Hugging Face's logo

timm
/
mobilenetv4_conv_medium.e180_ad_r384_in12k

Image Classification
timm
PyTorch
Safetensors
Model card Files Community
rwightman HF staff commited on
Commit
c99e2c3
•
1 Parent(s): 6f503b6

Add model

Browse files
Files changed (4) hide show
  1. README.md +207 -0
  2. config.json +35 -0
  3. model.safetensors +3 -0
  4. pytorch_model.bin +3 -0
README.md ADDED
@@ -0,0 +1,207 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ ---
2
+ tags:
3
+ - image-classification
4
+ - timm
5
+ library_name: timm
6
+ license: apache-2.0
7
+ datasets:
8
+ - imagenet-12k
9
+ ---
10
+ # Model card for mobilenetv4_conv_medium.e180_ad_r384_in12k
11
+
12
+ A MobileNet-V4 image classification model. Trained on ImageNet-12k by Ross Wightman.
13
+
14
+ A collection of hparams (timm .yaml config files) for this training series can be found here: https://gist.github.com/rwightman/f6705cb65c03daeebca8aa129b1b94ad
15
+
16
+ ## Model Details
17
+ - **Model Type:** Image classification / feature backbone
18
+ - **Model Stats:**
19
+ - Params (M): 23.6
20
+ - GMACs: 2.5
21
+ - Activations (M): 17.1
22
+ - Image size: 384 x 384
23
+ - **Dataset:** ImageNet-12k
24
+ - **Papers:**
25
+ - MobileNetV4 -- Universal Models for the Mobile Ecosystem: https://arxiv.org/abs/2404.10518
26
+ - PyTorch Image Models: https://github.com/huggingface/pytorch-image-models
27
+
28
+ ## Model Usage
29
+ ### Image Classification
30
+ ```python
31
+ from urllib.request import urlopen
32
+ from PIL import Image
33
+ import timm
34
+
35
+ img = Image.open(urlopen(
36
+ 'https://huggingface.co/datasets/huggingface/documentation-images/resolve/main/beignets-task-guide.png'
37
+ ))
38
+
39
+ model = timm.create_model('mobilenetv4_conv_medium.e180_ad_r384_in12k', pretrained=True)
40
+ model = model.eval()
41
+
42
+ # get model specific transforms (normalization, resize)
43
+ data_config = timm.data.resolve_model_data_config(model)
44
+ transforms = timm.data.create_transform(**data_config, is_training=False)
45
+
46
+ output = model(transforms(img).unsqueeze(0)) # unsqueeze single image into batch of 1
47
+
48
+ top5_probabilities, top5_class_indices = torch.topk(output.softmax(dim=1) * 100, k=5)
49
+ ```
50
+
51
+ ### Feature Map Extraction
52
+ ```python
53
+ from urllib.request import urlopen
54
+ from PIL import Image
55
+ import timm
56
+
57
+ img = Image.open(urlopen(
58
+ 'https://huggingface.co/datasets/huggingface/documentation-images/resolve/main/beignets-task-guide.png'
59
+ ))
60
+
61
+ model = timm.create_model(
62
+ 'mobilenetv4_conv_medium.e180_ad_r384_in12k',
63
+ pretrained=True,
64
+ features_only=True,
65
+ )
66
+ model = model.eval()
67
+
68
+ # get model specific transforms (normalization, resize)
69
+ data_config = timm.data.resolve_model_data_config(model)
70
+ transforms = timm.data.create_transform(**data_config, is_training=False)
71
+
72
+ output = model(transforms(img).unsqueeze(0)) # unsqueeze single image into batch of 1
73
+
74
+ for o in output:
75
+ # print shape of each feature map in output
76
+ # e.g.:
77
+ # torch.Size([1, 32, 192, 192])
78
+ # torch.Size([1, 48, 96, 96])
79
+ # torch.Size([1, 80, 48, 48])
80
+ # torch.Size([1, 160, 24, 24])
81
+ # torch.Size([1, 960, 12, 12])
82
+
83
+ print(o.shape)
84
+ ```
85
+
86
+ ### Image Embeddings
87
+ ```python
88
+ from urllib.request import urlopen
89
+ from PIL import Image
90
+ import timm
91
+
92
+ img = Image.open(urlopen(
93
+ 'https://huggingface.co/datasets/huggingface/documentation-images/resolve/main/beignets-task-guide.png'
94
+ ))
95
+
96
+ model = timm.create_model(
97
+ 'mobilenetv4_conv_medium.e180_ad_r384_in12k',
98
+ pretrained=True,
99
+ num_classes=0, # remove classifier nn.Linear
100
+ )
101
+ model = model.eval()
102
+
103
+ # get model specific transforms (normalization, resize)
104
+ data_config = timm.data.resolve_model_data_config(model)
105
+ transforms = timm.data.create_transform(**data_config, is_training=False)
106
+
107
+ output = model(transforms(img).unsqueeze(0)) # output is (batch_size, num_features) shaped tensor
108
+
109
+ # or equivalently (without needing to set num_classes=0)
110
+
111
+ output = model.forward_features(transforms(img).unsqueeze(0))
112
+ # output is unpooled, a (1, 960, 12, 12) shaped tensor
113
+
114
+ output = model.forward_head(output, pre_logits=True)
115
+ # output is a (1, num_features) shaped tensor
116
+ ```
117
+
118
+ ## Model Comparison
119
+ ### By Top-1
120
+
121
+ | model | top1 | top5 | param_count | img_size |
122
+ |--------------------------------------------------------------------------------------------------------------------------|--------|--------|-------------|----------|
123
+ | [mobilenetv4_conv_aa_large.e230_r448_in12k_ft_in1k](http://hf.co/timm/mobilenetv4_conv_aa_large.e230_r448_in12k_ft_in1k) | 84.99 | 97.294 | 32.59 | 544 |
124
+ | [mobilenetv4_conv_aa_large.e230_r384_in12k_ft_in1k](http://hf.co/timm/mobilenetv4_conv_aa_large.e230_r384_in12k_ft_in1k) | 84.772 | 97.344 | 32.59 | 480 |
125
+ | [mobilenetv4_conv_aa_large.e230_r448_in12k_ft_in1k](http://hf.co/timm/mobilenetv4_conv_aa_large.e230_r448_in12k_ft_in1k) | 84.64 | 97.114 | 32.59 | 448 |
126
+ | [mobilenetv4_hybrid_large.ix_e600_r384_in1k](http://hf.co/timm/mobilenetv4_hybrid_large.ix_e600_r384_in1k) | 84.356 | 96.892 | 37.76 | 448 |
127
+ | [mobilenetv4_conv_aa_large.e230_r384_in12k_ft_in1k](http://hf.co/timm/mobilenetv4_conv_aa_large.e230_r384_in12k_ft_in1k) | 84.314 | 97.102 | 32.59 | 384 |
128
+ | [mobilenetv4_hybrid_large.e600_r384_in1k](http://hf.co/timm/mobilenetv4_hybrid_large.e600_r384_in1k) | 84.266 | 96.936 | 37.76 | 448 |
129
+ | [mobilenetv4_hybrid_large.ix_e600_r384_in1k](http://hf.co/timm/mobilenetv4_hybrid_large.ix_e600_r384_in1k) | 83.990 | 96.702 | 37.76 | 384 |
130
+ | [mobilenetv4_conv_aa_large.e600_r384_in1k](http://hf.co/timm/mobilenetv4_conv_aa_large.e600_r384_in1k) | 83.824 | 96.734 | 32.59 | 480 |
131
+ | [mobilenetv4_hybrid_large.e600_r384_in1k](http://hf.co/timm/mobilenetv4_hybrid_large.e600_r384_in1k) | 83.800 | 96.770 | 37.76 | 384 |
132
+ | [mobilenetv4_hybrid_medium.ix_e550_r384_in1k](http://hf.co/timm/mobilenetv4_hybrid_medium.ix_e550_r384_in1k) | 83.394 | 96.760 | 11.07 | 448 |
133
+ | [mobilenetv4_conv_large.e600_r384_in1k](http://hf.co/timm/mobilenetv4_conv_large.e600_r384_in1k) | 83.392 | 96.622 | 32.59 | 448 |
134
+ | [mobilenetv4_conv_aa_large.e600_r384_in1k](http://hf.co/timm/mobilenetv4_conv_aa_large.e600_r384_in1k) | 83.244 | 96.392 | 32.59 | 384 |
135
+ | [mobilenetv4_hybrid_medium.e200_r256_in12k_ft_in1k](http://hf.co/timm/mobilenetv4_hybrid_medium.e200_r256_in12k_ft_in1k) | 82.99 | 96.67 | 11.07 | 320 |
136
+ | [mobilenetv4_hybrid_medium.ix_e550_r384_in1k](http://hf.co/timm/mobilenetv4_hybrid_medium.ix_e550_r384_in1k) | 82.968 | 96.474 | 11.07 | 384 |
137
+ | [mobilenetv4_conv_large.e600_r384_in1k](http://hf.co/timm/mobilenetv4_conv_large.e600_r384_in1k) | 82.952 | 96.266 | 32.59 | 384 |
138
+ | [mobilenetv4_conv_large.e500_r256_in1k](http://hf.co/timm/mobilenetv4_conv_large.e500_r256_in1k) | 82.674 | 96.31 | 32.59 | 320 |
139
+ | [mobilenetv4_hybrid_medium.ix_e550_r256_in1k](http://hf.co/timm/mobilenetv4_hybrid_medium.ix_e550_r256_in1k) | 82.492 | 96.278 | 11.07 | 320 |
140
+ | [mobilenetv4_hybrid_medium.e200_r256_in12k_ft_in1k](http://hf.co/timm/mobilenetv4_hybrid_medium.e200_r256_in12k_ft_in1k) | 82.364 | 96.256 | 11.07 | 256 |
141
+ | [mobilenetv4_conv_large.e500_r256_in1k](http://hf.co/timm/mobilenetv4_conv_large.e500_r256_in1k) | 81.862 | 95.69 | 32.59 | 256 |
142
+ | [resnet50d.ra4_e3600_r224_in1k](http://hf.co/timm/resnet50d.ra4_e3600_r224_in1k) | 81.838 | 95.922 | 25.58 | 288 |
143
+ | [mobilenetv3_large_150d.ra4_e3600_r256_in1k](http://hf.co/timm/mobilenetv3_large_150d.ra4_e3600_r256_in1k) | 81.806 | 95.9 | 14.62 | 320 |
144
+ | [mobilenetv4_hybrid_medium.ix_e550_r256_in1k](http://hf.co/timm/mobilenetv4_hybrid_medium.ix_e550_r256_in1k) | 81.446 | 95.704 | 11.07 | 256 |
145
+ | [efficientnet_b1.ra4_e3600_r240_in1k](http://hf.co/timm/efficientnet_b1.ra4_e3600_r240_in1k) | 81.440 | 95.700 | 7.79 | 288 |
146
+ | [mobilenetv4_hybrid_medium.e500_r224_in1k](http://hf.co/timm/mobilenetv4_hybrid_medium.e500_r224_in1k) | 81.276 | 95.742 | 11.07 | 256 |
147
+ | [resnet50d.ra4_e3600_r224_in1k](http://hf.co/timm/resnet50d.ra4_e3600_r224_in1k) | 80.952 | 95.384 | 25.58 | 224 |
148
+ | [mobilenetv3_large_150d.ra4_e3600_r256_in1k](http://hf.co/timm/mobilenetv3_large_150d.ra4_e3600_r256_in1k) | 80.944 | 95.448 | 14.62 | 256 |
149
+ | [mobilenetv4_conv_medium.e500_r256_in1k](http://hf.co/timm/mobilenetv4_conv_medium.e500_r256_in1k) | 80.858 | 95.768 | 9.72 | 320 |
150
+ | [mobilenet_edgetpu_v2_m.ra4_e3600_r224_in1k](http://hf.co/timm/mobilenet_edgetpu_v2_m.ra4_e3600_r224_in1k) | 80.680 | 95.442 | 8.46 | 256 |
151
+ | [mobilenetv4_hybrid_medium.e500_r224_in1k](http://hf.co/timm/mobilenetv4_hybrid_medium.e500_r224_in1k) | 80.442 | 95.38 | 11.07 | 224 |
152
+ | [efficientnet_b1.ra4_e3600_r240_in1k](http://hf.co/timm/efficientnet_b1.ra4_e3600_r240_in1k) | 80.406 | 95.152 | 7.79 | 240 |
153
+ | [mobilenetv4_conv_blur_medium.e500_r224_in1k](http://hf.co/timm/mobilenetv4_conv_blur_medium.e500_r224_in1k) | 80.142 | 95.298 | 9.72 | 256 |
154
+ | [mobilenet_edgetpu_v2_m.ra4_e3600_r224_in1k](http://hf.co/timm/mobilenet_edgetpu_v2_m.ra4_e3600_r224_in1k) | 80.130 | 95.002 | 8.46 | 224 |
155
+ | [mobilenetv4_conv_medium.e500_r256_in1k](http://hf.co/timm/mobilenetv4_conv_medium.e500_r256_in1k) | 79.928 | 95.184 | 9.72 | 256 |
156
+ | [mobilenetv4_conv_medium.e500_r224_in1k](http://hf.co/timm/mobilenetv4_conv_medium.e500_r224_in1k) | 79.808 | 95.186 | 9.72 | 256 |
157
+ | [resnetv2_34d.ra4_e3600_r224_in1k](http://hf.co/timm/resnetv2_34d.ra4_e3600_r224_in1k) | 79.590 | 94.770 | 21.82 | 288 |
158
+ | [mobilenetv4_conv_blur_medium.e500_r224_in1k](http://hf.co/timm/mobilenetv4_conv_blur_medium.e500_r224_in1k) | 79.438 | 94.932 | 9.72 | 224 |
159
+ | [efficientnet_b0.ra4_e3600_r224_in1k](http://hf.co/timm/efficientnet_b0.ra4_e3600_r224_in1k) | 79.364 | 94.754 | 5.29 | 256 |
160
+ | [mobilenetv4_conv_medium.e500_r224_in1k](http://hf.co/timm/mobilenetv4_conv_medium.e500_r224_in1k) | 79.094 | 94.77 | 9.72 | 224 |
161
+ | [resnetv2_34.ra4_e3600_r224_in1k](http://hf.co/timm/resnetv2_34.ra4_e3600_r224_in1k) | 79.072 | 94.566 | 21.80 | 288 |
162
+ | [resnet34.ra4_e3600_r224_in1k](http://hf.co/timm/resnet34.ra4_e3600_r224_in1k) | 78.952 | 94.450 | 21.80 | 288 |
163
+ | [efficientnet_b0.ra4_e3600_r224_in1k](http://hf.co/timm/efficientnet_b0.ra4_e3600_r224_in1k) | 78.584 | 94.338 | 5.29 | 224 |
164
+ | [resnetv2_34d.ra4_e3600_r224_in1k](http://hf.co/timm/resnetv2_34d.ra4_e3600_r224_in1k) | 78.268 | 93.952 | 21.82 | 224 |
165
+ | [resnetv2_34.ra4_e3600_r224_in1k](http://hf.co/timm/resnetv2_34.ra4_e3600_r224_in1k) | 77.636 | 93.528 | 21.80 | 224 |
166
+ | [mobilenetv1_125.ra4_e3600_r224_in1k](http://hf.co/timm/mobilenetv1_125.ra4_e3600_r224_in1k) | 77.600 | 93.804 | 6.27 | 256 |
167
+ | [resnet34.ra4_e3600_r224_in1k](http://hf.co/timm/resnet34.ra4_e3600_r224_in1k) | 77.448 | 93.502 | 21.80 | 224 |
168
+ | [mobilenetv3_large_100.ra4_e3600_r224_in1k](http://hf.co/timm/mobilenetv3_large_100.ra4_e3600_r224_in1k) | 77.164 | 93.336 | 5.48 | 256 |
169
+ | [mobilenetv1_125.ra4_e3600_r224_in1k](http://hf.co/timm/mobilenetv1_125.ra4_e3600_r224_in1k) | 76.924 | 93.234 | 6.27 | 224 |
170
+ | [mobilenetv1_100h.ra4_e3600_r224_in1k](http://hf.co/timm/mobilenetv1_100h.ra4_e3600_r224_in1k) | 76.596 | 93.272 | 5.28 | 256 |
171
+ | [mobilenetv3_large_100.ra4_e3600_r224_in1k](http://hf.co/timm/mobilenetv3_large_100.ra4_e3600_r224_in1k) | 76.310 | 92.846 | 5.48 | 224 |
172
+ | [mobilenetv1_100.ra4_e3600_r224_in1k](http://hf.co/timm/mobilenetv1_100.ra4_e3600_r224_in1k) | 76.094 | 93.004 | 4.23 | 256 |
173
+ | [resnetv2_18d.ra4_e3600_r224_in1k](http://hf.co/timm/resnetv2_18d.ra4_e3600_r224_in1k) | 76.044 | 93.020 | 11.71 | 288 |
174
+ | [resnet18d.ra4_e3600_r224_in1k](http://hf.co/timm/resnet18d.ra4_e3600_r224_in1k) | 76.024 | 92.780 | 11.71 | 288 |
175
+ | [mobilenetv1_100h.ra4_e3600_r224_in1k](http://hf.co/timm/mobilenetv1_100h.ra4_e3600_r224_in1k) | 75.662 | 92.504 | 5.28 | 224 |
176
+ | [mobilenetv1_100.ra4_e3600_r224_in1k](http://hf.co/timm/mobilenetv1_100.ra4_e3600_r224_in1k) | 75.382 | 92.312 | 4.23 | 224 |
177
+ | [resnetv2_18.ra4_e3600_r224_in1k](http://hf.co/timm/resnetv2_18.ra4_e3600_r224_in1k) | 75.340 | 92.678 | 11.69 | 288 |
178
+ | [mobilenetv4_conv_small.e2400_r224_in1k](http://hf.co/timm/mobilenetv4_conv_small.e2400_r224_in1k) | 74.616 | 92.072 | 3.77 | 256 |
179
+ | [resnetv2_18d.ra4_e3600_r224_in1k](http://hf.co/timm/resnetv2_18d.ra4_e3600_r224_in1k) | 74.412 | 91.936 | 11.71 | 224 |
180
+ | [resnet18d.ra4_e3600_r224_in1k](http://hf.co/timm/resnet18d.ra4_e3600_r224_in1k) | 74.322 | 91.832 | 11.71 | 224 |
181
+ | [mobilenetv4_conv_small.e1200_r224_in1k](http://hf.co/timm/mobilenetv4_conv_small.e1200_r224_in1k) | 74.292 | 92.116 | 3.77 | 256 |
182
+ | [mobilenetv4_conv_small.e2400_r224_in1k](http://hf.co/timm/mobilenetv4_conv_small.e2400_r224_in1k) | 73.756 | 91.422 | 3.77 | 224 |
183
+ | [resnetv2_18.ra4_e3600_r224_in1k](http://hf.co/timm/resnetv2_18.ra4_e3600_r224_in1k) | 73.578 | 91.352 | 11.69 | 224 |
184
+ | [mobilenetv4_conv_small.e1200_r224_in1k](http://hf.co/timm/mobilenetv4_conv_small.e1200_r224_in1k) | 73.454 | 91.34 | 3.77 | 224 |
185
+ | [mobilenetv4_conv_small_050.e3000_r224_in1k](http://hf.co/timm/mobilenetv4_conv_small_050.e3000_r224_in1k) | 65.810 | 86.424 | 2.24 | 256 |
186
+ | [mobilenetv4_conv_small_050.e3000_r224_in1k](http://hf.co/timm/mobilenetv4_conv_small_050.e3000_r224_in1k) | 64.762 | 85.514 | 2.24 | 224 |
187
+
188
+ ## Citation
189
+ ```bibtex
190
+ @article{qin2024mobilenetv4,
191
+ title={MobileNetV4-Universal Models for the Mobile Ecosystem},
192
+ author={Qin, Danfeng and Leichner, Chas and Delakis, Manolis and Fornoni, Marco and Luo, Shixin and Yang, Fan and Wang, Weijun and Banbury, Colby and Ye, Chengxi and Akin, Berkin and others},
193
+ journal={arXiv preprint arXiv:2404.10518},
194
+ year={2024}
195
+ }
196
+ ```
197
+ ```bibtex
198
+ @misc{rw2019timm,
199
+ author = {Ross Wightman},
200
+ title = {PyTorch Image Models},
201
+ year = {2019},
202
+ publisher = {GitHub},
203
+ journal = {GitHub repository},
204
+ doi = {10.5281/zenodo.4414861},
205
+ howpublished = {\url{https://github.com/huggingface/pytorch-image-models}}
206
+ }
207
+ ```
config.json ADDED
@@ -0,0 +1,35 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "architecture": "mobilenetv4_conv_medium",
3
+ "num_classes": 11821,
4
+ "num_features": 960,
5
+ "pretrained_cfg": {
6
+ "tag": "e180_ad_r384_in12k",
7
+ "custom_load": false,
8
+ "input_size": [
9
+ 3,
10
+ 384,
11
+ 384
12
+ ],
13
+ "fixed_input_size": false,
14
+ "interpolation": "bicubic",
15
+ "crop_pct": 0.95,
16
+ "crop_mode": "center",
17
+ "mean": [
18
+ 0.485,
19
+ 0.456,
20
+ 0.406
21
+ ],
22
+ "std": [
23
+ 0.229,
24
+ 0.224,
25
+ 0.225
26
+ ],
27
+ "num_classes": 11821,
28
+ "pool_size": [
29
+ 12,
30
+ 12
31
+ ],
32
+ "first_conv": "conv_stem",
33
+ "classifier": "classifier"
34
+ }
35
+ }
model.safetensors ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:43c3022f9f4047c47e8db559c8bd8da5812eeab0391c888e25b92a09f088a1cf
3
+ size 94625988
pytorch_model.bin ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:842ed66bd2d1d6e33f422e3794e58bc6c656950bf28be52e236db532b1da89f5
3
+ size 94747466