fix readme.md

README.md

@@ -8,20 +8,80 @@ tags:
 - kerasformers
 - image-classification
 - nextvit
-- tf
-- jax
+- backbone
+- arxiv:2207.05501
 - pytorch
+- jax
+- tf
 ---
 
-# nextvit_base_bd_in1k
+## ***See [our collection](https://huggingface.co/collections/kerasformers/next-vit-6a6bd9ad0b496b30d1aa3343) for all versions of Next-ViT.***
+
+# Run Next-ViT with Keras 3: JAX, PyTorch, or TensorFlow
+
+[![GitHub](https://img.shields.io/badge/GitHub-KerasFormers-black?logo=github)](https://github.com/IMvision12/KerasFormers) [![Docs](https://img.shields.io/badge/Docs-Next--ViT-blue)](https://imvision12.github.io/KerasFormers/classification_backbones/) [![Collection](https://img.shields.io/badge/HF-Next--ViT%20collection-yellow)](https://huggingface.co/collections/kerasformers/next-vit-6a6bd9ad0b496b30d1aa3343)
+
+# kerasformers/nextvit_base_bd_in1k
+
+Paper: [Next-ViT: Next Generation Vision Transformer for Efficient Deployment in Realistic Industrial Scenarios (arXiv:2207.05501)](https://arxiv.org/abs/2207.05501) · [HF Papers](https://huggingface.co/papers/2207.05501)
+
+Next-ViT targets efficient industrial deployment with a hybrid CNN/Transformer stack. Classifier or hierarchical backbone.
+
+For more details on the model, please go to the upstream [model card](https://huggingface.co/timm/nextvit_base.bd_in1k).
 
-Pure-Keras 3 image-classification weight for [kerasformers](https://github.com/IMvision12/KerasFormers), converted from [timm/nextvit_base.bd_in1k](https://huggingface.co/timm/nextvit_base.bd_in1k).
+Pure-**Keras 3** conversion of [`timm/nextvit_base.bd_in1k`](https://huggingface.co/timm/nextvit_base.bd_in1k) for [kerasformers](https://github.com/IMvision12/KerasFormers). One implementation runs unmodified on **TensorFlow / Torch / JAX**.
 
-## Usage
+This is an **image-classification / backbone** checkpoint (`NextViTImageClassify` / `NextViTModel`).
+
+## ✨ Quick start
 
 ```python
-from kerasformers.models.nextvit import NextViTImageClassify
-model = NextViTImageClassify.from_weights("nextvit_base_bd_in1k")
+import os
+os.environ["KERAS_BACKEND"] = "torch"  # or "jax" / "tensorflow"
+
+from PIL import Image
+import numpy as np
+from kerasformers.models.nextvit import NextViTImageClassify, NextViTModel
+
+model = NextViTImageClassify.from_weights("kerasformers/nextvit_base_bd_in1k")
+backbone = NextViTModel.from_weights(
+    "kerasformers/nextvit_base_bd_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: **apache-2.0**, inherited from the upstream source.
+Load any Next-ViT variant the same way with `from_weights("kerasformers/<variant>")`:
+
+| Variant | Hub |
+|---|---|
+| `nextvit_base_bd_in1k` | [`kerasformers/nextvit_base_bd_in1k`](https://huggingface.co/kerasformers/nextvit_base_bd_in1k) |
+| `nextvit_base_bd_in1k_384` | [`kerasformers/nextvit_base_bd_in1k_384`](https://huggingface.co/kerasformers/nextvit_base_bd_in1k_384) |
+| `nextvit_base_bd_ssld_6m_in1k` | [`kerasformers/nextvit_base_bd_ssld_6m_in1k`](https://huggingface.co/kerasformers/nextvit_base_bd_ssld_6m_in1k) |
+| `nextvit_base_bd_ssld_6m_in1k_384` | [`kerasformers/nextvit_base_bd_ssld_6m_in1k_384`](https://huggingface.co/kerasformers/nextvit_base_bd_ssld_6m_in1k_384) |
+| `nextvit_large_bd_in1k` | [`kerasformers/nextvit_large_bd_in1k`](https://huggingface.co/kerasformers/nextvit_large_bd_in1k) |
+| `nextvit_large_bd_in1k_384` | [`kerasformers/nextvit_large_bd_in1k_384`](https://huggingface.co/kerasformers/nextvit_large_bd_in1k_384) |
+| `nextvit_large_bd_ssld_6m_in1k` | [`kerasformers/nextvit_large_bd_ssld_6m_in1k`](https://huggingface.co/kerasformers/nextvit_large_bd_ssld_6m_in1k) |
+| `nextvit_large_bd_ssld_6m_in1k_384` | [`kerasformers/nextvit_large_bd_ssld_6m_in1k_384`](https://huggingface.co/kerasformers/nextvit_large_bd_ssld_6m_in1k_384) |
+| `nextvit_small_bd_in1k` | [`kerasformers/nextvit_small_bd_in1k`](https://huggingface.co/kerasformers/nextvit_small_bd_in1k) |
+| `nextvit_small_bd_in1k_384` | [`kerasformers/nextvit_small_bd_in1k_384`](https://huggingface.co/kerasformers/nextvit_small_bd_in1k_384) |
+| `nextvit_small_bd_ssld_6m_in1k` | [`kerasformers/nextvit_small_bd_ssld_6m_in1k`](https://huggingface.co/kerasformers/nextvit_small_bd_ssld_6m_in1k) |
+| `nextvit_small_bd_ssld_6m_in1k_384` | [`kerasformers/nextvit_small_bd_ssld_6m_in1k_384`](https://huggingface.co/kerasformers/nextvit_small_bd_ssld_6m_in1k_384) |
+
+## Tips
+
+- Set `KERAS_BACKEND` **before** importing Keras / kerasformers.
+- `NextViTImageClassify` returns class logits; `NextViTModel` 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: `NextViTImageClassify.from_weights("hf:timm/nextvit_base.bd_in1k")`.
+
+## Special Thanks
+
+A huge thank you to the Next-ViT authors and the timm / Hub communities for creating and releasing these models.
+
+License: see YAML `license` (usually matches the upstream checkpoint).