timm
/

Image Classification
timm
PyTorch
Safetensors
rwightman HF staff commited on
Commit
507d04d
1 Parent(s): 49a71d2
Files changed (4) hide show
  1. README.md +162 -0
  2. config.json +40 -0
  3. model.safetensors +3 -0
  4. pytorch_model.bin +3 -0
README.md ADDED
@@ -0,0 +1,162 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ ---
2
+ tags:
3
+ - image-classification
4
+ - timm
5
+ library_tag: timm
6
+ license: apache-2.0
7
+ datasets:
8
+ - imagenet-1k
9
+ ---
10
+ # Model card for resnetv2_50d_evos.ah_in1k
11
+
12
+ A ResNet-V2 (pre-activation ResNet) image classification model. Trained on ImageNet-1k by Ross Wightman in `timm` using ResNet strikes back (RSB) `A1` based recipe.
13
+
14
+ This model uses:
15
+ * A 3x3 3-layer stem, avg-pool in shortcut downsample.
16
+ * EvoNorm-S0 normalization-activation layers instead of Batch Normalization with ReLU activations.
17
+
18
+
19
+ ## Model Details
20
+ - **Model Type:** Image classification / feature backbone
21
+ - **Model Stats:**
22
+ - Params (M): 25.6
23
+ - GMACs: 4.3
24
+ - Activations (M): 11.9
25
+ - Image size: train = 224 x 224, test = 288 x 288
26
+ - **Papers:**
27
+ - ResNet strikes back: An improved training procedure in timm: https://arxiv.org/abs/2110.00476
28
+ - Identity Mappings in Deep Residual Networks: https://arxiv.org/abs/1603.05027
29
+ - Evolving Normalization-Activation Layers: https://arxiv.org/abs/2004.02967
30
+ - **Dataset:** ImageNet-1k
31
+ - **Original:** https://github.com/huggingface/pytorch-image-models
32
+
33
+ ## Model Usage
34
+ ### Image Classification
35
+ ```python
36
+ from urllib.request import urlopen
37
+ from PIL import Image
38
+ import timm
39
+
40
+ img = Image.open(urlopen(
41
+ 'https://huggingface.co/datasets/huggingface/documentation-images/resolve/main/beignets-task-guide.png'
42
+ ))
43
+
44
+ model = timm.create_model('resnetv2_50d_evos.ah_in1k', pretrained=True)
45
+ model = model.eval()
46
+
47
+ # get model specific transforms (normalization, resize)
48
+ data_config = timm.data.resolve_model_data_config(model)
49
+ transforms = timm.data.create_transform(**data_config, is_training=False)
50
+
51
+ output = model(transforms(img).unsqueeze(0)) # unsqueeze single image into batch of 1
52
+
53
+ top5_probabilities, top5_class_indices = torch.topk(output.softmax(dim=1) * 100, k=5)
54
+ ```
55
+
56
+ ### Feature Map Extraction
57
+ ```python
58
+ from urllib.request import urlopen
59
+ from PIL import Image
60
+ import timm
61
+
62
+ img = Image.open(urlopen(
63
+ 'https://huggingface.co/datasets/huggingface/documentation-images/resolve/main/beignets-task-guide.png'
64
+ ))
65
+
66
+ model = timm.create_model(
67
+ 'resnetv2_50d_evos.ah_in1k',
68
+ pretrained=True,
69
+ features_only=True,
70
+ )
71
+ model = model.eval()
72
+
73
+ # get model specific transforms (normalization, resize)
74
+ data_config = timm.data.resolve_model_data_config(model)
75
+ transforms = timm.data.create_transform(**data_config, is_training=False)
76
+
77
+ output = model(transforms(img).unsqueeze(0)) # unsqueeze single image into batch of 1
78
+
79
+ for o in output:
80
+ # print shape of each feature map in output
81
+ # e.g.:
82
+ # torch.Size([1, 64, 112, 112])
83
+ # torch.Size([1, 256, 56, 56])
84
+ # torch.Size([1, 512, 28, 28])
85
+ # torch.Size([1, 1024, 14, 14])
86
+ # torch.Size([1, 2048, 7, 7])
87
+
88
+ print(o.shape)
89
+ ```
90
+
91
+ ### Image Embeddings
92
+ ```python
93
+ from urllib.request import urlopen
94
+ from PIL import Image
95
+ import timm
96
+
97
+ img = Image.open(urlopen(
98
+ 'https://huggingface.co/datasets/huggingface/documentation-images/resolve/main/beignets-task-guide.png'
99
+ ))
100
+
101
+ model = timm.create_model(
102
+ 'resnetv2_50d_evos.ah_in1k',
103
+ pretrained=True,
104
+ num_classes=0, # remove classifier nn.Linear
105
+ )
106
+ model = model.eval()
107
+
108
+ # get model specific transforms (normalization, resize)
109
+ data_config = timm.data.resolve_model_data_config(model)
110
+ transforms = timm.data.create_transform(**data_config, is_training=False)
111
+
112
+ output = model(transforms(img).unsqueeze(0)) # output is (batch_size, num_features) shaped tensor
113
+
114
+ # or equivalently (without needing to set num_classes=0)
115
+
116
+ output = model.forward_features(transforms(img).unsqueeze(0))
117
+ # output is unpooled, a (1, 2048, 7, 7) shaped tensor
118
+
119
+ output = model.forward_head(output, pre_logits=True)
120
+ # output is a (1, num_features) shaped tensor
121
+ ```
122
+
123
+ ## Model Comparison
124
+ Explore the dataset and runtime metrics of this model in timm [model results](https://github.com/huggingface/pytorch-image-models/tree/main/results).
125
+
126
+ ## Citation
127
+ ```bibtex
128
+ @inproceedings{wightman2021resnet,
129
+ title={ResNet strikes back: An improved training procedure in timm},
130
+ author={Wightman, Ross and Touvron, Hugo and Jegou, Herve},
131
+ booktitle={NeurIPS 2021 Workshop on ImageNet: Past, Present, and Future}
132
+ }
133
+ ```
134
+ ```bibtex
135
+ @article{He2016,
136
+ author = {Kaiming He and Xiangyu Zhang and Shaoqing Ren and Jian Sun},
137
+ title = {Identity Mappings in Deep Residual Networks},
138
+ journal = {arXiv preprint arXiv:1603.05027},
139
+ year = {2016}
140
+ }
141
+ ```
142
+ ```bibtex
143
+ @article{liu2020evolving,
144
+ title={Evolving normalization-activation layers},
145
+ author={Liu, Hanxiao and Brock, Andy and Simonyan, Karen and Le, Quoc},
146
+ journal={Advances in Neural Information Processing Systems},
147
+ volume={33},
148
+ pages={13539--13550},
149
+ year={2020}
150
+ }
151
+ ```
152
+ ```bibtex
153
+ @misc{rw2019timm,
154
+ author = {Ross Wightman},
155
+ title = {PyTorch Image Models},
156
+ year = {2019},
157
+ publisher = {GitHub},
158
+ journal = {GitHub repository},
159
+ doi = {10.5281/zenodo.4414861},
160
+ howpublished = {\url{https://github.com/huggingface/pytorch-image-models}}
161
+ }
162
+ ```
config.json ADDED
@@ -0,0 +1,40 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "architecture": "resnetv2_50d_evos",
3
+ "num_classes": 1000,
4
+ "num_features": 2048,
5
+ "pretrained_cfg": {
6
+ "tag": "ah_in1k",
7
+ "custom_load": false,
8
+ "input_size": [
9
+ 3,
10
+ 224,
11
+ 224
12
+ ],
13
+ "test_input_size": [
14
+ 3,
15
+ 288,
16
+ 288
17
+ ],
18
+ "fixed_input_size": false,
19
+ "interpolation": "bicubic",
20
+ "crop_pct": 0.95,
21
+ "crop_mode": "center",
22
+ "mean": [
23
+ 0.5,
24
+ 0.5,
25
+ 0.5
26
+ ],
27
+ "std": [
28
+ 0.5,
29
+ 0.5,
30
+ 0.5
31
+ ],
32
+ "num_classes": 1000,
33
+ "pool_size": [
34
+ 7,
35
+ 7
36
+ ],
37
+ "first_conv": "stem.conv1",
38
+ "classifier": "head.fc"
39
+ }
40
+ }
model.safetensors ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:56a087097bd0e863ceb3154e91fc7787517efa4818f36c25ca6698ee370d7560
3
+ size 102385924
pytorch_model.bin ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:712196929a71a2107a7fbaf4a2e45df75f9ad0ef47ee8c6dac7247196ff56039
3
+ size 102438165