Replace all None by nn.Identity() in all models reset_classifier when False-values num_classes is given.

Make small code refactoring

timm/models/densenet.py

@@ -2,17 +2,17 @@
 This file is a copy of https://github.com/pytorch/vision 'densenet.py' (BSD-3-Clause) with
 fixed kwargs passthrough and addition of dynamic global avg/max pool.
 """
+import re
 from collections import OrderedDict
 
 import torch
 import torch.nn as nn
 import torch.nn.functional as F
 
-from .registry import register_model
+from timm.data import IMAGENET_DEFAULT_MEAN, IMAGENET_DEFAULT_STD
 from .helpers import load_pretrained
 from .layers import SelectAdaptivePool2d
-from timm.data import IMAGENET_DEFAULT_MEAN, IMAGENET_DEFAULT_STD
+from .registry import register_model
-import re
 
 __all__ = ['DenseNet']
 
@@ -85,6 +85,7 @@ class DenseNet(nn.Module):
         drop_rate (float) - dropout rate after each dense layer
         num_classes (int) - number of classification classes
     """
+
     def __init__(self, growth_rate=32, block_config=(6, 12, 24, 16),
                  num_init_features=64, bn_size=4, drop_rate=0,
                  num_classes=1000, in_chans=3, global_pool='avg'):
@@ -127,8 +128,11 @@ class DenseNet(nn.Module):
     def reset_classifier(self, num_classes, global_pool='avg'):
         self.num_classes = num_classes
         self.global_pool = SelectAdaptivePool2d(pool_type=global_pool)
-        self.classifier = nn.Linear(
+        if num_classes:
-            self.num_features * self.global_pool.feat_mult(), num_classes) if num_classes else None
+            num_features = self.num_features * self.global_pool.feat_mult()
+            self.classifier = nn.Linear(num_features, num_classes)
+        else:
+            self.classifier = nn.Identity()
 
     def forward_features(self, x):
         x = self.features(x)
@@ -157,7 +161,6 @@ def _filter_pretrained(state_dict):
     return state_dict
 
 
-
 @register_model
 def densenet121(pretrained=False, num_classes=1000, in_chans=3, **kwargs):
     r"""Densenet-121 model from

timm/models/dla.py

@@ -11,11 +11,10 @@ import torch
 import torch.nn as nn
 import torch.nn.functional as F
 
-from .registry import register_model
+from timm.data import IMAGENET_DEFAULT_MEAN, IMAGENET_DEFAULT_STD
 from .helpers import load_pretrained
 from .layers import SelectAdaptivePool2d
-from timm.data import IMAGENET_DEFAULT_MEAN, IMAGENET_DEFAULT_STD
+from .registry import register_model
-
 
 __all__ = ['DLA']
 
@@ -51,6 +50,7 @@ default_cfgs = {
 
 class DlaBasic(nn.Module):
     """DLA Basic"""
+
     def __init__(self, inplanes, planes, stride=1, dilation=1, **_):
         super(DlaBasic, self).__init__()
         self.conv1 = nn.Conv2d(
@@ -304,9 +304,10 @@ class DLA(nn.Module):
         self.num_classes = num_classes
         self.global_pool = SelectAdaptivePool2d(pool_type=global_pool)
         if num_classes:
-            self.fc = nn.Conv2d(self.num_features * self.global_pool.feat_mult(), num_classes, 1, bias=True)
+            num_features = self.num_features * self.global_pool.feat_mult()
+            self.fc = nn.Conv2d(num_features, num_classes, kernel_size=1, bias=True)
         else:
-            self.fc = None
+            self.fc = nn.Identity()
 
     def forward_features(self, x):
         x = self.base_layer(x)

timm/models/dpn.py

@@ -9,16 +9,16 @@ from __future__ import absolute_import
 from __future__ import division
 from __future__ import print_function
 
+from collections import OrderedDict
+
 import torch
 import torch.nn as nn
 import torch.nn.functional as F
-from collections import OrderedDict
 
-from .registry import register_model
+from timm.data import IMAGENET_DPN_MEAN, IMAGENET_DPN_STD
 from .helpers import load_pretrained
 from .layers import SelectAdaptivePool2d
-from timm.data import IMAGENET_DPN_MEAN, IMAGENET_DPN_STD
+from .registry import register_model
-
 
 __all__ = ['DPN']
 
@@ -218,8 +218,8 @@ class DPN(nn.Module):
 
         # Using 1x1 conv for the FC layer to allow the extra pooling scheme
         self.global_pool = SelectAdaptivePool2d(pool_type=global_pool)
-        self.classifier = nn.Conv2d(
+        num_features = self.num_features * self.global_pool.feat_mult()
-            self.num_features * self.global_pool.feat_mult(), num_classes, kernel_size=1, bias=True)
+        self.classifier = nn.Conv2d(num_features, num_classes, kernel_size=1, bias=True)
 
     def get_classifier(self):
         return self.classifier
@@ -228,10 +228,10 @@ class DPN(nn.Module):
         self.num_classes = num_classes
         self.global_pool = SelectAdaptivePool2d(pool_type=global_pool)
         if num_classes:
-            self.classifier = nn.Conv2d(
+            num_features = self.num_features * self.global_pool.feat_mult()
-                self.num_features * self.global_pool.feat_mult(), num_classes, kernel_size=1, bias=True)
+            self.classifier = nn.Conv2d(num_features, num_classes, kernel_size=1, bias=True)
         else:
-            self.classifier = None
+            self.classifier = nn.Identity()
 
     def forward_features(self, x):
         return self.features(x)

timm/models/efficientnet.py

@@ -24,14 +24,12 @@ An implementation of EfficienNet that covers variety of related models with effi
 
 Hacked together by Ross Wightman
 """
+from timm.data import IMAGENET_DEFAULT_MEAN, IMAGENET_DEFAULT_STD, IMAGENET_INCEPTION_MEAN, IMAGENET_INCEPTION_STD
 from .efficientnet_builder import *
 from .feature_hooks import FeatureHooks
-from .registry import register_model
 from .helpers import load_pretrained, adapt_model_from_file
 from .layers import SelectAdaptivePool2d
-from timm.models.layers import create_conv2d
+from .registry import register_model
-from timm.data import IMAGENET_DEFAULT_MEAN, IMAGENET_DEFAULT_STD, IMAGENET_INCEPTION_MEAN, IMAGENET_INCEPTION_STD
-
 
 __all__ = ['EfficientNet']
 
@@ -373,8 +371,11 @@ class EfficientNet(nn.Module):
     def reset_classifier(self, num_classes, global_pool='avg'):
         self.num_classes = num_classes
         self.global_pool = SelectAdaptivePool2d(pool_type=global_pool)
-        self.classifier = nn.Linear(
+        if num_classes:
-            self.num_features * self.global_pool.feat_mult(), num_classes) if num_classes else None
+            num_features = self.num_features * self.global_pool.feat_mult()
+            self.classifier = nn.Linear(num_features, num_classes)
+        else:
+            self.classifier = nn.Identity()
 
     def forward_features(self, x):
         x = self.conv_stem(x)
@@ -1187,6 +1188,7 @@ def efficientnet_cc_b0_8e(pretrained=False, **kwargs):
         pretrained=pretrained, **kwargs)
     return model
 
+
 @register_model
 def efficientnet_cc_b1_8e(pretrained=False, **kwargs):
     """ EfficientNet-CondConv-B1 w/ 8 Experts """
@@ -1242,8 +1244,6 @@ def efficientnet_lite4(pretrained=False, **kwargs):
     return model
 
 
-
-
 @register_model
 def efficientnet_b1_pruned(pretrained=False, **kwargs):
     """ EfficientNet-B1 Pruned. The pruning has been obtained using https://arxiv.org/pdf/2002.08258.pdf  """
@@ -1275,8 +1275,6 @@ def efficientnet_b3_pruned(pretrained=False, **kwargs):
     return model
 
 
-
-
 @register_model
 def tf_efficientnet_b0(pretrained=False, **kwargs):
     """ EfficientNet-B0. Tensorflow compatible variant  """
@@ -1619,6 +1617,7 @@ def tf_efficientnet_cc_b0_8e(pretrained=False, **kwargs):
         pretrained=pretrained, **kwargs)
     return model
 
+
 @register_model
 def tf_efficientnet_cc_b1_8e(pretrained=False, **kwargs):
     """ EfficientNet-CondConv-B1 w/ 8 Experts. Tensorflow compatible variant """
@@ -1764,4 +1763,3 @@ def tf_mixnet_l(pretrained=False, **kwargs):
     model = _gen_mixnet_m(
         'tf_mixnet_l', channel_multiplier=1.3, pretrained=pretrained, **kwargs)
     return model
-

timm/models/gluon_resnet.py

@@ -3,17 +3,11 @@ This file evolved from https://github.com/pytorch/vision 'resnet.py' with (SE)-R
 and ports of Gluon variations (https://github.com/dmlc/gluon-cv/blob/master/gluoncv/model_zoo/resnet.py) 
 by Ross Wightman
 """
-import math
 
-import torch
+from timm.data import IMAGENET_DEFAULT_MEAN, IMAGENET_DEFAULT_STD
-import torch.nn as nn
-import torch.nn.functional as F
-
-from .registry import register_model
 from .helpers import load_pretrained
 from .layers import SEModule
-from timm.data import IMAGENET_DEFAULT_MEAN, IMAGENET_DEFAULT_STD
+from .registry import register_model
-
 from .resnet import ResNet, Bottleneck, BasicBlock
 
 
@@ -202,8 +196,8 @@ def gluon_resnet50_v1e(pretrained=False, num_classes=1000, in_chans=3, **kwargs)
     model = ResNet(Bottleneck, [3, 4, 6, 3], num_classes=num_classes, in_chans=in_chans,
                    stem_width=64, stem_type='deep', avg_down=True, **kwargs)
     model.default_cfg = default_cfg
-    #if pretrained:
+    if pretrained:
-    #    load_pretrained(model, default_cfg, num_classes, in_chans)
+        load_pretrained(model, default_cfg, num_classes, in_chans)
     return model
 
 

timm/models/gluon_xception.py

@@ -6,15 +6,15 @@ Original PyTorch DeepLab impl: https://github.com/jfzhang95/pytorch-deeplab-xcep
 
 Hacked together by Ross Wightman
 """
-import torch
+from collections import OrderedDict
+
 import torch.nn as nn
 import torch.nn.functional as F
-from collections import OrderedDict
 
-from .registry import register_model
+from timm.data import IMAGENET_DEFAULT_MEAN, IMAGENET_DEFAULT_STD
 from .helpers import load_pretrained
 from .layers import SelectAdaptivePool2d
-from timm.data import IMAGENET_DEFAULT_MEAN, IMAGENET_DEFAULT_STD
+from .registry import register_model
 
 __all__ = ['Xception65', 'Xception71']
 
@@ -47,7 +47,6 @@ default_cfgs = {
     }
 }
 
-
 """ PADDING NOTES
 The original PyTorch and Gluon impl of these models dutifully reproduced the 
 aligned padding added to Tensorflow models for Deeplab. This padding was compensating
@@ -394,7 +393,11 @@ class Xception71(nn.Module):
     def reset_classifier(self, num_classes, global_pool='avg'):
         self.num_classes = num_classes
         self.global_pool = SelectAdaptivePool2d(pool_type=global_pool)
-        self.fc = nn.Linear(self.num_features * self.global_pool.feat_mult(), num_classes) if num_classes else None
+        if num_classes:
+            num_features = self.num_features * self.global_pool.feat_mult()
+            self.fc = nn.Linear(num_features, num_classes)
+        else:
+            self.fc = nn.Identity()
 
     def forward_features(self, x):
         # Entry flow
@@ -465,4 +468,3 @@ def gluon_xception71(pretrained=False, num_classes=1000, in_chans=3, **kwargs):
     if pretrained:
         load_pretrained(model, default_cfg, num_classes, in_chans)
     return model
-

timm/models/inception_resnet_v2.py

@@ -6,10 +6,10 @@ import torch
 import torch.nn as nn
 import torch.nn.functional as F
 
-from .registry import register_model
+from timm.data import IMAGENET_INCEPTION_MEAN, IMAGENET_INCEPTION_STD
 from .helpers import load_pretrained
 from .layers import SelectAdaptivePool2d
-from timm.data import IMAGENET_INCEPTION_MEAN, IMAGENET_INCEPTION_STD
+from .registry import register_model
 
 __all__ = ['InceptionResnetV2']
 
@@ -296,8 +296,11 @@ class InceptionResnetV2(nn.Module):
     def reset_classifier(self, num_classes, global_pool='avg'):
         self.global_pool = SelectAdaptivePool2d(pool_type=global_pool)
         self.num_classes = num_classes
-        self.classif = nn.Linear(
+        if num_classes:
-            self.num_features * self.global_pool.feat_mult(), num_classes) if num_classes else None
+            num_features = self.num_features * self.global_pool.feat_mult()
+            self.classif = nn.Linear(num_features, num_classes)
+        else:
+            self.classif = nn.Identity()
 
     def forward_features(self, x):
         x = self.conv2d_1a(x)

timm/models/inception_v3.py

@@ -1,7 +1,8 @@
 from torchvision.models import Inception3
-from .registry import register_model
+
-from .helpers import load_pretrained
 from timm.data import IMAGENET_DEFAULT_STD, IMAGENET_DEFAULT_MEAN, IMAGENET_INCEPTION_MEAN, IMAGENET_INCEPTION_STD
+from .helpers import load_pretrained
+from .registry import register_model
 
 __all__ = []
 

timm/models/inception_v4.py

@@ -6,10 +6,10 @@ import torch
 import torch.nn as nn
 import torch.nn.functional as F
 
-from .registry import register_model
+from timm.data import IMAGENET_INCEPTION_MEAN, IMAGENET_INCEPTION_STD
 from .helpers import load_pretrained
 from .layers import SelectAdaptivePool2d
-from timm.data import IMAGENET_INCEPTION_MEAN, IMAGENET_INCEPTION_STD
+from .registry import register_model
 
 __all__ = ['InceptionV4']
 
@@ -280,8 +280,11 @@ class InceptionV4(nn.Module):
     def reset_classifier(self, num_classes, global_pool='avg'):
         self.global_pool = SelectAdaptivePool2d(pool_type=global_pool)
         self.num_classes = num_classes
-        self.last_linear = nn.Linear(
+        if num_classes:
-            self.num_features * self.global_pool.feat_mult(), num_classes) if num_classes else None
+            num_features = self.num_features * self.global_pool.feat_mult()
+            self.last_linear = nn.Linear(num_features, num_classes)
+        else:
+            self.last_linear = nn.Identity()
 
     def forward_features(self, x):
         return self.features(x)
@@ -303,6 +306,3 @@ def inception_v4(pretrained=False, num_classes=1000, in_chans=3, **kwargs):
     if pretrained:
         load_pretrained(model, default_cfg, num_classes, in_chans)
     return model
-
-
-

timm/models/mobilenetv3.py

@@ -8,13 +8,13 @@ Paper: Searching for MobileNetV3 - https://arxiv.org/abs/1905.02244
 Hacked together by Ross Wightman
 """
 
+from timm.data import IMAGENET_DEFAULT_MEAN, IMAGENET_DEFAULT_STD, IMAGENET_INCEPTION_MEAN, IMAGENET_INCEPTION_STD
 from .efficientnet_builder import *
-from .registry import register_model
+from .feature_hooks import FeatureHooks
 from .helpers import load_pretrained
 from .layers import SelectAdaptivePool2d, create_conv2d
 from .layers.activations import HardSwish, hard_sigmoid
-from .feature_hooks import FeatureHooks
+from .registry import register_model
-from timm.data import IMAGENET_DEFAULT_MEAN, IMAGENET_DEFAULT_STD, IMAGENET_INCEPTION_MEAN, IMAGENET_INCEPTION_STD
 
 __all__ = ['MobileNetV3']
 
@@ -120,8 +120,11 @@ class MobileNetV3(nn.Module):
     def reset_classifier(self, num_classes, global_pool='avg'):
         self.global_pool = SelectAdaptivePool2d(pool_type=global_pool)
         self.num_classes = num_classes
-        self.classifier = nn.Linear(
+        if num_classes:
-            self.num_features * self.global_pool.feat_mult(), num_classes) if self.num_classes else None
+            num_features = self.num_features * self.global_pool.feat_mult()
+            self.classifier = nn.Linear(num_features, num_classes)
+        else:
+            self.classifier = nn.Identity()
 
     def forward_features(self, x):
         x = self.conv_stem(x)
@@ -397,7 +400,6 @@ def mobilenetv3_small_075(pretrained=False, **kwargs):
 
 @register_model
 def mobilenetv3_small_100(pretrained=False, **kwargs):
-    print(kwargs)
     """ MobileNet V3 """
     model = _gen_mobilenet_v3('mobilenetv3_small_100', 1.0, pretrained=pretrained, **kwargs)
     return model

timm/models/nasnet.py

@@ -2,10 +2,9 @@ import torch
 import torch.nn as nn
 import torch.nn.functional as F
 
-from .registry import register_model
 from .helpers import load_pretrained
 from .layers import SelectAdaptivePool2d
-
+from .registry import register_model
 
 __all__ = ['NASNetALarge']
 
@@ -562,9 +561,11 @@ class NASNetALarge(nn.Module):
     def reset_classifier(self, num_classes, global_pool='avg'):
         self.num_classes = num_classes
         self.global_pool = SelectAdaptivePool2d(pool_type=global_pool)
-        del self.last_linear
+        if num_classes:
-        self.last_linear = nn.Linear(
+            num_features = self.num_features * self.global_pool.feat_mult()
-            self.num_features * self.global_pool.feat_mult(), num_classes) if num_classes else None
+            self.last_linear = nn.Linear(num_features, num_classes)
+        else:
+            self.last_linear = nn.Identity()
 
     def forward_features(self, x):
         x_conv0 = self.conv0(x)

timm/models/pnasnet.py

@@ -6,15 +6,16 @@
 
 """
 from __future__ import print_function, division, absolute_import
+
 from collections import OrderedDict
 
 import torch
 import torch.nn as nn
 import torch.nn.functional as F
 
-from .registry import register_model
 from .helpers import load_pretrained
 from .layers import SelectAdaptivePool2d
+from .registry import register_model
 
 __all__ = ['PNASNet5Large']
 
@@ -349,11 +350,11 @@ class PNASNet5Large(nn.Module):
     def reset_classifier(self, num_classes, global_pool='avg'):
         self.num_classes = num_classes
         self.global_pool = SelectAdaptivePool2d(pool_type=global_pool)
-        del self.last_linear
         if num_classes:
-            self.last_linear = nn.Linear(self.num_features * self.global_pool.feat_mult(), num_classes)
+            num_features = self.num_features * self.global_pool.feat_mult()
+            self.last_linear = nn.Linear(num_features, num_classes)
         else:
-            self.last_linear = None
+            self.last_linear = nn.Identity()
 
     def forward_features(self, x):
         x_conv_0 = self.conv_0(x)

timm/models/res2net.py

@@ -6,13 +6,11 @@ import math
 
 import torch
 import torch.nn as nn
-import torch.nn.functional as F
 
-from .resnet import ResNet
-from .registry import register_model
-from .helpers import load_pretrained
-from .layers import SEModule
 from timm.data import IMAGENET_DEFAULT_MEAN, IMAGENET_DEFAULT_STD
+from .helpers import load_pretrained
+from .registry import register_model
+from .resnet import ResNet
 
 __all__ = []
 

timm/models/resnet.py

@@ -10,10 +10,10 @@ import math
 import torch.nn as nn
 import torch.nn.functional as F
 
-from .registry import register_model
+from timm.data import IMAGENET_DEFAULT_MEAN, IMAGENET_DEFAULT_STD
 from .helpers import load_pretrained, adapt_model_from_file
 from .layers import SelectAdaptivePool2d, DropBlock2d, DropPath, AvgPool2dSame, create_attn, BlurPool2d
-from timm.data import IMAGENET_DEFAULT_MEAN, IMAGENET_DEFAULT_STD
+from .registry import register_model
 
 __all__ = ['ResNet', 'BasicBlock', 'Bottleneck']  # model_registry will add each entrypoint fn to this
 
@@ -377,6 +377,7 @@ class ResNet(nn.Module):
     global_pool : str, default 'avg'
         Global pooling type. One of 'avg', 'max', 'avgmax', 'catavgmax'
     """
+
     def __init__(self, block, layers, num_classes=1000, in_chans=3,
                  cardinality=1, base_width=64, stem_width=64, stem_type='',
                  block_reduce_first=1, down_kernel_size=1, avg_down=False, output_stride=32,
@@ -482,8 +483,11 @@ class ResNet(nn.Module):
     def reset_classifier(self, num_classes, global_pool='avg'):
         self.global_pool = SelectAdaptivePool2d(pool_type=global_pool)
         self.num_classes = num_classes
-        del self.fc
+        if num_classes:
-        self.fc = nn.Linear(self.num_features * self.global_pool.feat_mult(), num_classes) if num_classes else None
+            num_features = self.num_features * self.global_pool.feat_mult()
+            self.fc = nn.Linear(num_features, num_classes)
+        else:
+            self.fc = nn.Identity()
 
     def forward_features(self, x):
         x = self.conv1(x)

timm/models/selecsls.py

@@ -9,16 +9,15 @@ https://arxiv.org/abs/1907.00837
 Based on ResNet implementation in https://github.com/rwightman/pytorch-image-models
 and SelecSLS Net implementation in https://github.com/mehtadushy/SelecSLS-Pytorch
 """
-import math
 
 import torch
 import torch.nn as nn
 import torch.nn.functional as F
 
-from .registry import register_model
+from timm.data import IMAGENET_DEFAULT_MEAN, IMAGENET_DEFAULT_STD
 from .helpers import load_pretrained
 from .layers import SelectAdaptivePool2d
-from timm.data import IMAGENET_DEFAULT_MEAN, IMAGENET_DEFAULT_STD
+from .registry import register_model
 
 __all__ = ['SelecSLS']  # model_registry will add each entrypoint fn to this
 
@@ -134,11 +133,11 @@ class SelecSLS(nn.Module):
     def reset_classifier(self, num_classes, global_pool='avg'):
         self.global_pool = SelectAdaptivePool2d(pool_type=global_pool)
         self.num_classes = num_classes
-        del self.fc
         if num_classes:
-            self.fc = nn.Linear(self.num_features * self.global_pool.feat_mult(), num_classes)
+            num_features = self.num_features * self.global_pool.feat_mult()
+            self.fc = nn.Linear(num_features, num_classes)
         else:
-            self.fc = None
+            self.fc = nn.Identity()
 
     def forward_features(self, x):
         x = self.stem(x)

timm/models/senet.py

@@ -8,16 +8,16 @@ Original model: https://github.com/hujie-frank/SENet
 ResNet code gently borrowed from
 https://github.com/pytorch/vision/blob/master/torchvision/models/resnet.py
 """
-from collections import OrderedDict
 import math
+from collections import OrderedDict
 
 import torch.nn as nn
 import torch.nn.functional as F
 
-from .registry import register_model
+from timm.data import IMAGENET_DEFAULT_MEAN, IMAGENET_DEFAULT_STD
 from .helpers import load_pretrained
 from .layers import SelectAdaptivePool2d
-from timm.data import IMAGENET_DEFAULT_MEAN, IMAGENET_DEFAULT_STD
+from .registry import register_model
 
 __all__ = ['SENet']
 
@@ -369,11 +369,11 @@ class SENet(nn.Module):
     def reset_classifier(self, num_classes, global_pool='avg'):
         self.num_classes = num_classes
         self.avg_pool = SelectAdaptivePool2d(pool_type=global_pool)
-        del self.last_linear
         if num_classes:
-            self.last_linear = nn.Linear(self.num_features * self.avg_pool.feat_mult(), num_classes)
+            num_features = self.num_features * self.avg_pool.feat_mult()
+            self.last_linear = nn.Linear(num_features, num_classes)
         else:
-            self.last_linear = None
+            self.last_linear = nn.Identity()
 
     def forward_features(self, x):
         x = self.layer0(x)

timm/models/tresnet.py

@@ -5,14 +5,16 @@ https://arxiv.org/pdf/2003.13630.pdf
 Original model: https://github.com/mrT23/TResNet
 
 """
+from collections import OrderedDict
 from functools import partial
+
 import torch
 import torch.nn as nn
 import torch.nn.functional as F
-from collections import OrderedDict
+
+from .helpers import load_pretrained
 from .layers import SpaceToDepthModule, AntiAliasDownsampleLayer, SelectAdaptivePool2d
 from .registry import register_model
-from .helpers import load_pretrained
 
 try:
     from inplace_abn import InPlaceABN
@@ -88,7 +90,7 @@ class FastSEModule(nn.Module):
 
 
 def IABN2Float(module: nn.Module) -> nn.Module:
-    "If `module` is IABN don't use half precision."
+    """If `module` is IABN don't use half precision."""
     if isinstance(module, InPlaceABN):
         module.float()
     for child in module.children():
@@ -277,8 +279,10 @@ class TResNet(nn.Module):
         self.num_classes = num_classes
         self.head = None
         if num_classes:
-            self.head = nn.Sequential(OrderedDict([
+            num_features = self.num_features * self.global_pool.feat_mult()
-                ('fc', nn.Linear(self.num_features * self.global_pool.feat_mult(), num_classes))]))
+            self.head = nn.Sequential(OrderedDict([('fc', nn.Linear(num_features, num_classes))]))
+        else:
+            self.head = nn.Sequential(OrderedDict([('fc', nn.Identity())]))
 
     def forward_features(self, x):
         return self.body(x)

timm/models/xception.py

@@ -21,15 +21,13 @@ normalize = transforms.Normalize(mean=[0.5, 0.5, 0.5],
 
 The resize parameter of the validation transform should be 333, and make sure to center crop at 299x299
 """
-import math
 
-import torch
 import torch.nn as nn
 import torch.nn.functional as F
 
-from .registry import register_model
 from .helpers import load_pretrained
 from .layers import SelectAdaptivePool2d
+from .registry import register_model
 
 __all__ = ['Xception']
 
@@ -180,8 +178,11 @@ class Xception(nn.Module):
     def reset_classifier(self, num_classes, global_pool='avg'):
         self.num_classes = num_classes
         self.global_pool = SelectAdaptivePool2d(pool_type=global_pool)
-        del self.fc
+        if num_classes:
-        self.fc = nn.Linear(self.num_features * self.global_pool.feat_mult(), num_classes) if num_classes else None
+            num_features = self.num_features * self.global_pool.feat_mult()
+            self.fc = nn.Linear(num_features, num_classes)
+        else:
+            self.fc = nn.Identity()
 
     def forward_features(self, x):
         x = self.conv1(x)