fix readme.md

README.md

@@ -8,20 +8,80 @@ tags:
 - kerasformers
 - image-classification
 - swinv2
+- backbone
+- arxiv:2111.09883
 - pytorch
 - jax
 - tf
 ---
 
-# swinv2_small_window8_256_ms_in1k
+## ***See [our collection](https://huggingface.co/collections/kerasformers/swin-transformer-v2-6a6d0b232b59a65efe584700) for all versions of Swin Transformer V2.***
 
-Pure-Keras 3 image-classification weight for [kerasformers](https://github.com/IMvision12/KerasFormers), converted from [timm/swinv2_small_window8_256.ms_in1k](https://huggingface.co/timm/swinv2_small_window8_256.ms_in1k).
+# Run Swin Transformer V2 with Keras 3: JAX, PyTorch, or TensorFlow
 
-## Usage
+[![GitHub](https://img.shields.io/badge/GitHub-KerasFormers-black?logo=github)](https://github.com/IMvision12/KerasFormers) [![Docs](https://img.shields.io/badge/Docs-Swin--V2-blue)](https://imvision12.github.io/KerasFormers/classification_backbones/) [![Collection](https://img.shields.io/badge/HF-Swin--V2%20collection-yellow)](https://huggingface.co/collections/kerasformers/swin-transformer-v2-6a6d0b232b59a65efe584700)
+
+# kerasformers/swinv2_small_window8_256_ms_in1k
+
+Paper: [Swin Transformer V2: Scaling Up Capacity and Resolution (arXiv:2111.09883)](https://arxiv.org/abs/2111.09883) · [HF Papers](https://huggingface.co/papers/2111.09883)
+
+Swin V2 scales capacity and resolution with residual post-norm and scaled cosine attention. Classifier + hierarchical backbone.
+
+For more details on the model, please go to the upstream [model card](https://huggingface.co/timm/swinv2_small_window8_256.ms_in1k).
+
+Pure-**Keras 3** conversion of [`timm/swinv2_small_window8_256.ms_in1k`](https://huggingface.co/timm/swinv2_small_window8_256.ms_in1k) for [kerasformers](https://github.com/IMvision12/KerasFormers). One implementation runs unmodified on **TensorFlow / Torch / JAX**.
+
+This is an **image-classification / backbone** checkpoint (`SwinV2ImageClassify` / `SwinV2Model`).
+
+## ✨ Quick start
 
 ```python
-from kerasformers.models.swinv2 import SwinV2ImageClassify
-model = SwinV2ImageClassify.from_weights("swinv2_small_window8_256_ms_in1k")
+import os
+os.environ["KERAS_BACKEND"] = "torch"  # or "jax" / "tensorflow"
+
+from PIL import Image
+import numpy as np
+from kerasformers.models.swinv2 import SwinV2ImageClassify, SwinV2Model
+
+model = SwinV2ImageClassify.from_weights("kerasformers/swinv2_small_window8_256_ms_in1k")
+backbone = SwinV2Model.from_weights(
+    "kerasformers/swinv2_small_window8_256_ms_in1k", as_backbone=True
+)
+
+image = Image.open("your_image.jpg").convert("RGB")
+image = image.resize((224, 224))
+x = np.asarray(image, dtype="float32")[None]  # (1, H, W, 3)
+print(model(x).shape)  # (1, num_classes)
+feats = backbone(x)
+print(len(feats), [tuple(f.shape) for f in feats])
 ```
 
-License: **mit**, inherited from the upstream source.
+Load any Swin Transformer V2 variant the same way with `from_weights("kerasformers/<variant>")`:
+
+| Variant | Hub |
+|---|---|
+| `swinv2_base_window12_192_ms_in22k` | [`kerasformers/swinv2_base_window12_192_ms_in22k`](https://huggingface.co/kerasformers/swinv2_base_window12_192_ms_in22k) |
+| `swinv2_base_window12to16_192to256_ms_in22k_ft_in1k` | [`kerasformers/swinv2_base_window12to16_192to256_ms_in22k_ft_in1k`](https://huggingface.co/kerasformers/swinv2_base_window12to16_192to256_ms_in22k_ft_in1k) |
+| `swinv2_base_window12to24_192to384_ms_in22k_ft_in1k` | [`kerasformers/swinv2_base_window12to24_192to384_ms_in22k_ft_in1k`](https://huggingface.co/kerasformers/swinv2_base_window12to24_192to384_ms_in22k_ft_in1k) |
+| `swinv2_base_window16_256_ms_in1k` | [`kerasformers/swinv2_base_window16_256_ms_in1k`](https://huggingface.co/kerasformers/swinv2_base_window16_256_ms_in1k) |
+| `swinv2_base_window8_256_ms_in1k` | [`kerasformers/swinv2_base_window8_256_ms_in1k`](https://huggingface.co/kerasformers/swinv2_base_window8_256_ms_in1k) |
+| `swinv2_large_window12_192_ms_in22k` | [`kerasformers/swinv2_large_window12_192_ms_in22k`](https://huggingface.co/kerasformers/swinv2_large_window12_192_ms_in22k) |
+| `swinv2_large_window12to16_192to256_ms_in22k_ft_in1k` | [`kerasformers/swinv2_large_window12to16_192to256_ms_in22k_ft_in1k`](https://huggingface.co/kerasformers/swinv2_large_window12to16_192to256_ms_in22k_ft_in1k) |
+| `swinv2_large_window12to24_192to384_ms_in22k_ft_in1k` | [`kerasformers/swinv2_large_window12to24_192to384_ms_in22k_ft_in1k`](https://huggingface.co/kerasformers/swinv2_large_window12to24_192to384_ms_in22k_ft_in1k) |
+| `swinv2_small_window16_256_ms_in1k` | [`kerasformers/swinv2_small_window16_256_ms_in1k`](https://huggingface.co/kerasformers/swinv2_small_window16_256_ms_in1k) |
+| `swinv2_small_window8_256_ms_in1k` | [`kerasformers/swinv2_small_window8_256_ms_in1k`](https://huggingface.co/kerasformers/swinv2_small_window8_256_ms_in1k) |
+| `swinv2_tiny_window16_256_ms_in1k` | [`kerasformers/swinv2_tiny_window16_256_ms_in1k`](https://huggingface.co/kerasformers/swinv2_tiny_window16_256_ms_in1k) |
+| `swinv2_tiny_window8_256_ms_in1k` | [`kerasformers/swinv2_tiny_window8_256_ms_in1k`](https://huggingface.co/kerasformers/swinv2_tiny_window8_256_ms_in1k) |
+
+## Tips
+
+- Set `KERAS_BACKEND` **before** importing Keras / kerasformers.
+- `SwinV2ImageClassify` returns class logits; `SwinV2Model` returns features (`as_backbone=True` for multi-scale stages).
+- See [docs](https://imvision12.github.io/KerasFormers/classification_backbones/) and [Loading Weights](https://imvision12.github.io/KerasFormers/loading_weights/).
+- Upstream / timm checkpoints: `SwinV2ImageClassify.from_weights("hf:timm/swinv2_small_window8_256.ms_in1k")`.
+
+## Special Thanks
+
+A huge thank you to the Swin Transformer V2 authors and the timm / Hub communities for creating and releasing these models.
+
+License: see YAML `license` (usually matches the upstream checkpoint).