New PyTorch trained EfficientNet-B2 weights with my RandAugment impl

README.md

@@ -2,6 +2,11 @@
 
 ## What's New
 
+
+### Dec 4, 2019
+* Added weights from the first training from scratch of an EfficientNet (B2) with my new RandAugment implementation. Much better than my previous B2 and very close to the official AdvProp ones (80.4 top-1, 95.08 top-5).
+  * For those interested in hparams, I trained with the following: `./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 --aa rand-m9-noise0.5 --remode pixel --reprob 0.2 --amp --lr .016`
+
 ### Nov 29, 2019
 * Brought EfficientNet and MobileNetV3 up to date with my https://github.com/rwightman/gen-efficientnet-pytorch code. Torchscript and ONNX export compat excluded.
   * AdvProp weights added
@@ -99,8 +104,8 @@ I've leveraged the training scripts in this repository to train a few of the mod
 
 |Model | Prec@1 (Err) | Prec@5 (Err) | Param # | Image Scaling  | Image Size |
 |---|---|---|---|---|---|
+| efficientnet_b2 | 80.402 (19.598) | 95.076 (4.924) | 9.11M | bicubic | 260 |
 | mixnet_xl | 80.120 (19.880) | 95.022 (4.978) | 11.90M | bicubic | 224 |
-| efficientnet_b2 | 79.760 (20.240) | 94.714 (5.286) | 9.11M | bicubic | 260 |
 | resnext50d_32x4d | 79.674 (20.326) | 94.868 (5.132) | 25.1M | bicubic | 224 |
 | mixnet_l | 78.976 (21.024 | 94.184 (5.816) | 7.33M | bicubic | 224 |
 | efficientnet_b1 | 78.692 (21.308) | 94.086 (5.914) | 7.79M | bicubic | 240 |

timm/models/efficientnet.py

@@ -68,10 +68,10 @@ default_cfgs = {
         url='https://github.com/rwightman/pytorch-image-models/releases/download/v0.1-weights/efficientnet_b0-d6904d92.pth'),
     'efficientnet_b1': _cfg(
         url='https://github.com/rwightman/pytorch-image-models/releases/download/v0.1-weights/efficientnet_b1-533bc792.pth',
-        input_size=(3, 240, 240), pool_size=(8, 8), crop_pct=0.882),
+        input_size=(3, 240, 240), pool_size=(8, 8)),
     'efficientnet_b2': _cfg(
-        url='https://github.com/rwightman/pytorch-image-models/releases/download/v0.1-weights/efficientnet_b2-cf78dc4d.pth',
-        input_size=(3, 260, 260), pool_size=(9, 9), crop_pct=0.890),
+        url='https://github.com/rwightman/pytorch-image-models/releases/download/v0.1-weights/efficientnet_b2_ra-bcdf34b7.pth',
+        input_size=(3, 260, 260), pool_size=(9, 9)),
     'efficientnet_b3': _cfg(
         url='', input_size=(3, 300, 300), pool_size=(10, 10), crop_pct=0.904),
     'efficientnet_b4': _cfg(