, 3, 4, 5))
     model = build_model_with_cfg(
         VGG, variant, pretrained,
         model_cfg=cfgs[cfg],

timm/models/visformer.py

@@ -4,6 +4,7 @@ Paper: Visformer: The Vision-friendly Transformer - https://arxiv.org/abs/2104.1
 
 From original at https://github.com/danczs/Visformer
 
+Modifications and additions for timm hacked together by / Copyright 2021, Ross Wightman
 """
 from copy import deepcopy
 
@@ -23,7 +24,7 @@ __all__ = ['Visformer']
 def _cfg(url='', **kwargs):
     return {
         'url': url,
-        'num_classes': 1000, 'input_size': (3, 224, 224), 'pool_size': None,
+        'num_classes': 1000, 'input_size': (3, 224, 224), 'pool_size': (7, 7),
         'crop_pct': .9, 'interpolation': 'bicubic', 'fixed_input_size': True,
         'mean': IMAGENET_DEFAULT_MEAN, 'std': IMAGENET_DEFAULT_STD,
         'first_conv': 'stem.0', 'classifier': 'head',

timm/models/vision_transformer.py

@@ -20,7 +20,7 @@ for some einops/einsum fun
 * Simple transformer style inspired by Andrej Karpathy's https://github.com/karpathy/minGPT
 * Bert reference code checks against Huggingface Transformers and Tensorflow Bert
 
-Hacked together by / Copyright 2021 Ross Wightman
+Hacked together by / Copyright 2020, Ross Wightman
 """
 import math
 import logging
@@ -105,6 +105,12 @@ default_cfgs = {
             'L_16-i21k-300ep-lr_0.001-aug_medium1-wd_0.1-do_0.1-sd_0.1--imagenet2012-steps_20k-lr_0.01-res_384.npz',
         input_size=(3, 384, 384), crop_pct=1.0),
 
+    'vit_huge_patch14_224': _cfg(url=''),
+    'vit_giant_patch14_224': _cfg(url=''),
+    'vit_gigantic_patch14_224': _cfg(url=''),
+
+    'vit_base2_patch32_256': _cfg(url='', input_size=(3, 256, 256), crop_pct=0.95),
+
     # patch models, imagenet21k (weights from official Google JAX impl)
     'vit_tiny_patch16_224_in21k': _cfg(
         url='https://storage.googleapis.com/vit_models/augreg/Ti_16-i21k-300ep-lr_0.001-aug_none-wd_0.03-do_0.0-sd_0.0.npz',
@@ -136,11 +142,25 @@ default_cfgs = {
         num_classes=21843),
 
     # SAM trained models (https://arxiv.org/abs/2106.01548)
-    'vit_base_patch32_sam_224': _cfg(
+    'vit_base_patch32_224_sam': _cfg(
         url='https://storage.googleapis.com/vit_models/sam/ViT-B_32.npz'),
-    'vit_base_patch16_sam_224': _cfg(
+    'vit_base_patch16_224_sam': _cfg(
         url='https://storage.googleapis.com/vit_models/sam/ViT-B_16.npz'),
 
+    # DINO pretrained - https://arxiv.org/abs/2104.14294 (no classifier head, for fine-tune only)
+    'vit_small_patch16_224_dino': _cfg(
+        url='https://dl.fbaipublicfiles.com/dino/dino_deitsmall16_pretrain/dino_deitsmall16_pretrain.pth',
+        mean=IMAGENET_DEFAULT_MEAN, std=IMAGENET_DEFAULT_STD, num_classes=0),
+    'vit_small_patch8_224_dino': _cfg(
+        url='https://dl.fbaipublicfiles.com/dino/dino_deitsmall8_pretrain/dino_deitsmall8_pretrain.pth',
+        mean=IMAGENET_DEFAULT_MEAN, std=IMAGENET_DEFAULT_STD, num_classes=0),
+    'vit_base_patch16_224_dino': _cfg(
+        url='https://dl.fbaipublicfiles.com/dino/dino_vitbase16_pretrain/dino_vitbase16_pretrain.pth',
+        mean=IMAGENET_DEFAULT_MEAN, std=IMAGENET_DEFAULT_STD, num_classes=0),
+    'vit_base_patch8_224_dino': _cfg(
+        url='https://dl.fbaipublicfiles.com/dino/dino_vitbase8_pretrain/dino_vitbase8_pretrain.pth',
+        mean=IMAGENET_DEFAULT_MEAN, std=IMAGENET_DEFAULT_STD, num_classes=0),
+
     # deit models (FB weights)
     'deit_tiny_patch16_224': _cfg(
         url='https://dl.fbaipublicfiles.com/deit/deit_tiny_patch16_224-a1311bcf.pth',
@@ -184,6 +204,7 @@ default_cfgs = {
 class Attention(nn.Module):
     def __init__(self, dim, num_heads=8, qkv_bias=False, attn_drop=0., proj_drop=0.):
         super().__init__()
+        assert dim % num_heads == 0, 'dim should be divisible by num_heads'
         self.num_heads = num_heads
         head_dim = dim // num_heads
         self.scale = head_dim ** -0.5
@@ -616,6 +637,16 @@ def vit_base_patch32_224(pretrained=False, **kwargs):
     return model
 
 
+@register_model
+def vit_base2_patch32_256(pretrained=False, **kwargs):
+    """ ViT-Base (ViT-B/32)
+    # FIXME experiment
+    """
+    model_kwargs = dict(patch_size=32, embed_dim=896, depth=12, num_heads=14, **kwargs)
+    model = _create_vision_transformer('vit_base2_patch32_256', pretrained=pretrained, **model_kwargs)
+    return model
+
+
 @register_model
 def vit_base_patch32_384(pretrained=False, **kwargs):
     """ ViT-Base model (ViT-B/32) from original paper (https://arxiv.org/abs/2010.11929).
@@ -696,22 +727,29 @@ def vit_large_patch16_384(pretrained=False, **kwargs):
 
 
 @register_model
-def vit_base_patch16_sam_224(pretrained=False, **kwargs):
-    """ ViT-Base (ViT-B/16) w/ SAM pretrained weights. Paper: https://arxiv.org/abs/2106.01548
+def vit_huge_patch14_224(pretrained=False, **kwargs):
+    """ ViT-Huge model (ViT-H/14) from original paper (https://arxiv.org/abs/2010.11929).
     """
-    # NOTE original SAM weights release worked with representation_size=768
-    model_kwargs = dict(patch_size=16, embed_dim=768, depth=12, num_heads=12, representation_size=0, **kwargs)
-    model = _create_vision_transformer('vit_base_patch16_sam_224', pretrained=pretrained, **model_kwargs)
+    model_kwargs = dict(patch_size=14, embed_dim=1280, depth=32, num_heads=16, **kwargs)
+    model = _create_vision_transformer('vit_huge_patch14_224', pretrained=pretrained, **model_kwargs)
     return model
 
 
 @register_model
-def vit_base_patch32_sam_224(pretrained=False, **kwargs):
-    """ ViT-Base (ViT-B/32) w/ SAM pretrained weights. Paper: https://arxiv.org/abs/2106.01548
+def vit_giant_patch14_224(pretrained=False, **kwargs):
+    """ ViT-Giant model (ViT-g/14) from `Scaling Vision Transformers` - https://arxiv.org/abs/2106.04560
     """
-    # NOTE original SAM weights release worked with representation_size=768
-    model_kwargs = dict(patch_size=32, embed_dim=768, depth=12, num_heads=12, representation_size=0, **kwargs)
-    model = _create_vision_transformer('vit_base_patch32_sam_224', pretrained=pretrained, **model_kwargs)
+    model_kwargs = dict(patch_size=14, embed_dim=1408, mlp_ratio=48/11, depth=40, num_heads=16, **kwargs)
+    model = _create_vision_transformer('vit_giant_patch14_224', pretrained=pretrained, **model_kwargs)
+    return model
+
+
+@register_model
+def vit_gigantic_patch14_224(pretrained=False, **kwargs):
+    """ ViT-Gigantic model (ViT-G/14) from `Scaling Vision Transformers` - https://arxiv.org/abs/2106.04560
+    """
+    model_kwargs = dict(patch_size=14, embed_dim=1664, mlp_ratio=64/13, depth=48, num_heads=16, **kwargs)
+    model = _create_vision_transformer('vit_gigantic_patch14_224', pretrained=pretrained, **model_kwargs)
     return model
 
 
@@ -820,6 +858,62 @@ def vit_huge_patch14_224_in21k(pretrained=False, **kwargs):
     return model
 
 
+@register_model
+def vit_base_patch16_224_sam(pretrained=False, **kwargs):
+    """ ViT-Base (ViT-B/16) w/ SAM pretrained weights. Paper: https://arxiv.org/abs/2106.01548
+    """
+    # NOTE original SAM weights release worked with representation_size=768
+    model_kwargs = dict(patch_size=16, embed_dim=768, depth=12, num_heads=12, **kwargs)
+    model = _create_vision_transformer('vit_base_patch16_224_sam', pretrained=pretrained, **model_kwargs)
+    return model
+
+
+@register_model
+def vit_base_patch32_224_sam(pretrained=False, **kwargs):
+    """ ViT-Base (ViT-B/32) w/ SAM pretrained weights. Paper: https://arxiv.org/abs/2106.01548
+    """
+    # NOTE original SAM weights release worked with representation_size=768
+    model_kwargs = dict(patch_size=32, embed_dim=768, depth=12, num_heads=12, **kwargs)
+    model = _create_vision_transformer('vit_base_patch32_224_sam', pretrained=pretrained, **model_kwargs)
+    return model
+
+
+@register_model
+def vit_small_patch16_224_dino(pretrained=False, **kwargs):
+    """ ViT-Small (ViT-S/16) w/ DINO pretrained weights (no head) - https://arxiv.org/abs/2104.14294
+    """
+    model_kwargs = dict(patch_size=16, embed_dim=384, depth=12, num_heads=6, **kwargs)
+    model = _create_vision_transformer('vit_small_patch16_224_dino', pretrained=pretrained, **model_kwargs)
+    return model
+
+
+@register_model
+def vit_small_patch8_224_dino(pretrained=False, **kwargs):
+    """ ViT-Small (ViT-S/8) w/ DINO pretrained weights (no head) - https://arxiv.org/abs/2104.14294
+    """
+    model_kwargs = dict(patch_size=8, embed_dim=384, depth=12, num_heads=6, **kwargs)
+    model = _create_vision_transformer('vit_small_patch8_224_dino', pretrained=pretrained, **model_kwargs)
+    return model
+
+
+@register_model
+def vit_base_patch16_224_dino(pretrained=False, **kwargs):
+    """ ViT-Base (ViT-B/16) /w DINO pretrained weights (no head) - https://arxiv.org/abs/2104.14294
+    """
+    model_kwargs = dict(patch_size=16, embed_dim=768, depth=12, num_heads=12, **kwargs)
+    model = _create_vision_transformer('vit_base_patch16_224_dino', pretrained=pretrained, **model_kwargs)
+    return model
+
+
+@register_model
+def vit_base_patch8_224_dino(pretrained=False, **kwargs):
+    """ ViT-Base (ViT-B/8) w/ DINO pretrained weights (no head) - https://arxiv.org/abs/2104.14294
+    """
+    model_kwargs = dict(patch_size=8, embed_dim=768, depth=12, num_heads=12, **kwargs)
+    model = _create_vision_transformer('vit_base_patch8_224_dino', pretrained=pretrained, **model_kwargs)
+    return model
+
+
 @register_model
 def deit_tiny_patch16_224(pretrained=False, **kwargs):
     """ DeiT-tiny model @ 224x224 from paper (https://arxiv.org/abs/2012.12877).

timm/models/vision_transformer_hybrid.py

@@ -11,7 +11,7 @@ A PyTorch implement of the Hybrid Vision Transformers as described in:
 NOTE These hybrid model definitions depend on code in vision_transformer.py.
 They were moved here to keep file sizes sane.
 
-Hacked together by / Copyright 2021 Ross Wightman
+Hacked together by / Copyright 2020, Ross Wightman
 """
 from copy import deepcopy
 from functools import partial

timm/models/xcit.py

@@ -1,10 +1,12 @@
 """ Cross-Covariance Image Transformer (XCiT) in PyTorch
 
-Same as the official implementation, with some minor adaptations.
-    - https://github.com/facebookresearch/xcit/blob/master/xcit.py
-
 Paper:
     - https://arxiv.org/abs/2106.09681
+
+Same as the official implementation, with some minor adaptations, original copyright below
+    - https://github.com/facebookresearch/xcit/blob/master/xcit.py
+
+Modifications and additions for timm hacked together by / Copyright 2021, Ross Wightman
 """
 # Copyright (c) 2015-present, Facebook, Inc.
 # All rights reserved.

timm/optim/lars.py

@@ -114,7 +114,7 @@ class Lars(Optimizer):
                     )
                     if group['trust_clip']:
                         trust_ratio = torch.minimum(trust_ratio / group['lr'], one_tensor)
-                    grad.add(p, alpha=weight_decay)
+                    grad.add_(p, alpha=weight_decay)
                     grad.mul_(trust_ratio)
 
                 # apply SGD update https://github.com/pytorch/pytorch/blob/1.7/torch/optim/sgd.py#L100

timm/utils/__init__.py

@@ -3,7 +3,7 @@ from .checkpoint_saver import CheckpointSaver
 from .clip_grad import dispatch_clip_grad
 from .cuda import ApexScaler, NativeScaler
 from .distributed import distribute_bn, reduce_tensor
-from .jit import set_jit_legacy
+from .jit import set_jit_legacy, set_jit_fuser
 from .log import setup_default_logging, FormatterNoInfo
 from .metrics import AverageMeter, accuracy
 from .misc import natural_key, add_bool_arg

timm/utils/jit.py

@@ -2,6 +2,8 @@
 
 Hacked together by / Copyright 2020 Ross Wightman
 """
+import os
+
 import torch
 
 
@@ -16,3 +18,33 @@ def set_jit_legacy():
     torch._C._jit_set_profiling_mode(False)
     torch._C._jit_override_can_fuse_on_gpu(True)
     #torch._C._jit_set_texpr_fuser_enabled(True)
+
+
+def set_jit_fuser(fuser):
+    if fuser == "te":
+        # default fuser should be == 'te'
+        torch._C._jit_set_profiling_executor(True)
+        torch._C._jit_set_profiling_mode(True)
+        torch._C._jit_override_can_fuse_on_cpu(False)
+        torch._C._jit_override_can_fuse_on_gpu(True)
+        torch._C._jit_set_texpr_fuser_enabled(True)
+    elif fuser == "old" or fuser == "legacy":
+        torch._C._jit_set_profiling_executor(False)
+        torch._C._jit_set_profiling_mode(False)
+        torch._C._jit_override_can_fuse_on_gpu(True)
+        torch._C._jit_set_texpr_fuser_enabled(False)
+    elif fuser == "nvfuser" or fuser == "nvf":
+        os.environ['PYTORCH_CUDA_FUSER_DISABLE_FALLBACK'] = '1'
+        os.environ['PYTORCH_CUDA_FUSER_DISABLE_FMA'] = '1'
+        os.environ['PYTORCH_CUDA_FUSER_JIT_OPT_LEVEL'] = '0'
+        torch._C._jit_set_texpr_fuser_enabled(False)
+        torch._C._jit_set_profiling_executor(True)
+        torch._C._jit_set_profiling_mode(True)
+        torch._C._jit_can_fuse_on_cpu()
+        torch._C._jit_can_fuse_on_gpu()
+        torch._C._jit_override_can_fuse_on_cpu(False)
+        torch._C._jit_override_can_fuse_on_gpu(False)
+        torch._C._jit_set_nvfuser_guard_mode(True)
+        torch._C._jit_set_nvfuser_enabled(True)
+    else:
+        assert False, f"Invalid jit fuser ({fuser})"

timm/version.py

@@ -1 +1 @@
-__version__ = '0.5.0'
+__version__ = '0.5.5'

train.py

@@ -295,6 +295,8 @@ parser.add_argument('--use-multi-epochs-loader', action='store_true', default=Fa
                     help='use the multi-epochs-loader to save time at the beginning of every epoch')
 parser.add_argument('--torchscript', dest='torchscript', action='store_true',
                     help='convert model torchscript for inference')
+parser.add_argument('--fuser', default='', type=str,
+                    help="Select jit fuser. One of ('', 'te', 'old', 'nvfuser')")
 parser.add_argument('--log-wandb', action='store_true', default=False,
                     help='log training and validation metrics to wandb')
 
@@ -364,6 +366,9 @@ def main():
 
     random_seed(args.seed, args.rank)
 
+    if args.fuser:
+        set_jit_fuser(args.fuser)
+
     model = create_model(
         args.model,
         pretrained=args.pretrained,

validate.py

@@ -11,6 +11,7 @@ import argparse
 import os
 import csv
 import glob
+import json
 import time
 import logging
 import torch
@@ -21,7 +22,7 @@ from contextlib import suppress
 
 from timm.models import create_model, apply_test_time_pool, load_checkpoint, is_model, list_models
 from timm.data import create_dataset, create_loader, resolve_data_config, RealLabelsImagenet
-from timm.utils import accuracy, AverageMeter, natural_key, setup_default_logging, set_jit_legacy
+from timm.utils import accuracy, AverageMeter, natural_key, setup_default_logging, set_jit_fuser
 
 has_apex = False
 try:
@@ -102,8 +103,8 @@ parser.add_argument('--use-ema', dest='use_ema', action='store_true',
                     help='use ema version of weights if present')
 parser.add_argument('--torchscript', dest='torchscript', action='store_true',
                     help='convert model torchscript for inference')
-parser.add_argument('--legacy-jit', dest='legacy_jit', action='store_true',
-                    help='use legacy jit mode for pytorch 1.5/1.5.1/1.6 to get back fusion performance')
+parser.add_argument('--fuser', default='', type=str,
+                    help="Select jit fuser. One of ('', 'te', 'old', 'nvfuser')")
 parser.add_argument('--results-file', default='', type=str, metavar='FILENAME',
                     help='Output csv file for validation results (summary)')
 parser.add_argument('--real-labels', default='', type=str, metavar='FILENAME',
@@ -133,8 +134,8 @@ def validate(args):
     else:
         _logger.info('Validating in float32. AMP not enabled.')
 
-    if args.legacy_jit:
-        set_jit_legacy()
+    if args.fuser:
+        set_jit_fuser(args.fuser)
 
     # create model
     model = create_model(
@@ -265,6 +266,7 @@ def validate(args):
     else:
         top1a, top5a = top1.avg, top5.avg
     results = OrderedDict(
+        model=args.model,
         top1=round(top1a, 4), top1_err=round(100 - top1a, 4),
         top5=round(top5a, 4), top5_err=round(100 - top5a, 4),
         param_count=round(param_count / 1e6, 2),
@@ -278,6 +280,27 @@ def validate(args):
     return results
 
 
+def _try_run(args, initial_batch_size):
+    batch_size = initial_batch_size
+    results = OrderedDict()
+    error_str = 'Unknown'
+    while batch_size >= 1:
+        args.batch_size = batch_size
+        torch.cuda.empty_cache()
+        try:
+            results = validate(args)
+            return results
+        except RuntimeError as e:
+            error_str = str(e)
+            if 'channels_last' in error_str:
+                break
+            _logger.warning(f'"{error_str}" while running validation. Reducing batch size to {batch_size} for retry.')
+        batch_size = batch_size // 2
+    results['error'] = error_str
+    _logger.error(f'{args.model} failed to validate ({error_str}).')
+    return results
+
+
 def main():
     setup_default_logging()
     args = parser.parse_args()
@@ -293,7 +316,7 @@ def main():
         if args.model == 'all':
             # validate all models in a list of names with pretrained checkpoints
             args.pretrained = True
-            model_names = list_models(pretrained=True, exclude_filters=['*_in21k', '*_in22k'])
+            model_names = list_models(pretrained=True, exclude_filters=['*_in21k', '*_in22k', '*_dino'])
             model_cfgs = [(n, '') for n in model_names]
         elif not is_model(args.model):
             # model name doesn't exist, try as wildcard filter
@@ -310,36 +333,25 @@ def main():
         _logger.info('Running bulk validation on these pretrained models: {}'.format(', '.join(model_names)))
         results = []
         try:
-            start_batch_size = args.batch_size
+            initial_batch_size = args.batch_size
             for m, c in model_cfgs:
-                batch_size = start_batch_size
                 args.model = m
                 args.checkpoint = c
-                result = OrderedDict(model=args.model)
-                r = {}
-                while not r and batch_size >= args.num_gpu:
-                    torch.cuda.empty_cache()
-                    try:
-                        args.batch_size = batch_size
-                        print('Validating with batch size: %d' % args.batch_size)
-                        r = validate(args)
-                    except RuntimeError as e:
-                        if batch_size <= args.num_gpu:
-                            print("Validation failed with no ability to reduce batch size. Exiting.")
-                            raise e
-                        batch_size = max(batch_size // 2, args.num_gpu)
-                        print("Validation failed, reducing batch size by 50%")
-                result.update(r)
+                r = _try_run(args, initial_batch_size)
+                if 'error' in r:
+                    continue
                 if args.checkpoint:
-                    result['checkpoint'] = args.checkpoint
-                results.append(result)
+                    r['checkpoint'] = args.checkpoint
+                results.append(r)
         except KeyboardInterrupt as e:
             pass
         results = sorted(results, key=lambda x: x['top1'], reverse=True)
         if len(results):
             write_results(results_file, results)
     else:
-        validate(args)
+        results = validate(args)
+    # output results in JSON to stdout w/ delimiter for runner script
+    print(f'--result\n{json.dumps(results, indent=4)}')
 
 
 def write_results(results_file, results):