Hugging Face's logo

zeromodels
/
maxvit_tiny_tf_224_in1k

Image Classification
Keras
PyTorch
JAX
TensorFlow
zeromodels
maxvit
backbone
Model card Files
xet
 Community
IMvision12 commited on
Commit
6911bde
·
verified ·
1 Parent(s): 8b6a7b6

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 +33 -33
  2. kf_config.json → zm_config.json +34 -34
README.md CHANGED
@@ -2,10 +2,10 @@
2
  pipeline_tag: image-classification
3
  license: apache-2.0
4
  base_model: timm/maxvit_tiny_tf_224.in1k
5
- library_name: kerasformers
6
  tags:
7
  - keras
8
- - kerasformers
9
  - image-classification
10
  - maxvit
11
  - backbone
@@ -15,13 +15,13 @@ tags:
15
  - tf
16
  ---
17
 
18
- ## ***See [our collection](https://huggingface.co/collections/kerasformers/maxvit-6a6bd734b6d773748325c03c) for all versions of MaxViT.***
19
 
20
  # Run MaxViT 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-MaxViT-blue)](https://imvision12.github.io/KerasFormers/classification_backbones/) [![Collection](https://img.shields.io/badge/HF-MaxViT%20collection-yellow)](https://huggingface.co/collections/kerasformers/maxvit-6a6bd734b6d773748325c03c)
23
 
24
- # kerasformers/maxvit_tiny_tf_224_in1k
25
 
26
  Paper: [MaxViT: Multi-Axis Vision Transformer (arXiv:2204.01697)](https://arxiv.org/abs/2204.01697) · [HF Papers](https://huggingface.co/papers/2204.01697)
27
 
@@ -29,7 +29,7 @@ MaxViT combines blocked local and dilated global attention (multi-axis) in a hie
29
 
30
  For more details on the model, please go to the upstream [model card](https://huggingface.co/timm/maxvit_tiny_tf_224.in1k).
31
 
32
- Pure-**Keras 3** conversion of [`timm/maxvit_tiny_tf_224.in1k`](https://huggingface.co/timm/maxvit_tiny_tf_224.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 (`MaxViTImageClassify` / `MaxViTModel`).
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.maxvit import MaxViTImageClassify, MaxViTModel
45
 
46
- model = MaxViTImageClassify.from_weights("kerasformers/maxvit_tiny_tf_224_in1k")
47
  backbone = MaxViTModel.from_weights(
48
- "kerasformers/maxvit_tiny_tf_224_in1k", as_backbone=True
49
  )
50
 
51
  image = Image.open("your_image.jpg").convert("RGB")
@@ -56,37 +56,37 @@ feats = backbone(x)
56
  print(len(feats), [tuple(f.shape) for f in feats])
57
  ```
58
 
59
- Load any MaxViT variant the same way with `from_weights("kerasformers/<variant>")`:
60
 
61
  | Variant | Hub |
62
  |---|---|
63
- | `maxvit_base_tf_224_in1k` | [`kerasformers/maxvit_base_tf_224_in1k`](https://huggingface.co/kerasformers/maxvit_base_tf_224_in1k) |
64
- | `maxvit_base_tf_224_in21k` | [`kerasformers/maxvit_base_tf_224_in21k`](https://huggingface.co/kerasformers/maxvit_base_tf_224_in21k) |
65
- | `maxvit_base_tf_384_in1k` | [`kerasformers/maxvit_base_tf_384_in1k`](https://huggingface.co/kerasformers/maxvit_base_tf_384_in1k) |
66
- | `maxvit_base_tf_384_in21k_ft_in1k` | [`kerasformers/maxvit_base_tf_384_in21k_ft_in1k`](https://huggingface.co/kerasformers/maxvit_base_tf_384_in21k_ft_in1k) |
67
- | `maxvit_base_tf_512_in1k` | [`kerasformers/maxvit_base_tf_512_in1k`](https://huggingface.co/kerasformers/maxvit_base_tf_512_in1k) |
68
- | `maxvit_base_tf_512_in21k_ft_in1k` | [`kerasformers/maxvit_base_tf_512_in21k_ft_in1k`](https://huggingface.co/kerasformers/maxvit_base_tf_512_in21k_ft_in1k) |
69
- | `maxvit_large_tf_224_in1k` | [`kerasformers/maxvit_large_tf_224_in1k`](https://huggingface.co/kerasformers/maxvit_large_tf_224_in1k) |
70
- | `maxvit_large_tf_224_in21k` | [`kerasformers/maxvit_large_tf_224_in21k`](https://huggingface.co/kerasformers/maxvit_large_tf_224_in21k) |
71
- | `maxvit_large_tf_384_in1k` | [`kerasformers/maxvit_large_tf_384_in1k`](https://huggingface.co/kerasformers/maxvit_large_tf_384_in1k) |
72
- | `maxvit_large_tf_384_in21k_ft_in1k` | [`kerasformers/maxvit_large_tf_384_in21k_ft_in1k`](https://huggingface.co/kerasformers/maxvit_large_tf_384_in21k_ft_in1k) |
73
- | `maxvit_large_tf_512_in1k` | [`kerasformers/maxvit_large_tf_512_in1k`](https://huggingface.co/kerasformers/maxvit_large_tf_512_in1k) |
74
- | `maxvit_large_tf_512_in21k_ft_in1k` | [`kerasformers/maxvit_large_tf_512_in21k_ft_in1k`](https://huggingface.co/kerasformers/maxvit_large_tf_512_in21k_ft_in1k) |
75
- | `maxvit_small_tf_224_in1k` | [`kerasformers/maxvit_small_tf_224_in1k`](https://huggingface.co/kerasformers/maxvit_small_tf_224_in1k) |
76
- | `maxvit_small_tf_384_in1k` | [`kerasformers/maxvit_small_tf_384_in1k`](https://huggingface.co/kerasformers/maxvit_small_tf_384_in1k) |
77
- | `maxvit_small_tf_512_in1k` | [`kerasformers/maxvit_small_tf_512_in1k`](https://huggingface.co/kerasformers/maxvit_small_tf_512_in1k) |
78
- | `maxvit_tiny_tf_224_in1k` | [`kerasformers/maxvit_tiny_tf_224_in1k`](https://huggingface.co/kerasformers/maxvit_tiny_tf_224_in1k) |
79
- | `maxvit_tiny_tf_384_in1k` | [`kerasformers/maxvit_tiny_tf_384_in1k`](https://huggingface.co/kerasformers/maxvit_tiny_tf_384_in1k) |
80
- | `maxvit_tiny_tf_512_in1k` | [`kerasformers/maxvit_tiny_tf_512_in1k`](https://huggingface.co/kerasformers/maxvit_tiny_tf_512_in1k) |
81
- | `maxvit_xlarge_tf_224_in21k` | [`kerasformers/maxvit_xlarge_tf_224_in21k`](https://huggingface.co/kerasformers/maxvit_xlarge_tf_224_in21k) |
82
- | `maxvit_xlarge_tf_384_in21k_ft_in1k` | [`kerasformers/maxvit_xlarge_tf_384_in21k_ft_in1k`](https://huggingface.co/kerasformers/maxvit_xlarge_tf_384_in21k_ft_in1k) |
83
- | `maxvit_xlarge_tf_512_in21k_ft_in1k` | [`kerasformers/maxvit_xlarge_tf_512_in21k_ft_in1k`](https://huggingface.co/kerasformers/maxvit_xlarge_tf_512_in21k_ft_in1k) |
84
 
85
  ## Tips
86
 
87
- - Set `KERAS_BACKEND` **before** importing Keras / kerasformers.
88
  - `MaxViTImageClassify` returns class logits; `MaxViTModel` returns features (`as_backbone=True` for multi-scale stages).
89
- - See [docs](https://imvision12.github.io/KerasFormers/classification_backbones/) and [Loading Weights](https://imvision12.github.io/KerasFormers/loading_weights/).
90
  - Upstream / timm checkpoints: `MaxViTImageClassify.from_weights("hf:timm/maxvit_tiny_tf_224.in1k")`.
91
 
92
  ## Special Thanks
 
2
  pipeline_tag: image-classification
3
  license: apache-2.0
4
  base_model: timm/maxvit_tiny_tf_224.in1k
5
+ library_name: zeromodels
6
  tags:
7
  - keras
8
+ - zeromodels
9
  - image-classification
10
  - maxvit
11
  - backbone
 
15
  - tf
16
  ---
17
 
18
+ ## ***See [our collection](https://huggingface.co/collections/zeromodels/maxvit-6a6bd734b6d773748325c03c) for all versions of MaxViT.***
19
 
20
  # Run MaxViT 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-MaxViT-blue)](https://imvision12.github.io/ZeroModels/classification_backbones/) [![Collection](https://img.shields.io/badge/HF-MaxViT%20collection-yellow)](https://huggingface.co/collections/zeromodels/maxvit-6a6bd734b6d773748325c03c)
23
 
24
+ # zeromodels/maxvit_tiny_tf_224_in1k
25
 
26
  Paper: [MaxViT: Multi-Axis Vision Transformer (arXiv:2204.01697)](https://arxiv.org/abs/2204.01697) · [HF Papers](https://huggingface.co/papers/2204.01697)
27
 
 
29
 
30
  For more details on the model, please go to the upstream [model card](https://huggingface.co/timm/maxvit_tiny_tf_224.in1k).
31
 
32
+ Pure-**Keras 3** conversion of [`timm/maxvit_tiny_tf_224.in1k`](https://huggingface.co/timm/maxvit_tiny_tf_224.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 (`MaxViTImageClassify` / `MaxViTModel`).
35
 
 
41
 
42
  from PIL import Image
43
  import numpy as np
44
+ from zeromodels.models.maxvit import MaxViTImageClassify, MaxViTModel
45
 
46
+ model = MaxViTImageClassify.from_weights("zeromodels/maxvit_tiny_tf_224_in1k")
47
  backbone = MaxViTModel.from_weights(
48
+ "zeromodels/maxvit_tiny_tf_224_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 MaxViT variant the same way with `from_weights("zeromodels/<variant>")`:
60
 
61
  | Variant | Hub |
62
  |---|---|
63
+ | `maxvit_base_tf_224_in1k` | [`zeromodels/maxvit_base_tf_224_in1k`](https://huggingface.co/zeromodels/maxvit_base_tf_224_in1k) |
64
+ | `maxvit_base_tf_224_in21k` | [`zeromodels/maxvit_base_tf_224_in21k`](https://huggingface.co/zeromodels/maxvit_base_tf_224_in21k) |
65
+ | `maxvit_base_tf_384_in1k` | [`zeromodels/maxvit_base_tf_384_in1k`](https://huggingface.co/zeromodels/maxvit_base_tf_384_in1k) |
66
+ | `maxvit_base_tf_384_in21k_ft_in1k` | [`zeromodels/maxvit_base_tf_384_in21k_ft_in1k`](https://huggingface.co/zeromodels/maxvit_base_tf_384_in21k_ft_in1k) |
67
+ | `maxvit_base_tf_512_in1k` | [`zeromodels/maxvit_base_tf_512_in1k`](https://huggingface.co/zeromodels/maxvit_base_tf_512_in1k) |
68
+ | `maxvit_base_tf_512_in21k_ft_in1k` | [`zeromodels/maxvit_base_tf_512_in21k_ft_in1k`](https://huggingface.co/zeromodels/maxvit_base_tf_512_in21k_ft_in1k) |
69
+ | `maxvit_large_tf_224_in1k` | [`zeromodels/maxvit_large_tf_224_in1k`](https://huggingface.co/zeromodels/maxvit_large_tf_224_in1k) |
70
+ | `maxvit_large_tf_224_in21k` | [`zeromodels/maxvit_large_tf_224_in21k`](https://huggingface.co/zeromodels/maxvit_large_tf_224_in21k) |
71
+ | `maxvit_large_tf_384_in1k` | [`zeromodels/maxvit_large_tf_384_in1k`](https://huggingface.co/zeromodels/maxvit_large_tf_384_in1k) |
72
+ | `maxvit_large_tf_384_in21k_ft_in1k` | [`zeromodels/maxvit_large_tf_384_in21k_ft_in1k`](https://huggingface.co/zeromodels/maxvit_large_tf_384_in21k_ft_in1k) |
73
+ | `maxvit_large_tf_512_in1k` | [`zeromodels/maxvit_large_tf_512_in1k`](https://huggingface.co/zeromodels/maxvit_large_tf_512_in1k) |
74
+ | `maxvit_large_tf_512_in21k_ft_in1k` | [`zeromodels/maxvit_large_tf_512_in21k_ft_in1k`](https://huggingface.co/zeromodels/maxvit_large_tf_512_in21k_ft_in1k) |
75
+ | `maxvit_small_tf_224_in1k` | [`zeromodels/maxvit_small_tf_224_in1k`](https://huggingface.co/zeromodels/maxvit_small_tf_224_in1k) |
76
+ | `maxvit_small_tf_384_in1k` | [`zeromodels/maxvit_small_tf_384_in1k`](https://huggingface.co/zeromodels/maxvit_small_tf_384_in1k) |
77
+ | `maxvit_small_tf_512_in1k` | [`zeromodels/maxvit_small_tf_512_in1k`](https://huggingface.co/zeromodels/maxvit_small_tf_512_in1k) |
78
+ | `maxvit_tiny_tf_224_in1k` | [`zeromodels/maxvit_tiny_tf_224_in1k`](https://huggingface.co/zeromodels/maxvit_tiny_tf_224_in1k) |
79
+ | `maxvit_tiny_tf_384_in1k` | [`zeromodels/maxvit_tiny_tf_384_in1k`](https://huggingface.co/zeromodels/maxvit_tiny_tf_384_in1k) |
80
+ | `maxvit_tiny_tf_512_in1k` | [`zeromodels/maxvit_tiny_tf_512_in1k`](https://huggingface.co/zeromodels/maxvit_tiny_tf_512_in1k) |
81
+ | `maxvit_xlarge_tf_224_in21k` | [`zeromodels/maxvit_xlarge_tf_224_in21k`](https://huggingface.co/zeromodels/maxvit_xlarge_tf_224_in21k) |
82
+ | `maxvit_xlarge_tf_384_in21k_ft_in1k` | [`zeromodels/maxvit_xlarge_tf_384_in21k_ft_in1k`](https://huggingface.co/zeromodels/maxvit_xlarge_tf_384_in21k_ft_in1k) |
83
+ | `maxvit_xlarge_tf_512_in21k_ft_in1k` | [`zeromodels/maxvit_xlarge_tf_512_in21k_ft_in1k`](https://huggingface.co/zeromodels/maxvit_xlarge_tf_512_in21k_ft_in1k) |
84
 
85
  ## Tips
86
 
87
+ - Set `KERAS_BACKEND` **before** importing Keras / zeromodels.
88
  - `MaxViTImageClassify` returns class logits; `MaxViTModel` returns features (`as_backbone=True` for multi-scale stages).
89
+ - See [docs](https://imvision12.github.io/ZeroModels/classification_backbones/) and [Loading Weights](https://imvision12.github.io/ZeroModels/loading_weights/).
90
  - Upstream / timm checkpoints: `MaxViTImageClassify.from_weights("hf:timm/maxvit_tiny_tf_224.in1k")`.
91
 
92
  ## Special Thanks
kf_config.json → zm_config.json RENAMED
@@ -1,35 +1,35 @@
1
- {
2
- "library_name": "kerasformers",
3
- "kerasformers_version": "1.2.1",
4
- "model_module": "kerasformers.models.maxvit",
5
- "model_class": "MaxViTImageClassify",
6
- "variant": "maxvit_tiny_tf_224_in1k",
7
- "weights": "model.weights.h5",
8
- "schema_version": 2,
9
- "weight_dtype": "float32",
10
- "model_type": "maxvit",
11
- "vision_config": {
12
- "stem_width": 64,
13
- "depths": [
14
- 2,
15
- 2,
16
- 5,
17
- 2
18
- ],
19
- "embed_dim": [
20
- 64,
21
- 128,
22
- 256,
23
- 512
24
- ],
25
- "num_heads": [
26
- 2,
27
- 4,
28
- 8,
29
- 16
30
- ],
31
- "window_size": 7,
32
- "image_size": 224,
33
- "num_classes": 1000
34
- }
35
  }
 
1
+ {
2
+ "library_name": "zeromodels",
3
+ "zeromodels_version": "1.2.1",
4
+ "model_module": "zeromodels.models.maxvit",
5
+ "model_class": "MaxViTImageClassify",
6
+ "variant": "maxvit_tiny_tf_224_in1k",
7
+ "weights": "model.weights.h5",
8
+ "schema_version": 2,
9
+ "weight_dtype": "float32",
10
+ "model_type": "maxvit",
11
+ "vision_config": {
12
+ "stem_width": 64,
13
+ "depths": [
14
+ 2,
15
+ 2,
16
+ 5,
17
+ 2
18
+ ],
19
+ "embed_dim": [
20
+ 64,
21
+ 128,
22
+ 256,
23
+ 512
24
+ ],
25
+ "num_heads": [
26
+ 2,
27
+ 4,
28
+ 8,
29
+ 16
30
+ ],
31
+ "window_size": 7,
32
+ "image_size": 224,
33
+ "num_classes": 1000
34
+ }
35
  }