Add Iwin model config, weights (pth + safetensors), and README

.gitattributes

@@ -27,8 +27,9 @@ saved_model/**/* filter=lfs diff=lfs merge=lfs -text
 *.tar.* filter=lfs diff=lfs merge=lfs -text
 *.tar filter=lfs diff=lfs merge=lfs -text
 *.tflite filter=lfs diff=lfs merge=lfs -text
-*.tgz filter=lfs diff=lfs merge=lfs -text
-*.wasm filter=lfs diff=lfs merge=lfs -text
+*.tif filter=lfs diff=lfs merge=lfs -text
+*.tiff filter=lfs diff=lfs merge=lfs -text
+*.txt filter=lfs diff=lfs merge=lfs -text
 *.xz filter=lfs diff=lfs merge=lfs -text
 *.zip filter=lfs diff=lfs merge=lfs -text
 *.zst filter=lfs diff=lfs merge=lfs -text

README.md

@@ -0,0 +1,41 @@
+---
+license: mit
+tags:
+- vision
+- image-classification
+- arxiv:2507.18405
+datasets:
+- imagenet-22k
+---
+
+
+# Iwin Transformer (large-sized model, imagenet-22k, 224x224)
+
+<div style="display:flex;justify-content: left">
+  <a href="https://arxiv.org/abs/2507.18405"><img src="https://img.shields.io/static/v1?label=Paper&message=Arxiv:Iwin&color=red&logo=arxiv"></a> &ensp;
+   <a href="https://github.com/Cominder/Iwin-Transformer/"><img src="https://img.shields.io/static/v1?label=Repository&message=Github&color=blue&logo=github-pages"></a> &ensp;
+</div>
+
+## Model Source
+Pre-trained Iwin Transformer models on ImageNet-1k and ImageNet-22k. For research purposes, we recommend our Github repository (https://github.com/Cominder/Iwin-Transformer), which is more suitable for both training and inference.
+
+## Model Descprition
+We introduce Iwin Transformer, a novel positionembedding-free hierarchical vision transformer, which can be fine-tuned directly from low to high resolution, through the
+collaboration of innovative interleaved window attention and depthwise separable convolution. This approach uses attention to connect distant tokens and applies convolution to link neighboring tokens, enabling global information exchange within a
+single module, overcoming Swin Transformer's limitation of requiring two consecutive blocks to approximate global attention. Extensive experiments on visual benchmarks demonstrate that Iwin Transformer exhibits strong competitiveness in tasks such as image classification (87.4 top-1 accuracy on ImageNet-1K),
+semantic segmentation, and video action recognition. We also validate the effectiveness of the core component in Iwin as a standalone module that can seamlessly replace the self-attention module in class-conditional image generation. The concepts
+and methods introduced by the Iwin Transformer have the potential to inspire future research, like Iwin 3D Attention in video generation. The code and models are available at
+https://github.com/cominder/Iwin-Transformer.
+
+## Citation
+```
+@misc{huo2025iwin,
+      title={Iwin Transformer: Hierarchical Vision Transformer using Interleaved Windows}, 
+      author={Simin Huo and Ning Li},
+      year={2025},
+      eprint={2507.18405},
+      archivePrefix={arXiv},
+      primaryClass={cs.CV},
+      url={https://arxiv.org/abs/2507.18405}, 
+}
+```

config.json

@@ -0,0 +1,35 @@
+{
+  "architectures": [
+    "IwinForImageClassification"
+  ],
+  "attention_probs_dropout_prob": 0.0,
+  "depths": [
+    2,
+    2,
+    18,
+    2
+  ],
+  "drop_path_rate": 0.2,
+  "embed_dim": 192,
+  "hidden_act": "gelu",
+  "hidden_dropout_prob": 0.0,
+  "image_size": 224,
+  "initializer_range": 0.02,
+  "layer_norm_eps": 1e-05,
+  "mlp_ratio": 4.0,
+  "model_type": "iwin",
+  "num_channels": 3,
+  "num_heads": [
+    6,
+    12,
+    24,
+    48
+  ],
+  "num_labels": 21841,
+  "patch_size": 4,
+  "path_norm": true,
+  "qkv_bias": true,
+  "torch_dtype": "float32",
+  "use_absolute_embeddings": false,
+  "window_size": 7
+}

iwin_large_patch4_window7_224_22k.pth

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:ef6de2839858159f8b5227fd069ad38bf5b40dfa6781ae5f8af9e768195f5054
+size 945561150

model.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:dad7b16ff1bc59e96336cf3be476720dd22bb64594994e6ab487fba9eae066b6
+size 945458972

preprocessor_config.json

@@ -0,0 +1,17 @@
+{
+  "do_normalize": true,
+  "do_resize": true,
+  "feature_extractor_type": "ViTFeatureExtractor",
+  "image_mean": [
+    0.485,
+    0.456,
+    0.406
+  ],
+  "image_std": [
+    0.229,
+    0.224,
+    0.225
+  ],
+  "resample": 3,
+  "size": 224
+}