first commit

Code.ipynb

@@ -0,0 +1,224 @@
+{
+ "cells": [
+  {
+   "cell_type": "code",
+   "execution_count": 1,
+   "id": "a7dd2ed2",
+   "metadata": {},
+   "outputs": [
+    {
+     "name": "stdout",
+     "output_type": "stream",
+     "text": [
+      "Defaulting to user installation because normal site-packages is not writeable\n",
+      "Collecting timm\n",
+      "  Downloading timm-1.0.15-py3-none-any.whl.metadata (52 kB)\n",
+      "Requirement already satisfied: torch in /home/ubuntu/.local/lib/python3.12/site-packages (from timm) (2.6.0)\n",
+      "Requirement already satisfied: torchvision in /home/ubuntu/.local/lib/python3.12/site-packages (from timm) (0.21.0)\n",
+      "Requirement already satisfied: pyyaml in /opt/anaconda3/lib/python3.12/site-packages (from timm) (6.0.1)\n",
+      "Requirement already satisfied: huggingface_hub in /home/ubuntu/.local/lib/python3.12/site-packages (from timm) (0.29.3)\n",
+      "Requirement already satisfied: safetensors in /home/ubuntu/.local/lib/python3.12/site-packages (from timm) (0.5.3)\n",
+      "Requirement already satisfied: filelock in /opt/anaconda3/lib/python3.12/site-packages (from huggingface_hub->timm) (3.13.1)\n",
+      "Requirement already satisfied: fsspec>=2023.5.0 in /opt/anaconda3/lib/python3.12/site-packages (from huggingface_hub->timm) (2024.6.1)\n",
+      "Requirement already satisfied: packaging>=20.9 in /opt/anaconda3/lib/python3.12/site-packages (from huggingface_hub->timm) (24.1)\n",
+      "Requirement already satisfied: requests in /opt/anaconda3/lib/python3.12/site-packages (from huggingface_hub->timm) (2.32.3)\n",
+      "Requirement already satisfied: tqdm>=4.42.1 in /opt/anaconda3/lib/python3.12/site-packages (from huggingface_hub->timm) (4.66.5)\n",
+      "Requirement already satisfied: typing-extensions>=3.7.4.3 in /opt/anaconda3/lib/python3.12/site-packages (from huggingface_hub->timm) (4.11.0)\n",
+      "Requirement already satisfied: networkx in /opt/anaconda3/lib/python3.12/site-packages (from torch->timm) (3.3)\n",
+      "Requirement already satisfied: jinja2 in /opt/anaconda3/lib/python3.12/site-packages (from torch->timm) (3.1.4)\n",
+      "Requirement already satisfied: nvidia-cuda-nvrtc-cu12==12.4.127 in /home/ubuntu/.local/lib/python3.12/site-packages (from torch->timm) (12.4.127)\n",
+      "Requirement already satisfied: nvidia-cuda-runtime-cu12==12.4.127 in /home/ubuntu/.local/lib/python3.12/site-packages (from torch->timm) (12.4.127)\n",
+      "Requirement already satisfied: nvidia-cuda-cupti-cu12==12.4.127 in /home/ubuntu/.local/lib/python3.12/site-packages (from torch->timm) (12.4.127)\n",
+      "Requirement already satisfied: nvidia-cudnn-cu12==9.1.0.70 in /home/ubuntu/.local/lib/python3.12/site-packages (from torch->timm) (9.1.0.70)\n",
+      "Requirement already satisfied: nvidia-cublas-cu12==12.4.5.8 in /home/ubuntu/.local/lib/python3.12/site-packages (from torch->timm) (12.4.5.8)\n",
+      "Requirement already satisfied: nvidia-cufft-cu12==11.2.1.3 in /home/ubuntu/.local/lib/python3.12/site-packages (from torch->timm) (11.2.1.3)\n",
+      "Requirement already satisfied: nvidia-curand-cu12==10.3.5.147 in /home/ubuntu/.local/lib/python3.12/site-packages (from torch->timm) (10.3.5.147)\n",
+      "Requirement already satisfied: nvidia-cusolver-cu12==11.6.1.9 in /home/ubuntu/.local/lib/python3.12/site-packages (from torch->timm) (11.6.1.9)\n",
+      "Requirement already satisfied: nvidia-cusparse-cu12==12.3.1.170 in /home/ubuntu/.local/lib/python3.12/site-packages (from torch->timm) (12.3.1.170)\n",
+      "Requirement already satisfied: nvidia-cusparselt-cu12==0.6.2 in /home/ubuntu/.local/lib/python3.12/site-packages (from torch->timm) (0.6.2)\n",
+      "Requirement already satisfied: nvidia-nccl-cu12==2.21.5 in /home/ubuntu/.local/lib/python3.12/site-packages (from torch->timm) (2.21.5)\n",
+      "Requirement already satisfied: nvidia-nvtx-cu12==12.4.127 in /home/ubuntu/.local/lib/python3.12/site-packages (from torch->timm) (12.4.127)\n",
+      "Requirement already satisfied: nvidia-nvjitlink-cu12==12.4.127 in /home/ubuntu/.local/lib/python3.12/site-packages (from torch->timm) (12.4.127)\n",
+      "Requirement already satisfied: triton==3.2.0 in /home/ubuntu/.local/lib/python3.12/site-packages (from torch->timm) (3.2.0)\n",
+      "Requirement already satisfied: setuptools in /opt/anaconda3/lib/python3.12/site-packages (from torch->timm) (75.1.0)\n",
+      "Requirement already satisfied: sympy==1.13.1 in /home/ubuntu/.local/lib/python3.12/site-packages (from torch->timm) (1.13.1)\n",
+      "Requirement already satisfied: mpmath<1.4,>=1.1.0 in /opt/anaconda3/lib/python3.12/site-packages (from sympy==1.13.1->torch->timm) (1.3.0)\n",
+      "Requirement already satisfied: numpy in /opt/anaconda3/lib/python3.12/site-packages (from torchvision->timm) (1.26.4)\n",
+      "Requirement already satisfied: pillow!=8.3.*,>=5.3.0 in /opt/anaconda3/lib/python3.12/site-packages (from torchvision->timm) (10.4.0)\n",
+      "Requirement already satisfied: MarkupSafe>=2.0 in /opt/anaconda3/lib/python3.12/site-packages (from jinja2->torch->timm) (2.1.3)\n",
+      "Requirement already satisfied: charset-normalizer<4,>=2 in /opt/anaconda3/lib/python3.12/site-packages (from requests->huggingface_hub->timm) (3.3.2)\n",
+      "Requirement already satisfied: idna<4,>=2.5 in /opt/anaconda3/lib/python3.12/site-packages (from requests->huggingface_hub->timm) (3.7)\n",
+      "Requirement already satisfied: urllib3<3,>=1.21.1 in /opt/anaconda3/lib/python3.12/site-packages (from requests->huggingface_hub->timm) (2.2.3)\n",
+      "Requirement already satisfied: certifi>=2017.4.17 in /opt/anaconda3/lib/python3.12/site-packages (from requests->huggingface_hub->timm) (2024.8.30)\n",
+      "Downloading timm-1.0.15-py3-none-any.whl (2.4 MB)\n",
+      "\u001b[2K   \u001b[90m━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━\u001b[0m \u001b[32m2.4/2.4 MB\u001b[0m \u001b[31m13.6 MB/s\u001b[0m eta \u001b[36m0:00:00\u001b[0ma \u001b[36m0:00:01\u001b[0m\n",
+      "\u001b[?25hInstalling collected packages: timm\n",
+      "Successfully installed timm-1.0.15\n"
+     ]
+    }
+   ],
+   "source": [
+    "!pip install timm"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": 2,
+   "id": "ff0cd74c",
+   "metadata": {},
+   "outputs": [
+    {
+     "data": {
+      "application/vnd.jupyter.widget-view+json": {
+       "model_id": "e878415c59ba4f97a1fac91e71010469",
+       "version_major": 2,
+       "version_minor": 0
+      },
+      "text/plain": [
+       "config.json:   0%|          | 0.00/742 [00:00<?, ?B/s]"
+      ]
+     },
+     "metadata": {},
+     "output_type": "display_data"
+    },
+    {
+     "data": {
+      "application/vnd.jupyter.widget-view+json": {
+       "model_id": "356e15c8e1434fbe9d52f6ec94d6e875",
+       "version_major": 2,
+       "version_minor": 0
+      },
+      "text/plain": [
+       "model.safetensors:   0%|          | 0.00/2.53G [00:00<?, ?B/s]"
+      ]
+     },
+     "metadata": {},
+     "output_type": "display_data"
+    },
+    {
+     "ename": "RuntimeError",
+     "evalue": "Error(s) in loading state_dict for VisionTransformer:\n\tsize mismatch for blocks.0.mlp.fc2.weight: copying a param with shape torch.Size([1280, 3416]) from checkpoint, the shape in current model is torch.Size([1280, 6832]).\n\tsize mismatch for blocks.1.mlp.fc2.weight: copying a param with shape torch.Size([1280, 3416]) from checkpoint, the shape in current model is torch.Size([1280, 6832]).\n\tsize mismatch for blocks.2.mlp.fc2.weight: copying a param with shape torch.Size([1280, 3416]) from checkpoint, the shape in current model is torch.Size([1280, 6832]).\n\tsize mismatch for blocks.3.mlp.fc2.weight: copying a param with shape torch.Size([1280, 3416]) from checkpoint, the shape in current model is torch.Size([1280, 6832]).\n\tsize mismatch for blocks.4.mlp.fc2.weight: copying a param with shape torch.Size([1280, 3416]) from checkpoint, the shape in current model is torch.Size([1280, 6832]).\n\tsize mismatch for blocks.5.mlp.fc2.weight: copying a param with shape torch.Size([1280, 3416]) from checkpoint, the shape in current model is torch.Size([1280, 6832]).\n\tsize mismatch for blocks.6.mlp.fc2.weight: copying a param with shape torch.Size([1280, 3416]) from checkpoint, the shape in current model is torch.Size([1280, 6832]).\n\tsize mismatch for blocks.7.mlp.fc2.weight: copying a param with shape torch.Size([1280, 3416]) from checkpoint, the shape in current model is torch.Size([1280, 6832]).\n\tsize mismatch for blocks.8.mlp.fc2.weight: copying a param with shape torch.Size([1280, 3416]) from checkpoint, the shape in current model is torch.Size([1280, 6832]).\n\tsize mismatch for blocks.9.mlp.fc2.weight: copying a param with shape torch.Size([1280, 3416]) from checkpoint, the shape in current model is torch.Size([1280, 6832]).\n\tsize mismatch for blocks.10.mlp.fc2.weight: copying a param with shape torch.Size([1280, 3416]) from checkpoint, the shape in current model is torch.Size([1280, 6832]).\n\tsize mismatch for blocks.11.mlp.fc2.weight: copying a param with shape torch.Size([1280, 3416]) from checkpoint, the shape in current model is torch.Size([1280, 6832]).\n\tsize mismatch for blocks.12.mlp.fc2.weight: copying a param with shape torch.Size([1280, 3416]) from checkpoint, the shape in current model is torch.Size([1280, 6832]).\n\tsize mismatch for blocks.13.mlp.fc2.weight: copying a param with shape torch.Size([1280, 3416]) from checkpoint, the shape in current model is torch.Size([1280, 6832]).\n\tsize mismatch for blocks.14.mlp.fc2.weight: copying a param with shape torch.Size([1280, 3416]) from checkpoint, the shape in current model is torch.Size([1280, 6832]).\n\tsize mismatch for blocks.15.mlp.fc2.weight: copying a param with shape torch.Size([1280, 3416]) from checkpoint, the shape in current model is torch.Size([1280, 6832]).\n\tsize mismatch for blocks.16.mlp.fc2.weight: copying a param with shape torch.Size([1280, 3416]) from checkpoint, the shape in current model is torch.Size([1280, 6832]).\n\tsize mismatch for blocks.17.mlp.fc2.weight: copying a param with shape torch.Size([1280, 3416]) from checkpoint, the shape in current model is torch.Size([1280, 6832]).\n\tsize mismatch for blocks.18.mlp.fc2.weight: copying a param with shape torch.Size([1280, 3416]) from checkpoint, the shape in current model is torch.Size([1280, 6832]).\n\tsize mismatch for blocks.19.mlp.fc2.weight: copying a param with shape torch.Size([1280, 3416]) from checkpoint, the shape in current model is torch.Size([1280, 6832]).\n\tsize mismatch for blocks.20.mlp.fc2.weight: copying a param with shape torch.Size([1280, 3416]) from checkpoint, the shape in current model is torch.Size([1280, 6832]).\n\tsize mismatch for blocks.21.mlp.fc2.weight: copying a param with shape torch.Size([1280, 3416]) from checkpoint, the shape in current model is torch.Size([1280, 6832]).\n\tsize mismatch for blocks.22.mlp.fc2.weight: copying a param with shape torch.Size([1280, 3416]) from checkpoint, the shape in current model is torch.Size([1280, 6832]).\n\tsize mismatch for blocks.23.mlp.fc2.weight: copying a param with shape torch.Size([1280, 3416]) from checkpoint, the shape in current model is torch.Size([1280, 6832]).\n\tsize mismatch for blocks.24.mlp.fc2.weight: copying a param with shape torch.Size([1280, 3416]) from checkpoint, the shape in current model is torch.Size([1280, 6832]).\n\tsize mismatch for blocks.25.mlp.fc2.weight: copying a param with shape torch.Size([1280, 3416]) from checkpoint, the shape in current model is torch.Size([1280, 6832]).\n\tsize mismatch for blocks.26.mlp.fc2.weight: copying a param with shape torch.Size([1280, 3416]) from checkpoint, the shape in current model is torch.Size([1280, 6832]).\n\tsize mismatch for blocks.27.mlp.fc2.weight: copying a param with shape torch.Size([1280, 3416]) from checkpoint, the shape in current model is torch.Size([1280, 6832]).\n\tsize mismatch for blocks.28.mlp.fc2.weight: copying a param with shape torch.Size([1280, 3416]) from checkpoint, the shape in current model is torch.Size([1280, 6832]).\n\tsize mismatch for blocks.29.mlp.fc2.weight: copying a param with shape torch.Size([1280, 3416]) from checkpoint, the shape in current model is torch.Size([1280, 6832]).\n\tsize mismatch for blocks.30.mlp.fc2.weight: copying a param with shape torch.Size([1280, 3416]) from checkpoint, the shape in current model is torch.Size([1280, 6832]).\n\tsize mismatch for blocks.31.mlp.fc2.weight: copying a param with shape torch.Size([1280, 3416]) from checkpoint, the shape in current model is torch.Size([1280, 6832]).",
+     "output_type": "error",
+     "traceback": [
+      "\u001b[0;31m---------------------------------------------------------------------------\u001b[0m",
+      "\u001b[0;31mRuntimeError\u001b[0m                              Traceback (most recent call last)",
+      "Cell \u001b[0;32mIn[2], line 3\u001b[0m\n\u001b[1;32m      1\u001b[0m \u001b[38;5;28;01mimport\u001b[39;00m \u001b[38;5;21;01mtimm\u001b[39;00m\n\u001b[0;32m----> 3\u001b[0m model \u001b[38;5;241m=\u001b[39m timm\u001b[38;5;241m.\u001b[39mcreate_model(\u001b[38;5;124m\"\u001b[39m\u001b[38;5;124mhf_hub:paige-ai/Virchow2\u001b[39m\u001b[38;5;124m\"\u001b[39m, pretrained\u001b[38;5;241m=\u001b[39m\u001b[38;5;28;01mTrue\u001b[39;00m)\n",
+      "File \u001b[0;32m~/.local/lib/python3.12/site-packages/timm/models/_factory.py:126\u001b[0m, in \u001b[0;36mcreate_model\u001b[0;34m(model_name, pretrained, pretrained_cfg, pretrained_cfg_overlay, checkpoint_path, cache_dir, scriptable, exportable, no_jit, **kwargs)\u001b[0m\n\u001b[1;32m    124\u001b[0m create_fn \u001b[38;5;241m=\u001b[39m model_entrypoint(model_name)\n\u001b[1;32m    125\u001b[0m \u001b[38;5;28;01mwith\u001b[39;00m set_layer_config(scriptable\u001b[38;5;241m=\u001b[39mscriptable, exportable\u001b[38;5;241m=\u001b[39mexportable, no_jit\u001b[38;5;241m=\u001b[39mno_jit):\n\u001b[0;32m--> 126\u001b[0m     model \u001b[38;5;241m=\u001b[39m create_fn(\n\u001b[1;32m    127\u001b[0m         pretrained\u001b[38;5;241m=\u001b[39mpretrained,\n\u001b[1;32m    128\u001b[0m         pretrained_cfg\u001b[38;5;241m=\u001b[39mpretrained_cfg,\n\u001b[1;32m    129\u001b[0m         pretrained_cfg_overlay\u001b[38;5;241m=\u001b[39mpretrained_cfg_overlay,\n\u001b[1;32m    130\u001b[0m         cache_dir\u001b[38;5;241m=\u001b[39mcache_dir,\n\u001b[1;32m    131\u001b[0m         \u001b[38;5;241m*\u001b[39m\u001b[38;5;241m*\u001b[39mkwargs,\n\u001b[1;32m    132\u001b[0m     )\n\u001b[1;32m    134\u001b[0m \u001b[38;5;28;01mif\u001b[39;00m checkpoint_path:\n\u001b[1;32m    135\u001b[0m     load_checkpoint(model, checkpoint_path)\n",
+      "File \u001b[0;32m~/.local/lib/python3.12/site-packages/timm/models/vision_transformer.py:2583\u001b[0m, in \u001b[0;36mvit_huge_patch14_224\u001b[0;34m(pretrained, **kwargs)\u001b[0m\n\u001b[1;32m   2580\u001b[0m \u001b[38;5;250m\u001b[39m\u001b[38;5;124;03m\"\"\" ViT-Huge model (ViT-H/14) from original paper (https://arxiv.org/abs/2010.11929).\u001b[39;00m\n\u001b[1;32m   2581\u001b[0m \u001b[38;5;124;03m\"\"\"\u001b[39;00m\n\u001b[1;32m   2582\u001b[0m model_args \u001b[38;5;241m=\u001b[39m \u001b[38;5;28mdict\u001b[39m(patch_size\u001b[38;5;241m=\u001b[39m\u001b[38;5;241m14\u001b[39m, embed_dim\u001b[38;5;241m=\u001b[39m\u001b[38;5;241m1280\u001b[39m, depth\u001b[38;5;241m=\u001b[39m\u001b[38;5;241m32\u001b[39m, num_heads\u001b[38;5;241m=\u001b[39m\u001b[38;5;241m16\u001b[39m)\n\u001b[0;32m-> 2583\u001b[0m model \u001b[38;5;241m=\u001b[39m _create_vision_transformer(\u001b[38;5;124m'\u001b[39m\u001b[38;5;124mvit_huge_patch14_224\u001b[39m\u001b[38;5;124m'\u001b[39m, pretrained\u001b[38;5;241m=\u001b[39mpretrained, \u001b[38;5;241m*\u001b[39m\u001b[38;5;241m*\u001b[39m\u001b[38;5;28mdict\u001b[39m(model_args, \u001b[38;5;241m*\u001b[39m\u001b[38;5;241m*\u001b[39mkwargs))\n\u001b[1;32m   2584\u001b[0m \u001b[38;5;28;01mreturn\u001b[39;00m model\n",
+      "File \u001b[0;32m~/.local/lib/python3.12/site-packages/timm/models/vision_transformer.py:2406\u001b[0m, in \u001b[0;36m_create_vision_transformer\u001b[0;34m(variant, pretrained, **kwargs)\u001b[0m\n\u001b[1;32m   2403\u001b[0m \u001b[38;5;28;01mif\u001b[39;00m \u001b[38;5;124m'\u001b[39m\u001b[38;5;124msiglip\u001b[39m\u001b[38;5;124m'\u001b[39m \u001b[38;5;129;01min\u001b[39;00m variant \u001b[38;5;129;01mand\u001b[39;00m kwargs\u001b[38;5;241m.\u001b[39mget(\u001b[38;5;124m'\u001b[39m\u001b[38;5;124mglobal_pool\u001b[39m\u001b[38;5;124m'\u001b[39m, \u001b[38;5;28;01mNone\u001b[39;00m) \u001b[38;5;241m!=\u001b[39m \u001b[38;5;124m'\u001b[39m\u001b[38;5;124mmap\u001b[39m\u001b[38;5;124m'\u001b[39m:\n\u001b[1;32m   2404\u001b[0m     strict \u001b[38;5;241m=\u001b[39m \u001b[38;5;28;01mFalse\u001b[39;00m\n\u001b[0;32m-> 2406\u001b[0m \u001b[38;5;28;01mreturn\u001b[39;00m build_model_with_cfg(\n\u001b[1;32m   2407\u001b[0m     VisionTransformer,\n\u001b[1;32m   2408\u001b[0m     variant,\n\u001b[1;32m   2409\u001b[0m     pretrained,\n\u001b[1;32m   2410\u001b[0m     pretrained_filter_fn\u001b[38;5;241m=\u001b[39m_filter_fn,\n\u001b[1;32m   2411\u001b[0m     pretrained_strict\u001b[38;5;241m=\u001b[39mstrict,\n\u001b[1;32m   2412\u001b[0m     feature_cfg\u001b[38;5;241m=\u001b[39m\u001b[38;5;28mdict\u001b[39m(out_indices\u001b[38;5;241m=\u001b[39mout_indices, feature_cls\u001b[38;5;241m=\u001b[39m\u001b[38;5;124m'\u001b[39m\u001b[38;5;124mgetter\u001b[39m\u001b[38;5;124m'\u001b[39m),\n\u001b[1;32m   2413\u001b[0m     \u001b[38;5;241m*\u001b[39m\u001b[38;5;241m*\u001b[39mkwargs,\n\u001b[1;32m   2414\u001b[0m )\n",
+      "File \u001b[0;32m~/.local/lib/python3.12/site-packages/timm/models/_builder.py:436\u001b[0m, in \u001b[0;36mbuild_model_with_cfg\u001b[0;34m(model_cls, variant, pretrained, pretrained_cfg, pretrained_cfg_overlay, model_cfg, feature_cfg, pretrained_strict, pretrained_filter_fn, cache_dir, kwargs_filter, **kwargs)\u001b[0m\n\u001b[1;32m    434\u001b[0m num_classes_pretrained \u001b[38;5;241m=\u001b[39m \u001b[38;5;241m0\u001b[39m \u001b[38;5;28;01mif\u001b[39;00m features \u001b[38;5;28;01melse\u001b[39;00m \u001b[38;5;28mgetattr\u001b[39m(model, \u001b[38;5;124m'\u001b[39m\u001b[38;5;124mnum_classes\u001b[39m\u001b[38;5;124m'\u001b[39m, kwargs\u001b[38;5;241m.\u001b[39mget(\u001b[38;5;124m'\u001b[39m\u001b[38;5;124mnum_classes\u001b[39m\u001b[38;5;124m'\u001b[39m, \u001b[38;5;241m1000\u001b[39m))\n\u001b[1;32m    435\u001b[0m \u001b[38;5;28;01mif\u001b[39;00m pretrained:\n\u001b[0;32m--> 436\u001b[0m     load_pretrained(\n\u001b[1;32m    437\u001b[0m         model,\n\u001b[1;32m    438\u001b[0m         pretrained_cfg\u001b[38;5;241m=\u001b[39mpretrained_cfg,\n\u001b[1;32m    439\u001b[0m         num_classes\u001b[38;5;241m=\u001b[39mnum_classes_pretrained,\n\u001b[1;32m    440\u001b[0m         in_chans\u001b[38;5;241m=\u001b[39mkwargs\u001b[38;5;241m.\u001b[39mget(\u001b[38;5;124m'\u001b[39m\u001b[38;5;124min_chans\u001b[39m\u001b[38;5;124m'\u001b[39m, \u001b[38;5;241m3\u001b[39m),\n\u001b[1;32m    441\u001b[0m         filter_fn\u001b[38;5;241m=\u001b[39mpretrained_filter_fn,\n\u001b[1;32m    442\u001b[0m         strict\u001b[38;5;241m=\u001b[39mpretrained_strict,\n\u001b[1;32m    443\u001b[0m         cache_dir\u001b[38;5;241m=\u001b[39mcache_dir,\n\u001b[1;32m    444\u001b[0m     )\n\u001b[1;32m    446\u001b[0m \u001b[38;5;66;03m# Wrap the model in a feature extraction module if enabled\u001b[39;00m\n\u001b[1;32m    447\u001b[0m \u001b[38;5;28;01mif\u001b[39;00m features:\n",
+      "File \u001b[0;32m~/.local/lib/python3.12/site-packages/timm/models/_builder.py:260\u001b[0m, in \u001b[0;36mload_pretrained\u001b[0;34m(model, pretrained_cfg, num_classes, in_chans, filter_fn, strict, cache_dir)\u001b[0m\n\u001b[1;32m    257\u001b[0m             classifier_bias \u001b[38;5;241m=\u001b[39m state_dict[classifier_name \u001b[38;5;241m+\u001b[39m \u001b[38;5;124m'\u001b[39m\u001b[38;5;124m.bias\u001b[39m\u001b[38;5;124m'\u001b[39m]\n\u001b[1;32m    258\u001b[0m             state_dict[classifier_name \u001b[38;5;241m+\u001b[39m \u001b[38;5;124m'\u001b[39m\u001b[38;5;124m.bias\u001b[39m\u001b[38;5;124m'\u001b[39m] \u001b[38;5;241m=\u001b[39m classifier_bias[label_offset:]\n\u001b[0;32m--> 260\u001b[0m load_result \u001b[38;5;241m=\u001b[39m model\u001b[38;5;241m.\u001b[39mload_state_dict(state_dict, strict\u001b[38;5;241m=\u001b[39mstrict)\n\u001b[1;32m    261\u001b[0m \u001b[38;5;28;01mif\u001b[39;00m load_result\u001b[38;5;241m.\u001b[39mmissing_keys:\n\u001b[1;32m    262\u001b[0m     _logger\u001b[38;5;241m.\u001b[39minfo(\n\u001b[1;32m    263\u001b[0m         \u001b[38;5;124mf\u001b[39m\u001b[38;5;124m'\u001b[39m\u001b[38;5;124mMissing keys (\u001b[39m\u001b[38;5;132;01m{\u001b[39;00m\u001b[38;5;124m\"\u001b[39m\u001b[38;5;124m, \u001b[39m\u001b[38;5;124m\"\u001b[39m\u001b[38;5;241m.\u001b[39mjoin(load_result\u001b[38;5;241m.\u001b[39mmissing_keys)\u001b[38;5;132;01m}\u001b[39;00m\u001b[38;5;124m) discovered while loading pretrained weights.\u001b[39m\u001b[38;5;124m'\u001b[39m\n\u001b[1;32m    264\u001b[0m         \u001b[38;5;124mf\u001b[39m\u001b[38;5;124m'\u001b[39m\u001b[38;5;124m This is expected if model is being adapted.\u001b[39m\u001b[38;5;124m'\u001b[39m)\n",
+      "File \u001b[0;32m~/.local/lib/python3.12/site-packages/torch/nn/modules/module.py:2581\u001b[0m, in \u001b[0;36mModule.load_state_dict\u001b[0;34m(self, state_dict, strict, assign)\u001b[0m\n\u001b[1;32m   2573\u001b[0m         error_msgs\u001b[38;5;241m.\u001b[39minsert(\n\u001b[1;32m   2574\u001b[0m             \u001b[38;5;241m0\u001b[39m,\n\u001b[1;32m   2575\u001b[0m             \u001b[38;5;124m\"\u001b[39m\u001b[38;5;124mMissing key(s) in state_dict: \u001b[39m\u001b[38;5;132;01m{}\u001b[39;00m\u001b[38;5;124m. \u001b[39m\u001b[38;5;124m\"\u001b[39m\u001b[38;5;241m.\u001b[39mformat(\n\u001b[1;32m   2576\u001b[0m                 \u001b[38;5;124m\"\u001b[39m\u001b[38;5;124m, \u001b[39m\u001b[38;5;124m\"\u001b[39m\u001b[38;5;241m.\u001b[39mjoin(\u001b[38;5;124mf\u001b[39m\u001b[38;5;124m'\u001b[39m\u001b[38;5;124m\"\u001b[39m\u001b[38;5;132;01m{\u001b[39;00mk\u001b[38;5;132;01m}\u001b[39;00m\u001b[38;5;124m\"\u001b[39m\u001b[38;5;124m'\u001b[39m \u001b[38;5;28;01mfor\u001b[39;00m k \u001b[38;5;129;01min\u001b[39;00m missing_keys)\n\u001b[1;32m   2577\u001b[0m             ),\n\u001b[1;32m   2578\u001b[0m         )\n\u001b[1;32m   2580\u001b[0m \u001b[38;5;28;01mif\u001b[39;00m \u001b[38;5;28mlen\u001b[39m(error_msgs) \u001b[38;5;241m>\u001b[39m \u001b[38;5;241m0\u001b[39m:\n\u001b[0;32m-> 2581\u001b[0m     \u001b[38;5;28;01mraise\u001b[39;00m \u001b[38;5;167;01mRuntimeError\u001b[39;00m(\n\u001b[1;32m   2582\u001b[0m         \u001b[38;5;124m\"\u001b[39m\u001b[38;5;124mError(s) in loading state_dict for \u001b[39m\u001b[38;5;132;01m{}\u001b[39;00m\u001b[38;5;124m:\u001b[39m\u001b[38;5;130;01m\\n\u001b[39;00m\u001b[38;5;130;01m\\t\u001b[39;00m\u001b[38;5;132;01m{}\u001b[39;00m\u001b[38;5;124m\"\u001b[39m\u001b[38;5;241m.\u001b[39mformat(\n\u001b[1;32m   2583\u001b[0m             \u001b[38;5;28mself\u001b[39m\u001b[38;5;241m.\u001b[39m\u001b[38;5;18m__class__\u001b[39m\u001b[38;5;241m.\u001b[39m\u001b[38;5;18m__name__\u001b[39m, \u001b[38;5;124m\"\u001b[39m\u001b[38;5;130;01m\\n\u001b[39;00m\u001b[38;5;130;01m\\t\u001b[39;00m\u001b[38;5;124m\"\u001b[39m\u001b[38;5;241m.\u001b[39mjoin(error_msgs)\n\u001b[1;32m   2584\u001b[0m         )\n\u001b[1;32m   2585\u001b[0m     )\n\u001b[1;32m   2586\u001b[0m \u001b[38;5;28;01mreturn\u001b[39;00m _IncompatibleKeys(missing_keys, unexpected_keys)\n",
+      "\u001b[0;31mRuntimeError\u001b[0m: Error(s) in loading state_dict for VisionTransformer:\n\tsize mismatch for blocks.0.mlp.fc2.weight: copying a param with shape torch.Size([1280, 3416]) from checkpoint, the shape in current model is torch.Size([1280, 6832]).\n\tsize mismatch for blocks.1.mlp.fc2.weight: copying a param with shape torch.Size([1280, 3416]) from checkpoint, the shape in current model is torch.Size([1280, 6832]).\n\tsize mismatch for blocks.2.mlp.fc2.weight: copying a param with shape torch.Size([1280, 3416]) from checkpoint, the shape in current model is torch.Size([1280, 6832]).\n\tsize mismatch for blocks.3.mlp.fc2.weight: copying a param with shape torch.Size([1280, 3416]) from checkpoint, the shape in current model is torch.Size([1280, 6832]).\n\tsize mismatch for blocks.4.mlp.fc2.weight: copying a param with shape torch.Size([1280, 3416]) from checkpoint, the shape in current model is torch.Size([1280, 6832]).\n\tsize mismatch for blocks.5.mlp.fc2.weight: copying a param with shape torch.Size([1280, 3416]) from checkpoint, the shape in current model is torch.Size([1280, 6832]).\n\tsize mismatch for blocks.6.mlp.fc2.weight: copying a param with shape torch.Size([1280, 3416]) from checkpoint, the shape in current model is torch.Size([1280, 6832]).\n\tsize mismatch for blocks.7.mlp.fc2.weight: copying a param with shape torch.Size([1280, 3416]) from checkpoint, the shape in current model is torch.Size([1280, 6832]).\n\tsize mismatch for blocks.8.mlp.fc2.weight: copying a param with shape torch.Size([1280, 3416]) from checkpoint, the shape in current model is torch.Size([1280, 6832]).\n\tsize mismatch for blocks.9.mlp.fc2.weight: copying a param with shape torch.Size([1280, 3416]) from checkpoint, the shape in current model is torch.Size([1280, 6832]).\n\tsize mismatch for blocks.10.mlp.fc2.weight: copying a param with shape torch.Size([1280, 3416]) from checkpoint, the shape in current model is torch.Size([1280, 6832]).\n\tsize mismatch for blocks.11.mlp.fc2.weight: copying a param with shape torch.Size([1280, 3416]) from checkpoint, the shape in current model is torch.Size([1280, 6832]).\n\tsize mismatch for blocks.12.mlp.fc2.weight: copying a param with shape torch.Size([1280, 3416]) from checkpoint, the shape in current model is torch.Size([1280, 6832]).\n\tsize mismatch for blocks.13.mlp.fc2.weight: copying a param with shape torch.Size([1280, 3416]) from checkpoint, the shape in current model is torch.Size([1280, 6832]).\n\tsize mismatch for blocks.14.mlp.fc2.weight: copying a param with shape torch.Size([1280, 3416]) from checkpoint, the shape in current model is torch.Size([1280, 6832]).\n\tsize mismatch for blocks.15.mlp.fc2.weight: copying a param with shape torch.Size([1280, 3416]) from checkpoint, the shape in current model is torch.Size([1280, 6832]).\n\tsize mismatch for blocks.16.mlp.fc2.weight: copying a param with shape torch.Size([1280, 3416]) from checkpoint, the shape in current model is torch.Size([1280, 6832]).\n\tsize mismatch for blocks.17.mlp.fc2.weight: copying a param with shape torch.Size([1280, 3416]) from checkpoint, the shape in current model is torch.Size([1280, 6832]).\n\tsize mismatch for blocks.18.mlp.fc2.weight: copying a param with shape torch.Size([1280, 3416]) from checkpoint, the shape in current model is torch.Size([1280, 6832]).\n\tsize mismatch for blocks.19.mlp.fc2.weight: copying a param with shape torch.Size([1280, 3416]) from checkpoint, the shape in current model is torch.Size([1280, 6832]).\n\tsize mismatch for blocks.20.mlp.fc2.weight: copying a param with shape torch.Size([1280, 3416]) from checkpoint, the shape in current model is torch.Size([1280, 6832]).\n\tsize mismatch for blocks.21.mlp.fc2.weight: copying a param with shape torch.Size([1280, 3416]) from checkpoint, the shape in current model is torch.Size([1280, 6832]).\n\tsize mismatch for blocks.22.mlp.fc2.weight: copying a param with shape torch.Size([1280, 3416]) from checkpoint, the shape in current model is torch.Size([1280, 6832]).\n\tsize mismatch for blocks.23.mlp.fc2.weight: copying a param with shape torch.Size([1280, 3416]) from checkpoint, the shape in current model is torch.Size([1280, 6832]).\n\tsize mismatch for blocks.24.mlp.fc2.weight: copying a param with shape torch.Size([1280, 3416]) from checkpoint, the shape in current model is torch.Size([1280, 6832]).\n\tsize mismatch for blocks.25.mlp.fc2.weight: copying a param with shape torch.Size([1280, 3416]) from checkpoint, the shape in current model is torch.Size([1280, 6832]).\n\tsize mismatch for blocks.26.mlp.fc2.weight: copying a param with shape torch.Size([1280, 3416]) from checkpoint, the shape in current model is torch.Size([1280, 6832]).\n\tsize mismatch for blocks.27.mlp.fc2.weight: copying a param with shape torch.Size([1280, 3416]) from checkpoint, the shape in current model is torch.Size([1280, 6832]).\n\tsize mismatch for blocks.28.mlp.fc2.weight: copying a param with shape torch.Size([1280, 3416]) from checkpoint, the shape in current model is torch.Size([1280, 6832]).\n\tsize mismatch for blocks.29.mlp.fc2.weight: copying a param with shape torch.Size([1280, 3416]) from checkpoint, the shape in current model is torch.Size([1280, 6832]).\n\tsize mismatch for blocks.30.mlp.fc2.weight: copying a param with shape torch.Size([1280, 3416]) from checkpoint, the shape in current model is torch.Size([1280, 6832]).\n\tsize mismatch for blocks.31.mlp.fc2.weight: copying a param with shape torch.Size([1280, 3416]) from checkpoint, the shape in current model is torch.Size([1280, 6832])."
+     ]
+    }
+   ],
+   "source": [
+    "import timm\n",
+    "\n",
+    "model = timm.create_model(\"hf_hub:paige-ai/Virchow2\", pretrained=True)"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": 3,
+   "id": "2999e249",
+   "metadata": {},
+   "outputs": [
+    {
+     "name": "stderr",
+     "output_type": "stream",
+     "text": [
+      "2025-04-18 16:09:01.177109: E external/local_xla/xla/stream_executor/cuda/cuda_fft.cc:467] Unable to register cuFFT factory: Attempting to register factory for plugin cuFFT when one has already been registered\n",
+      "WARNING: All log messages before absl::InitializeLog() is called are written to STDERR\n",
+      "E0000 00:00:1744967341.201028 2970384 cuda_dnn.cc:8579] Unable to register cuDNN factory: Attempting to register factory for plugin cuDNN when one has already been registered\n",
+      "E0000 00:00:1744967341.209888 2970384 cuda_blas.cc:1407] Unable to register cuBLAS factory: Attempting to register factory for plugin cuBLAS when one has already been registered\n",
+      "W0000 00:00:1744967341.235808 2970384 computation_placer.cc:177] computation placer already registered. Please check linkage and avoid linking the same target more than once.\n",
+      "W0000 00:00:1744967341.235834 2970384 computation_placer.cc:177] computation placer already registered. Please check linkage and avoid linking the same target more than once.\n",
+      "W0000 00:00:1744967341.235838 2970384 computation_placer.cc:177] computation placer already registered. Please check linkage and avoid linking the same target more than once.\n",
+      "W0000 00:00:1744967341.235840 2970384 computation_placer.cc:177] computation placer already registered. Please check linkage and avoid linking the same target more than once.\n",
+      "2025-04-18 16:09:01.243585: I tensorflow/core/platform/cpu_feature_guard.cc:210] This TensorFlow binary is optimized to use available CPU instructions in performance-critical operations.\n",
+      "To enable the following instructions: AVX2 FMA, in other operations, rebuild TensorFlow with the appropriate compiler flags.\n"
+     ]
+    },
+    {
+     "ename": "TypeError",
+     "evalue": "'NoneType' object cannot be interpreted as an integer",
+     "output_type": "error",
+     "traceback": [
+      "\u001b[0;31m---------------------------------------------------------------------------\u001b[0m",
+      "\u001b[0;31mTypeError\u001b[0m                                 Traceback (most recent call last)",
+      "Cell \u001b[0;32mIn[3], line 4\u001b[0m\n\u001b[1;32m      1\u001b[0m \u001b[38;5;28;01mfrom\u001b[39;00m \u001b[38;5;21;01mtransformers\u001b[39;00m \u001b[38;5;28;01mimport\u001b[39;00m AutoModel, AutoImageProcessor\n\u001b[1;32m      3\u001b[0m \u001b[38;5;66;03m# Load the model directly from Hugging Face\u001b[39;00m\n\u001b[0;32m----> 4\u001b[0m model \u001b[38;5;241m=\u001b[39m AutoModel\u001b[38;5;241m.\u001b[39mfrom_pretrained(\u001b[38;5;124m\"\u001b[39m\u001b[38;5;124mpaige-ai/Virchow2\u001b[39m\u001b[38;5;124m\"\u001b[39m)\n\u001b[1;32m      5\u001b[0m processor \u001b[38;5;241m=\u001b[39m AutoImageProcessor\u001b[38;5;241m.\u001b[39mfrom_pretrained(\u001b[38;5;124m\"\u001b[39m\u001b[38;5;124mpaige-ai/Virchow2\u001b[39m\u001b[38;5;124m\"\u001b[39m)\n",
+      "File \u001b[0;32m~/.local/lib/python3.12/site-packages/transformers/models/auto/auto_factory.py:531\u001b[0m, in \u001b[0;36m_BaseAutoModelClass.from_pretrained\u001b[0;34m(cls, pretrained_model_name_or_path, *model_args, **kwargs)\u001b[0m\n\u001b[1;32m    528\u001b[0m \u001b[38;5;28;01mif\u001b[39;00m kwargs\u001b[38;5;241m.\u001b[39mget(\u001b[38;5;124m\"\u001b[39m\u001b[38;5;124mquantization_config\u001b[39m\u001b[38;5;124m\"\u001b[39m, \u001b[38;5;28;01mNone\u001b[39;00m) \u001b[38;5;129;01mis\u001b[39;00m \u001b[38;5;129;01mnot\u001b[39;00m \u001b[38;5;28;01mNone\u001b[39;00m:\n\u001b[1;32m    529\u001b[0m     _ \u001b[38;5;241m=\u001b[39m kwargs\u001b[38;5;241m.\u001b[39mpop(\u001b[38;5;124m\"\u001b[39m\u001b[38;5;124mquantization_config\u001b[39m\u001b[38;5;124m\"\u001b[39m)\n\u001b[0;32m--> 531\u001b[0m config, kwargs \u001b[38;5;241m=\u001b[39m AutoConfig\u001b[38;5;241m.\u001b[39mfrom_pretrained(\n\u001b[1;32m    532\u001b[0m     pretrained_model_name_or_path,\n\u001b[1;32m    533\u001b[0m     return_unused_kwargs\u001b[38;5;241m=\u001b[39m\u001b[38;5;28;01mTrue\u001b[39;00m,\n\u001b[1;32m    534\u001b[0m     trust_remote_code\u001b[38;5;241m=\u001b[39mtrust_remote_code,\n\u001b[1;32m    535\u001b[0m     code_revision\u001b[38;5;241m=\u001b[39mcode_revision,\n\u001b[1;32m    536\u001b[0m     _commit_hash\u001b[38;5;241m=\u001b[39mcommit_hash,\n\u001b[1;32m    537\u001b[0m     \u001b[38;5;241m*\u001b[39m\u001b[38;5;241m*\u001b[39mhub_kwargs,\n\u001b[1;32m    538\u001b[0m     \u001b[38;5;241m*\u001b[39m\u001b[38;5;241m*\u001b[39mkwargs,\n\u001b[1;32m    539\u001b[0m )\n\u001b[1;32m    541\u001b[0m \u001b[38;5;66;03m# if torch_dtype=auto was passed here, ensure to pass it on\u001b[39;00m\n\u001b[1;32m    542\u001b[0m \u001b[38;5;28;01mif\u001b[39;00m kwargs_orig\u001b[38;5;241m.\u001b[39mget(\u001b[38;5;124m\"\u001b[39m\u001b[38;5;124mtorch_dtype\u001b[39m\u001b[38;5;124m\"\u001b[39m, \u001b[38;5;28;01mNone\u001b[39;00m) \u001b[38;5;241m==\u001b[39m \u001b[38;5;124m\"\u001b[39m\u001b[38;5;124mauto\u001b[39m\u001b[38;5;124m\"\u001b[39m:\n",
+      "File \u001b[0;32m~/.local/lib/python3.12/site-packages/transformers/models/auto/configuration_auto.py:1125\u001b[0m, in \u001b[0;36mAutoConfig.from_pretrained\u001b[0;34m(cls, pretrained_model_name_or_path, **kwargs)\u001b[0m\n\u001b[1;32m   1114\u001b[0m     \u001b[38;5;28;01mexcept\u001b[39;00m \u001b[38;5;167;01mKeyError\u001b[39;00m:\n\u001b[1;32m   1115\u001b[0m         \u001b[38;5;28;01mraise\u001b[39;00m \u001b[38;5;167;01mValueError\u001b[39;00m(\n\u001b[1;32m   1116\u001b[0m             \u001b[38;5;124mf\u001b[39m\u001b[38;5;124m\"\u001b[39m\u001b[38;5;124mThe checkpoint you are trying to load has model type `\u001b[39m\u001b[38;5;132;01m{\u001b[39;00mconfig_dict[\u001b[38;5;124m'\u001b[39m\u001b[38;5;124mmodel_type\u001b[39m\u001b[38;5;124m'\u001b[39m]\u001b[38;5;132;01m}\u001b[39;00m\u001b[38;5;124m` \u001b[39m\u001b[38;5;124m\"\u001b[39m\n\u001b[1;32m   1117\u001b[0m             \u001b[38;5;124m\"\u001b[39m\u001b[38;5;124mbut Transformers does not recognize this architecture. This could be because of an \u001b[39m\u001b[38;5;124m\"\u001b[39m\n\u001b[0;32m   (...)\u001b[0m\n\u001b[1;32m   1123\u001b[0m             \u001b[38;5;124m\"\u001b[39m\u001b[38;5;124m`pip install git+https://github.com/huggingface/transformers.git`\u001b[39m\u001b[38;5;124m\"\u001b[39m\n\u001b[1;32m   1124\u001b[0m         )\n\u001b[0;32m-> 1125\u001b[0m     \u001b[38;5;28;01mreturn\u001b[39;00m config_class\u001b[38;5;241m.\u001b[39mfrom_dict(config_dict, \u001b[38;5;241m*\u001b[39m\u001b[38;5;241m*\u001b[39munused_kwargs)\n\u001b[1;32m   1126\u001b[0m \u001b[38;5;28;01melse\u001b[39;00m:\n\u001b[1;32m   1127\u001b[0m     \u001b[38;5;66;03m# Fallback: use pattern matching on the string.\u001b[39;00m\n\u001b[1;32m   1128\u001b[0m     \u001b[38;5;66;03m# We go from longer names to shorter names to catch roberta before bert (for instance)\u001b[39;00m\n\u001b[1;32m   1129\u001b[0m     \u001b[38;5;28;01mfor\u001b[39;00m pattern \u001b[38;5;129;01min\u001b[39;00m \u001b[38;5;28msorted\u001b[39m(CONFIG_MAPPING\u001b[38;5;241m.\u001b[39mkeys(), key\u001b[38;5;241m=\u001b[39m\u001b[38;5;28mlen\u001b[39m, reverse\u001b[38;5;241m=\u001b[39m\u001b[38;5;28;01mTrue\u001b[39;00m):\n",
+      "File \u001b[0;32m~/.local/lib/python3.12/site-packages/transformers/models/timm_wrapper/configuration_timm_wrapper.py:106\u001b[0m, in \u001b[0;36mTimmWrapperConfig.from_dict\u001b[0;34m(cls, config_dict, **kwargs)\u001b[0m\n\u001b[1;32m    103\u001b[0m \u001b[38;5;28;01mif\u001b[39;00m \u001b[38;5;124m\"\u001b[39m\u001b[38;5;124mpretrained_cfg\u001b[39m\u001b[38;5;124m\"\u001b[39m \u001b[38;5;129;01min\u001b[39;00m config_dict \u001b[38;5;129;01mand\u001b[39;00m \u001b[38;5;124m\"\u001b[39m\u001b[38;5;124mnum_classes\u001b[39m\u001b[38;5;124m\"\u001b[39m \u001b[38;5;129;01min\u001b[39;00m config_dict[\u001b[38;5;124m\"\u001b[39m\u001b[38;5;124mpretrained_cfg\u001b[39m\u001b[38;5;124m\"\u001b[39m]:\n\u001b[1;32m    104\u001b[0m     config_dict[\u001b[38;5;124m\"\u001b[39m\u001b[38;5;124mpretrained_cfg\u001b[39m\u001b[38;5;124m\"\u001b[39m]\u001b[38;5;241m.\u001b[39mpop(\u001b[38;5;124m\"\u001b[39m\u001b[38;5;124mnum_classes\u001b[39m\u001b[38;5;124m\"\u001b[39m, \u001b[38;5;28;01mNone\u001b[39;00m)\n\u001b[0;32m--> 106\u001b[0m \u001b[38;5;28;01mreturn\u001b[39;00m \u001b[38;5;28msuper\u001b[39m()\u001b[38;5;241m.\u001b[39mfrom_dict(config_dict, \u001b[38;5;241m*\u001b[39m\u001b[38;5;241m*\u001b[39mkwargs)\n",
+      "File \u001b[0;32m~/.local/lib/python3.12/site-packages/transformers/configuration_utils.py:765\u001b[0m, in \u001b[0;36mPretrainedConfig.from_dict\u001b[0;34m(cls, config_dict, **kwargs)\u001b[0m\n\u001b[1;32m    763\u001b[0m \u001b[38;5;28;01mif\u001b[39;00m \u001b[38;5;28misinstance\u001b[39m(current_attr, PretrainedConfig) \u001b[38;5;129;01mand\u001b[39;00m \u001b[38;5;28misinstance\u001b[39m(value, \u001b[38;5;28mdict\u001b[39m):\n\u001b[1;32m    764\u001b[0m     value \u001b[38;5;241m=\u001b[39m current_attr\u001b[38;5;241m.\u001b[39m\u001b[38;5;18m__class__\u001b[39m(\u001b[38;5;241m*\u001b[39m\u001b[38;5;241m*\u001b[39mvalue)\n\u001b[0;32m--> 765\u001b[0m \u001b[38;5;28msetattr\u001b[39m(config, key, value)\n\u001b[1;32m    766\u001b[0m \u001b[38;5;28;01mif\u001b[39;00m key \u001b[38;5;241m!=\u001b[39m \u001b[38;5;124m\"\u001b[39m\u001b[38;5;124mtorch_dtype\u001b[39m\u001b[38;5;124m\"\u001b[39m:\n\u001b[1;32m    767\u001b[0m     to_remove\u001b[38;5;241m.\u001b[39mappend(key)\n",
+      "File \u001b[0;32m~/.local/lib/python3.12/site-packages/transformers/configuration_utils.py:209\u001b[0m, in \u001b[0;36mPretrainedConfig.__setattr__\u001b[0;34m(self, key, value)\u001b[0m\n\u001b[1;32m    207\u001b[0m \u001b[38;5;28;01mif\u001b[39;00m key \u001b[38;5;129;01min\u001b[39;00m \u001b[38;5;28msuper\u001b[39m()\u001b[38;5;241m.\u001b[39m\u001b[38;5;21m__getattribute__\u001b[39m(\u001b[38;5;124m\"\u001b[39m\u001b[38;5;124mattribute_map\u001b[39m\u001b[38;5;124m\"\u001b[39m):\n\u001b[1;32m    208\u001b[0m     key \u001b[38;5;241m=\u001b[39m \u001b[38;5;28msuper\u001b[39m()\u001b[38;5;241m.\u001b[39m\u001b[38;5;21m__getattribute__\u001b[39m(\u001b[38;5;124m\"\u001b[39m\u001b[38;5;124mattribute_map\u001b[39m\u001b[38;5;124m\"\u001b[39m)[key]\n\u001b[0;32m--> 209\u001b[0m \u001b[38;5;28msuper\u001b[39m()\u001b[38;5;241m.\u001b[39m\u001b[38;5;21m__setattr__\u001b[39m(key, value)\n",
+      "File \u001b[0;32m~/.local/lib/python3.12/site-packages/transformers/configuration_utils.py:355\u001b[0m, in \u001b[0;36mPretrainedConfig.num_labels\u001b[0;34m(self, num_labels)\u001b[0m\n\u001b[1;32m    352\u001b[0m \u001b[38;5;129m@num_labels\u001b[39m\u001b[38;5;241m.\u001b[39msetter\n\u001b[1;32m    353\u001b[0m \u001b[38;5;28;01mdef\u001b[39;00m \u001b[38;5;21mnum_labels\u001b[39m(\u001b[38;5;28mself\u001b[39m, num_labels: \u001b[38;5;28mint\u001b[39m):\n\u001b[1;32m    354\u001b[0m     \u001b[38;5;28;01mif\u001b[39;00m \u001b[38;5;129;01mnot\u001b[39;00m \u001b[38;5;28mhasattr\u001b[39m(\u001b[38;5;28mself\u001b[39m, \u001b[38;5;124m\"\u001b[39m\u001b[38;5;124mid2label\u001b[39m\u001b[38;5;124m\"\u001b[39m) \u001b[38;5;129;01mor\u001b[39;00m \u001b[38;5;28mself\u001b[39m\u001b[38;5;241m.\u001b[39mid2label \u001b[38;5;129;01mis\u001b[39;00m \u001b[38;5;28;01mNone\u001b[39;00m \u001b[38;5;129;01mor\u001b[39;00m \u001b[38;5;28mlen\u001b[39m(\u001b[38;5;28mself\u001b[39m\u001b[38;5;241m.\u001b[39mid2label) \u001b[38;5;241m!=\u001b[39m num_labels:\n\u001b[0;32m--> 355\u001b[0m         \u001b[38;5;28mself\u001b[39m\u001b[38;5;241m.\u001b[39mid2label \u001b[38;5;241m=\u001b[39m {i: \u001b[38;5;124mf\u001b[39m\u001b[38;5;124m\"\u001b[39m\u001b[38;5;124mLABEL_\u001b[39m\u001b[38;5;132;01m{\u001b[39;00mi\u001b[38;5;132;01m}\u001b[39;00m\u001b[38;5;124m\"\u001b[39m \u001b[38;5;28;01mfor\u001b[39;00m i \u001b[38;5;129;01min\u001b[39;00m \u001b[38;5;28mrange\u001b[39m(num_labels)}\n\u001b[1;32m    356\u001b[0m         \u001b[38;5;28mself\u001b[39m\u001b[38;5;241m.\u001b[39mlabel2id \u001b[38;5;241m=\u001b[39m \u001b[38;5;28mdict\u001b[39m(\u001b[38;5;28mzip\u001b[39m(\u001b[38;5;28mself\u001b[39m\u001b[38;5;241m.\u001b[39mid2label\u001b[38;5;241m.\u001b[39mvalues(), \u001b[38;5;28mself\u001b[39m\u001b[38;5;241m.\u001b[39mid2label\u001b[38;5;241m.\u001b[39mkeys()))\n",
+      "\u001b[0;31mTypeError\u001b[0m: 'NoneType' object cannot be interpreted as an integer"
+     ]
+    }
+   ],
+   "source": [
+    "from transformers import AutoModel, AutoImageProcessor\n",
+    "\n",
+    "# Load the model directly from Hugging Face\n",
+    "model = AutoModel.from_pretrained(\"paige-ai/Virchow2\")\n",
+    "processor = AutoImageProcessor.from_pretrained(\"paige-ai/Virchow2\")"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": null,
+   "id": "173d0ea6",
+   "metadata": {},
+   "outputs": [],
+   "source": [
+    "import timm\n",
+    "import torch\n",
+    "from timm.data import resolve_data_config\n",
+    "from timm.data.transforms_factory import create_transform\n",
+    "from timm.layers import SwiGLUPacked\n",
+    "from PIL import Image\n",
+    "\n",
+    "# need to specify MLP layer and activation function for proper init\n",
+    "model = timm.create_model(\"hf-hub:paige-ai/Virchow2\", pretrained=True, mlp_layer=SwiGLUPacked, act_layer=torch.nn.SiLU)\n",
+    "model = model.eval()\n",
+    "\n",
+    "transforms = create_transform(**resolve_data_config(model.pretrained_cfg, model=model))\n",
+    "\n",
+    "image = Image.open(r\"/home/ubuntu/vnet/TaoST/Brand.jpg\")\n",
+    "image = transforms(image).unsqueeze(0)  # size: 1 x 3 x 224 x 224\n",
+    "\n",
+    "output = model(image)  # size: 1 x 261 x 1280\n",
+    "\n",
+    "class_token = output[:, 0]    # size: 1 x 1280\n",
+    "patch_tokens = output[:, 5:]  # size: 1 x 256 x 1280, tokens 1-4 are register tokens so we ignore those\n",
+    "\n",
+    "# concatenate class token and average pool of patch tokens\n",
+    "embedding = torch.cat([class_token, patch_tokens.mean(1)], dim=-1)  # size: 1 x 2560"
+   ]
+  }
+ ],
+ "metadata": {
+  "kernelspec": {
+   "display_name": "base",
+   "language": "python",
+   "name": "python3"
+  },
+  "language_info": {
+   "codemirror_mode": {
+    "name": "ipython",
+    "version": 3
+   },
+   "file_extension": ".py",
+   "mimetype": "text/x-python",
+   "name": "python",
+   "nbconvert_exporter": "python",
+   "pygments_lexer": "ipython3",
+   "version": "3.12.7"
+  }
+ },
+ "nbformat": 4,
+ "nbformat_minor": 5
+}

Virchow2

@@ -0,0 +1 @@
+Subproject commit 3158645804b69e3f3bc4439d4116edddf0840a72