Hugging Face's logo

zeromodels
/
tf_efficientnet_b5_in1k

Image Classification
Keras
PyTorch
JAX
TensorFlow
zeromodels
efficientnet
backbone
Model card Files
xet
 Community
IMvision12 commited on
Commit
7ab3386
·
verified ·
1 Parent(s): 7611562

Migrate to zeromodels (rename kf_*.json -> zm_*.json, fix refs in config + README, ensure tag + badge)

Browse files
Files changed (2) hide show
  1. README.md +45 -45
  2. kf_config.json → zm_config.json +18 -18
README.md CHANGED
@@ -2,10 +2,10 @@
2
  pipeline_tag: image-classification
3
  license: apache-2.0
4
  base_model: timm/tf_efficientnet_b5.in1k
5
- library_name: kerasformers
6
  tags:
7
  - keras
8
- - kerasformers
9
  - image-classification
10
  - efficientnet
11
  - backbone
@@ -15,13 +15,13 @@ tags:
15
  - tf
16
  ---
17
 
18
- ## ***See [our collection](https://huggingface.co/collections/kerasformers/efficientnet-6a6d0d5e9b3756eaaca7dbe4) for all versions of EfficientNet.***
19
 
20
  # Run EfficientNet with Keras 3: JAX, PyTorch, or TensorFlow
21
 
22
- [![GitHub](https://img.shields.io/badge/GitHub-KerasFormers-black?logo=github)](https://github.com/IMvision12/KerasFormers) [![Docs](https://img.shields.io/badge/Docs-EfficientNet-blue)](https://imvision12.github.io/KerasFormers/classification_backbones/) [![Collection](https://img.shields.io/badge/HF-EfficientNet%20collection-yellow)](https://huggingface.co/collections/kerasformers/efficientnet-6a6d0d5e9b3756eaaca7dbe4)
23
 
24
- # kerasformers/tf_efficientnet_b5_in1k
25
 
26
  Paper: [EfficientNet: Rethinking Model Scaling for Convolutional Neural Networks (arXiv:1905.11946)](https://arxiv.org/abs/1905.11946) · [HF Papers](https://huggingface.co/papers/1905.11946)
27
 
@@ -29,7 +29,7 @@ EfficientNet compound-scales depth/width/resolution for strong accuracy/efficien
29
 
30
  For more details on the model, please go to the upstream [model card](https://huggingface.co/timm/tf_efficientnet_b5.in1k).
31
 
32
- Pure-**Keras 3** conversion of [`timm/tf_efficientnet_b5.in1k`](https://huggingface.co/timm/tf_efficientnet_b5.in1k) for [kerasformers](https://github.com/IMvision12/KerasFormers). One implementation runs unmodified on **TensorFlow / Torch / JAX**.
33
 
34
  This is an **image-classification / backbone** checkpoint (`EfficientNetImageClassify` / `EfficientNetModel`).
35
 
@@ -41,11 +41,11 @@ os.environ["KERAS_BACKEND"] = "torch" # or "jax" / "tensorflow"
41
 
42
  from PIL import Image
43
  import numpy as np
44
- from kerasformers.models.efficientnet import EfficientNetImageClassify, EfficientNetModel
45
 
46
- model = EfficientNetImageClassify.from_weights("kerasformers/tf_efficientnet_b5_in1k")
47
  backbone = EfficientNetModel.from_weights(
48
- "kerasformers/tf_efficientnet_b5_in1k", as_backbone=True
49
  )
50
 
51
  image = Image.open("your_image.jpg").convert("RGB")
@@ -56,49 +56,49 @@ feats = backbone(x)
56
  print(len(feats), [tuple(f.shape) for f in feats])
57
  ```
58
 
59
- Load any EfficientNet variant the same way with `from_weights("kerasformers/<variant>")`:
60
 
61
  | Variant | Hub |
62
  |---|---|
63
- | `tf_efficientnet_b0_aa_in1k` | [`kerasformers/tf_efficientnet_b0_aa_in1k`](https://huggingface.co/kerasformers/tf_efficientnet_b0_aa_in1k) |
64
- | `tf_efficientnet_b0_ap_in1k` | [`kerasformers/tf_efficientnet_b0_ap_in1k`](https://huggingface.co/kerasformers/tf_efficientnet_b0_ap_in1k) |
65
- | `tf_efficientnet_b0_in1k` | [`kerasformers/tf_efficientnet_b0_in1k`](https://huggingface.co/kerasformers/tf_efficientnet_b0_in1k) |
66
- | `tf_efficientnet_b0_ns_jft_in1k` | [`kerasformers/tf_efficientnet_b0_ns_jft_in1k`](https://huggingface.co/kerasformers/tf_efficientnet_b0_ns_jft_in1k) |
67
- | `tf_efficientnet_b1_aa_in1k` | [`kerasformers/tf_efficientnet_b1_aa_in1k`](https://huggingface.co/kerasformers/tf_efficientnet_b1_aa_in1k) |
68
- | `tf_efficientnet_b1_ap_in1k` | [`kerasformers/tf_efficientnet_b1_ap_in1k`](https://huggingface.co/kerasformers/tf_efficientnet_b1_ap_in1k) |
69
- | `tf_efficientnet_b1_in1k` | [`kerasformers/tf_efficientnet_b1_in1k`](https://huggingface.co/kerasformers/tf_efficientnet_b1_in1k) |
70
- | `tf_efficientnet_b1_ns_jft_in1k` | [`kerasformers/tf_efficientnet_b1_ns_jft_in1k`](https://huggingface.co/kerasformers/tf_efficientnet_b1_ns_jft_in1k) |
71
- | `tf_efficientnet_b2_aa_in1k` | [`kerasformers/tf_efficientnet_b2_aa_in1k`](https://huggingface.co/kerasformers/tf_efficientnet_b2_aa_in1k) |
72
- | `tf_efficientnet_b2_ap_in1k` | [`kerasformers/tf_efficientnet_b2_ap_in1k`](https://huggingface.co/kerasformers/tf_efficientnet_b2_ap_in1k) |
73
- | `tf_efficientnet_b2_in1k` | [`kerasformers/tf_efficientnet_b2_in1k`](https://huggingface.co/kerasformers/tf_efficientnet_b2_in1k) |
74
- | `tf_efficientnet_b2_ns_jft_in1k` | [`kerasformers/tf_efficientnet_b2_ns_jft_in1k`](https://huggingface.co/kerasformers/tf_efficientnet_b2_ns_jft_in1k) |
75
- | `tf_efficientnet_b3_aa_in1k` | [`kerasformers/tf_efficientnet_b3_aa_in1k`](https://huggingface.co/kerasformers/tf_efficientnet_b3_aa_in1k) |
76
- | `tf_efficientnet_b3_ap_in1k` | [`kerasformers/tf_efficientnet_b3_ap_in1k`](https://huggingface.co/kerasformers/tf_efficientnet_b3_ap_in1k) |
77
- | `tf_efficientnet_b3_in1k` | [`kerasformers/tf_efficientnet_b3_in1k`](https://huggingface.co/kerasformers/tf_efficientnet_b3_in1k) |
78
- | `tf_efficientnet_b3_ns_jft_in1k` | [`kerasformers/tf_efficientnet_b3_ns_jft_in1k`](https://huggingface.co/kerasformers/tf_efficientnet_b3_ns_jft_in1k) |
79
- | `tf_efficientnet_b4_aa_in1k` | [`kerasformers/tf_efficientnet_b4_aa_in1k`](https://huggingface.co/kerasformers/tf_efficientnet_b4_aa_in1k) |
80
- | `tf_efficientnet_b4_ap_in1k` | [`kerasformers/tf_efficientnet_b4_ap_in1k`](https://huggingface.co/kerasformers/tf_efficientnet_b4_ap_in1k) |
81
- | `tf_efficientnet_b4_in1k` | [`kerasformers/tf_efficientnet_b4_in1k`](https://huggingface.co/kerasformers/tf_efficientnet_b4_in1k) |
82
- | `tf_efficientnet_b4_ns_jft_in1k` | [`kerasformers/tf_efficientnet_b4_ns_jft_in1k`](https://huggingface.co/kerasformers/tf_efficientnet_b4_ns_jft_in1k) |
83
- | `tf_efficientnet_b5_aa_in1k` | [`kerasformers/tf_efficientnet_b5_aa_in1k`](https://huggingface.co/kerasformers/tf_efficientnet_b5_aa_in1k) |
84
- | `tf_efficientnet_b5_ap_in1k` | [`kerasformers/tf_efficientnet_b5_ap_in1k`](https://huggingface.co/kerasformers/tf_efficientnet_b5_ap_in1k) |
85
- | `tf_efficientnet_b5_in1k` | [`kerasformers/tf_efficientnet_b5_in1k`](https://huggingface.co/kerasformers/tf_efficientnet_b5_in1k) |
86
- | `tf_efficientnet_b5_ns_jft_in1k` | [`kerasformers/tf_efficientnet_b5_ns_jft_in1k`](https://huggingface.co/kerasformers/tf_efficientnet_b5_ns_jft_in1k) |
87
- | `tf_efficientnet_b6_aa_in1k` | [`kerasformers/tf_efficientnet_b6_aa_in1k`](https://huggingface.co/kerasformers/tf_efficientnet_b6_aa_in1k) |
88
- | `tf_efficientnet_b6_ap_in1k` | [`kerasformers/tf_efficientnet_b6_ap_in1k`](https://huggingface.co/kerasformers/tf_efficientnet_b6_ap_in1k) |
89
- | `tf_efficientnet_b6_ns_jft_in1k` | [`kerasformers/tf_efficientnet_b6_ns_jft_in1k`](https://huggingface.co/kerasformers/tf_efficientnet_b6_ns_jft_in1k) |
90
- | `tf_efficientnet_b7_aa_in1k` | [`kerasformers/tf_efficientnet_b7_aa_in1k`](https://huggingface.co/kerasformers/tf_efficientnet_b7_aa_in1k) |
91
- | `tf_efficientnet_b7_ap_in1k` | [`kerasformers/tf_efficientnet_b7_ap_in1k`](https://huggingface.co/kerasformers/tf_efficientnet_b7_ap_in1k) |
92
- | `tf_efficientnet_b7_ns_jft_in1k` | [`kerasformers/tf_efficientnet_b7_ns_jft_in1k`](https://huggingface.co/kerasformers/tf_efficientnet_b7_ns_jft_in1k) |
93
- | `tf_efficientnet_b8_ap_in1k` | [`kerasformers/tf_efficientnet_b8_ap_in1k`](https://huggingface.co/kerasformers/tf_efficientnet_b8_ap_in1k) |
94
- | `tf_efficientnet_l2_ns_jft_in1k` | [`kerasformers/tf_efficientnet_l2_ns_jft_in1k`](https://huggingface.co/kerasformers/tf_efficientnet_l2_ns_jft_in1k) |
95
- | `tf_efficientnet_l2_ns_jft_in1k_475` | [`kerasformers/tf_efficientnet_l2_ns_jft_in1k_475`](https://huggingface.co/kerasformers/tf_efficientnet_l2_ns_jft_in1k_475) |
96
 
97
  ## Tips
98
 
99
- - Set `KERAS_BACKEND` **before** importing Keras / kerasformers.
100
  - `EfficientNetImageClassify` returns class logits; `EfficientNetModel` returns features (`as_backbone=True` for multi-scale stages).
101
- - See [docs](https://imvision12.github.io/KerasFormers/classification_backbones/) and [Loading Weights](https://imvision12.github.io/KerasFormers/loading_weights/).
102
  - Upstream / timm checkpoints: `EfficientNetImageClassify.from_weights("hf:timm/tf_efficientnet_b5.in1k")`.
103
 
104
  ## Special Thanks
 
2
  pipeline_tag: image-classification
3
  license: apache-2.0
4
  base_model: timm/tf_efficientnet_b5.in1k
5
+ library_name: zeromodels
6
  tags:
7
  - keras
8
+ - zeromodels
9
  - image-classification
10
  - efficientnet
11
  - backbone
 
15
  - tf
16
  ---
17
 
18
+ ## ***See [our collection](https://huggingface.co/collections/zeromodels/efficientnet-6a6d0d5e9b3756eaaca7dbe4) for all versions of EfficientNet.***
19
 
20
  # Run EfficientNet with Keras 3: JAX, PyTorch, or TensorFlow
21
 
22
+ [![GitHub](https://img.shields.io/badge/GitHub-ZeroModels-black?logo=github)](https://github.com/IMvision12/ZeroModels) [![Docs](https://img.shields.io/badge/Docs-EfficientNet-blue)](https://imvision12.github.io/ZeroModels/classification_backbones/) [![Collection](https://img.shields.io/badge/HF-EfficientNet%20collection-yellow)](https://huggingface.co/collections/zeromodels/efficientnet-6a6d0d5e9b3756eaaca7dbe4)
23
 
24
+ # zeromodels/tf_efficientnet_b5_in1k
25
 
26
  Paper: [EfficientNet: Rethinking Model Scaling for Convolutional Neural Networks (arXiv:1905.11946)](https://arxiv.org/abs/1905.11946) · [HF Papers](https://huggingface.co/papers/1905.11946)
27
 
 
29
 
30
  For more details on the model, please go to the upstream [model card](https://huggingface.co/timm/tf_efficientnet_b5.in1k).
31
 
32
+ Pure-**Keras 3** conversion of [`timm/tf_efficientnet_b5.in1k`](https://huggingface.co/timm/tf_efficientnet_b5.in1k) for [zeromodels](https://github.com/IMvision12/ZeroModels). One implementation runs unmodified on **TensorFlow / Torch / JAX**.
33
 
34
  This is an **image-classification / backbone** checkpoint (`EfficientNetImageClassify` / `EfficientNetModel`).
35
 
 
41
 
42
  from PIL import Image
43
  import numpy as np
44
+ from zeromodels.models.efficientnet import EfficientNetImageClassify, EfficientNetModel
45
 
46
+ model = EfficientNetImageClassify.from_weights("zeromodels/tf_efficientnet_b5_in1k")
47
  backbone = EfficientNetModel.from_weights(
48
+ "zeromodels/tf_efficientnet_b5_in1k", as_backbone=True
49
  )
50
 
51
  image = Image.open("your_image.jpg").convert("RGB")
 
56
  print(len(feats), [tuple(f.shape) for f in feats])
57
  ```
58
 
59
+ Load any EfficientNet variant the same way with `from_weights("zeromodels/<variant>")`:
60
 
61
  | Variant | Hub |
62
  |---|---|
63
+ | `tf_efficientnet_b0_aa_in1k` | [`zeromodels/tf_efficientnet_b0_aa_in1k`](https://huggingface.co/zeromodels/tf_efficientnet_b0_aa_in1k) |
64
+ | `tf_efficientnet_b0_ap_in1k` | [`zeromodels/tf_efficientnet_b0_ap_in1k`](https://huggingface.co/zeromodels/tf_efficientnet_b0_ap_in1k) |
65
+ | `tf_efficientnet_b0_in1k` | [`zeromodels/tf_efficientnet_b0_in1k`](https://huggingface.co/zeromodels/tf_efficientnet_b0_in1k) |
66
+ | `tf_efficientnet_b0_ns_jft_in1k` | [`zeromodels/tf_efficientnet_b0_ns_jft_in1k`](https://huggingface.co/zeromodels/tf_efficientnet_b0_ns_jft_in1k) |
67
+ | `tf_efficientnet_b1_aa_in1k` | [`zeromodels/tf_efficientnet_b1_aa_in1k`](https://huggingface.co/zeromodels/tf_efficientnet_b1_aa_in1k) |
68
+ | `tf_efficientnet_b1_ap_in1k` | [`zeromodels/tf_efficientnet_b1_ap_in1k`](https://huggingface.co/zeromodels/tf_efficientnet_b1_ap_in1k) |
69
+ | `tf_efficientnet_b1_in1k` | [`zeromodels/tf_efficientnet_b1_in1k`](https://huggingface.co/zeromodels/tf_efficientnet_b1_in1k) |
70
+ | `tf_efficientnet_b1_ns_jft_in1k` | [`zeromodels/tf_efficientnet_b1_ns_jft_in1k`](https://huggingface.co/zeromodels/tf_efficientnet_b1_ns_jft_in1k) |
71
+ | `tf_efficientnet_b2_aa_in1k` | [`zeromodels/tf_efficientnet_b2_aa_in1k`](https://huggingface.co/zeromodels/tf_efficientnet_b2_aa_in1k) |
72
+ | `tf_efficientnet_b2_ap_in1k` | [`zeromodels/tf_efficientnet_b2_ap_in1k`](https://huggingface.co/zeromodels/tf_efficientnet_b2_ap_in1k) |
73
+ | `tf_efficientnet_b2_in1k` | [`zeromodels/tf_efficientnet_b2_in1k`](https://huggingface.co/zeromodels/tf_efficientnet_b2_in1k) |
74
+ | `tf_efficientnet_b2_ns_jft_in1k` | [`zeromodels/tf_efficientnet_b2_ns_jft_in1k`](https://huggingface.co/zeromodels/tf_efficientnet_b2_ns_jft_in1k) |
75
+ | `tf_efficientnet_b3_aa_in1k` | [`zeromodels/tf_efficientnet_b3_aa_in1k`](https://huggingface.co/zeromodels/tf_efficientnet_b3_aa_in1k) |
76
+ | `tf_efficientnet_b3_ap_in1k` | [`zeromodels/tf_efficientnet_b3_ap_in1k`](https://huggingface.co/zeromodels/tf_efficientnet_b3_ap_in1k) |
77
+ | `tf_efficientnet_b3_in1k` | [`zeromodels/tf_efficientnet_b3_in1k`](https://huggingface.co/zeromodels/tf_efficientnet_b3_in1k) |
78
+ | `tf_efficientnet_b3_ns_jft_in1k` | [`zeromodels/tf_efficientnet_b3_ns_jft_in1k`](https://huggingface.co/zeromodels/tf_efficientnet_b3_ns_jft_in1k) |
79
+ | `tf_efficientnet_b4_aa_in1k` | [`zeromodels/tf_efficientnet_b4_aa_in1k`](https://huggingface.co/zeromodels/tf_efficientnet_b4_aa_in1k) |
80
+ | `tf_efficientnet_b4_ap_in1k` | [`zeromodels/tf_efficientnet_b4_ap_in1k`](https://huggingface.co/zeromodels/tf_efficientnet_b4_ap_in1k) |
81
+ | `tf_efficientnet_b4_in1k` | [`zeromodels/tf_efficientnet_b4_in1k`](https://huggingface.co/zeromodels/tf_efficientnet_b4_in1k) |
82
+ | `tf_efficientnet_b4_ns_jft_in1k` | [`zeromodels/tf_efficientnet_b4_ns_jft_in1k`](https://huggingface.co/zeromodels/tf_efficientnet_b4_ns_jft_in1k) |
83
+ | `tf_efficientnet_b5_aa_in1k` | [`zeromodels/tf_efficientnet_b5_aa_in1k`](https://huggingface.co/zeromodels/tf_efficientnet_b5_aa_in1k) |
84
+ | `tf_efficientnet_b5_ap_in1k` | [`zeromodels/tf_efficientnet_b5_ap_in1k`](https://huggingface.co/zeromodels/tf_efficientnet_b5_ap_in1k) |
85
+ | `tf_efficientnet_b5_in1k` | [`zeromodels/tf_efficientnet_b5_in1k`](https://huggingface.co/zeromodels/tf_efficientnet_b5_in1k) |
86
+ | `tf_efficientnet_b5_ns_jft_in1k` | [`zeromodels/tf_efficientnet_b5_ns_jft_in1k`](https://huggingface.co/zeromodels/tf_efficientnet_b5_ns_jft_in1k) |
87
+ | `tf_efficientnet_b6_aa_in1k` | [`zeromodels/tf_efficientnet_b6_aa_in1k`](https://huggingface.co/zeromodels/tf_efficientnet_b6_aa_in1k) |
88
+ | `tf_efficientnet_b6_ap_in1k` | [`zeromodels/tf_efficientnet_b6_ap_in1k`](https://huggingface.co/zeromodels/tf_efficientnet_b6_ap_in1k) |
89
+ | `tf_efficientnet_b6_ns_jft_in1k` | [`zeromodels/tf_efficientnet_b6_ns_jft_in1k`](https://huggingface.co/zeromodels/tf_efficientnet_b6_ns_jft_in1k) |
90
+ | `tf_efficientnet_b7_aa_in1k` | [`zeromodels/tf_efficientnet_b7_aa_in1k`](https://huggingface.co/zeromodels/tf_efficientnet_b7_aa_in1k) |
91
+ | `tf_efficientnet_b7_ap_in1k` | [`zeromodels/tf_efficientnet_b7_ap_in1k`](https://huggingface.co/zeromodels/tf_efficientnet_b7_ap_in1k) |
92
+ | `tf_efficientnet_b7_ns_jft_in1k` | [`zeromodels/tf_efficientnet_b7_ns_jft_in1k`](https://huggingface.co/zeromodels/tf_efficientnet_b7_ns_jft_in1k) |
93
+ | `tf_efficientnet_b8_ap_in1k` | [`zeromodels/tf_efficientnet_b8_ap_in1k`](https://huggingface.co/zeromodels/tf_efficientnet_b8_ap_in1k) |
94
+ | `tf_efficientnet_l2_ns_jft_in1k` | [`zeromodels/tf_efficientnet_l2_ns_jft_in1k`](https://huggingface.co/zeromodels/tf_efficientnet_l2_ns_jft_in1k) |
95
+ | `tf_efficientnet_l2_ns_jft_in1k_475` | [`zeromodels/tf_efficientnet_l2_ns_jft_in1k_475`](https://huggingface.co/zeromodels/tf_efficientnet_l2_ns_jft_in1k_475) |
96
 
97
  ## Tips
98
 
99
+ - Set `KERAS_BACKEND` **before** importing Keras / zeromodels.
100
  - `EfficientNetImageClassify` returns class logits; `EfficientNetModel` returns features (`as_backbone=True` for multi-scale stages).
101
+ - See [docs](https://imvision12.github.io/ZeroModels/classification_backbones/) and [Loading Weights](https://imvision12.github.io/ZeroModels/loading_weights/).
102
  - Upstream / timm checkpoints: `EfficientNetImageClassify.from_weights("hf:timm/tf_efficientnet_b5.in1k")`.
103
 
104
  ## Special Thanks
kf_config.json → zm_config.json RENAMED
@@ -1,19 +1,19 @@
1
- {
2
- "library_name": "kerasformers",
3
- "kerasformers_version": "1.2.1",
4
- "model_module": "kerasformers.models.efficientnet",
5
- "model_class": "EfficientNetImageClassify",
6
- "variant": "tf_efficientnet_b5_in1k",
7
- "weights": "model.weights.h5",
8
- "schema_version": 2,
9
- "weight_dtype": "float32",
10
- "model_type": "efficientnet",
11
- "vision_config": {
12
- "width_coefficient": 1.6,
13
- "depth_coefficient": 2.2,
14
- "dropout_rate": 0.4,
15
- "default_size": 456,
16
- "image_size": 456,
17
- "num_classes": 1000
18
- }
19
  }
 
1
+ {
2
+ "library_name": "zeromodels",
3
+ "zeromodels_version": "1.2.1",
4
+ "model_module": "zeromodels.models.efficientnet",
5
+ "model_class": "EfficientNetImageClassify",
6
+ "variant": "tf_efficientnet_b5_in1k",
7
+ "weights": "model.weights.h5",
8
+ "schema_version": 2,
9
+ "weight_dtype": "float32",
10
+ "model_type": "efficientnet",
11
+ "vision_config": {
12
+ "width_coefficient": 1.6,
13
+ "depth_coefficient": 2.2,
14
+ "dropout_rate": 0.4,
15
+ "default_size": 456,
16
+ "image_size": 456,
17
+ "num_classes": 1000
18
+ }
19
  }