fix test notebook

finetuning.ipynb

@@ -940,7 +940,7 @@
   },
   {
    "cell_type": "code",
-   "execution_count": 32,
+   "execution_count": 33,
    "metadata": {},
    "outputs": [
     {
@@ -951,21 +951,50 @@
       "remote: Enforcing permissions...        \n",
       "remote: Allowed refs: all        \n",
       "To https://huggingface.co/ChainYo/segformer-sidewalk\n",
-      "   5d5f276..56db83f  main -> main\n",
-      "\n"
+      "   56db83f..fcb528d  main -> main\n",
+      "\n",
+      "Some weights of the model checkpoint at /home/chainyo/code/segformer-sidewalk/checkpoints/epoch=44-step=1125.ckpt were not used when initializing SegformerForSemanticSegmentation: ['pytorch-lightning_version', 'epoch', 'hyper_parameters', 'optimizer_states', 'loops', 'global_step', 'state_dict', 'callbacks', 'lr_schedulers', 'hparams_name']\n",
+      "- This IS expected if you are initializing SegformerForSemanticSegmentation from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model).\n",
+      "- This IS NOT expected if you are initializing SegformerForSemanticSegmentation from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model).\n",
+      "Some weights of SegformerForSemanticSegmentation were not initialized from the model checkpoint at /home/chainyo/code/segformer-sidewalk/checkpoints/epoch=44-step=1125.ckpt and are newly initialized: ['encoder.block.1.0.mlp.dwconv.dwconv.weight', 'encoder.block.1.0.layer_norm_2.bias', 'encoder.block.1.0.attention.output.dense.weight', 'encoder.block.0.0.mlp.dwconv.dwconv.bias', 'decode_head.batch_norm.running_var', 'decode_head.linear_c.1.proj.weight', 'encoder.block.1.0.mlp.dense1.weight', 'encoder.block.3.1.attention.self.value.weight', 'encoder.block.2.1.attention.output.dense.weight', 'encoder.block.3.1.mlp.dense1.bias', 'encoder.block.1.1.attention.self.sr.weight', 'encoder.block.2.1.attention.self.sr.bias', 'encoder.block.2.0.attention.self.value.weight', 'encoder.block.1.0.attention.self.sr.bias', 'encoder.block.2.0.layer_norm_2.bias', 'encoder.block.0.1.layer_norm_1.weight', 'encoder.patch_embeddings.2.proj.weight', 'encoder.block.0.1.layer_norm_2.weight', 'encoder.block.3.1.mlp.dwconv.dwconv.bias', 'encoder.block.2.1.layer_norm_2.weight', 'encoder.block.2.1.layer_norm_2.bias', 'encoder.block.3.0.layer_norm_1.bias', 'encoder.block.1.0.attention.self.layer_norm.bias', 'encoder.block.0.1.attention.self.sr.weight', 'encoder.block.2.0.attention.self.sr.weight', 'encoder.block.2.0.attention.self.layer_norm.bias', 'decode_head.classifier.weight', 'encoder.block.3.1.attention.self.query.weight', 'encoder.block.1.0.mlp.dense1.bias', 'encoder.block.0.0.layer_norm_1.bias', 'encoder.block.0.1.mlp.dense1.weight', 'encoder.block.2.0.layer_norm_1.weight', 'encoder.block.0.0.mlp.dense2.weight', 'encoder.block.1.0.layer_norm_2.weight', 'encoder.block.2.1.attention.output.dense.bias', 'encoder.block.0.0.attention.self.value.bias', 'encoder.block.3.0.attention.self.query.bias', 'encoder.block.1.1.attention.self.layer_norm.bias', 'decode_head.linear_c.2.proj.weight', 'encoder.block.3.0.mlp.dense1.weight', 'encoder.block.3.0.mlp.dense1.bias', 'encoder.patch_embeddings.1.layer_norm.bias', 'encoder.block.3.1.mlp.dense2.weight', 'encoder.block.1.0.mlp.dwconv.dwconv.bias', 'encoder.block.1.0.attention.self.query.weight', 'encoder.block.2.0.attention.self.query.bias', 'encoder.block.3.0.attention.output.dense.bias', 'encoder.block.2.1.attention.self.sr.weight', 'decode_head.batch_norm.running_mean', 'encoder.block.3.0.attention.self.key.weight', 'encoder.block.0.1.attention.self.layer_norm.bias', 'encoder.block.0.0.attention.self.sr.bias', 'encoder.block.3.1.layer_norm_1.bias', 'encoder.block.0.1.attention.self.value.bias', 'encoder.block.0.1.attention.self.sr.bias', 'encoder.block.0.1.attention.self.key.weight', 'encoder.block.1.0.attention.self.value.bias', 'encoder.block.0.0.attention.self.layer_norm.weight', 'encoder.patch_embeddings.0.layer_norm.bias', 'encoder.patch_embeddings.3.proj.bias', 'encoder.block.1.1.attention.self.key.weight', 'encoder.block.1.1.attention.self.sr.bias', 'encoder.block.2.0.attention.self.key.weight', 'decode_head.linear_fuse.weight', 'encoder.block.0.1.mlp.dwconv.dwconv.bias', 'encoder.block.0.1.attention.self.layer_norm.weight', 'encoder.block.0.1.mlp.dwconv.dwconv.weight', 'encoder.block.2.0.layer_norm_1.bias', 'encoder.block.0.0.attention.self.query.bias', 'encoder.block.3.1.attention.self.key.bias', 'encoder.block.3.0.mlp.dwconv.dwconv.weight', 'encoder.block.0.0.attention.self.layer_norm.bias', 'encoder.block.0.0.layer_norm_1.weight', 'encoder.block.1.1.mlp.dense2.bias', 'encoder.block.3.1.layer_norm_2.weight', 'encoder.block.3.0.attention.output.dense.weight', 'encoder.block.1.0.layer_norm_1.weight', 'encoder.block.2.0.attention.self.layer_norm.weight', 'encoder.block.2.1.attention.self.layer_norm.weight', 'encoder.block.0.0.layer_norm_2.weight', 'encoder.patch_embeddings.1.layer_norm.weight', 'encoder.block.2.1.mlp.dwconv.dwconv.weight', 'encoder.block.3.1.attention.output.dense.bias', 'encoder.block.1.1.attention.self.query.weight', 'encoder.block.1.0.attention.self.key.bias', 'encoder.block.0.1.attention.self.value.weight', 'encoder.patch_embeddings.1.proj.weight', 'encoder.block.1.1.layer_norm_1.bias', 'encoder.block.1.1.attention.self.value.weight', 'encoder.block.3.0.mlp.dense2.weight', 'encoder.block.1.0.attention.self.value.weight', 'decode_head.classifier.bias', 'encoder.block.2.0.mlp.dwconv.dwconv.bias', 'encoder.block.1.1.attention.output.dense.weight', 'encoder.block.2.1.attention.self.query.weight', 'decode_head.batch_norm.num_batches_tracked', 'encoder.patch_embeddings.0.layer_norm.weight', 'decode_head.linear_c.3.proj.bias', 'encoder.block.0.1.attention.output.dense.bias', 'encoder.block.1.1.mlp.dwconv.dwconv.weight', 'encoder.block.2.1.attention.self.layer_norm.bias', 'encoder.block.2.1.mlp.dwconv.dwconv.bias', 'encoder.layer_norm.0.bias', 'encoder.patch_embeddings.3.layer_norm.weight', 'encoder.block.0.1.mlp.dense1.bias', 'encoder.block.1.1.mlp.dense2.weight', 'decode_head.batch_norm.weight', 'encoder.block.3.1.attention.self.query.bias', 'encoder.block.0.0.attention.output.dense.weight', 'encoder.block.1.1.layer_norm_1.weight', 'encoder.layer_norm.1.weight', 'encoder.block.3.0.mlp.dense2.bias', 'encoder.layer_norm.3.bias', 'encoder.block.0.0.mlp.dwconv.dwconv.weight', 'encoder.block.1.1.attention.output.dense.bias', 'encoder.block.2.1.layer_norm_1.weight', 'encoder.block.1.1.attention.self.layer_norm.weight', 'encoder.block.1.1.attention.self.key.bias', 'encoder.block.2.0.attention.self.value.bias', 'encoder.block.0.0.attention.self.key.bias', 'encoder.block.2.1.attention.self.key.bias', 'encoder.block.0.0.layer_norm_2.bias', 'encoder.block.1.1.layer_norm_2.weight', 'encoder.layer_norm.2.weight', 'decode_head.linear_c.0.proj.weight', 'encoder.block.1.0.attention.self.sr.weight', 'encoder.block.1.0.mlp.dense2.weight', 'encoder.block.2.0.mlp.dense1.weight', 'encoder.block.3.0.layer_norm_1.weight', 'encoder.patch_embeddings.1.proj.bias', 'decode_head.linear_c.3.proj.weight', 'encoder.block.1.1.mlp.dwconv.dwconv.bias', 'decode_head.linear_c.1.proj.bias', 'encoder.block.2.1.mlp.dense2.bias', 'encoder.patch_embeddings.3.layer_norm.bias', 'encoder.block.0.0.mlp.dense2.bias', 'encoder.block.3.1.layer_norm_2.bias', 'encoder.block.1.0.attention.output.dense.bias', 'encoder.patch_embeddings.0.proj.bias', 'encoder.block.2.1.attention.self.query.bias', 'encoder.block.2.1.mlp.dense1.bias', 'encoder.block.3.1.mlp.dwconv.dwconv.weight', 'encoder.block.0.1.layer_norm_2.bias', 'encoder.block.3.1.mlp.dense1.weight', 'encoder.block.1.0.attention.self.key.weight', 'encoder.block.1.1.mlp.dense1.weight', 'encoder.block.0.1.mlp.dense2.bias', 'encoder.block.0.1.attention.self.query.weight', 'encoder.block.2.0.attention.self.sr.bias', 'encoder.block.2.0.attention.output.dense.bias', 'encoder.block.2.1.mlp.dense1.weight', 'encoder.block.2.1.attention.self.value.bias', 'encoder.block.0.0.attention.self.key.weight', 'encoder.block.2.0.attention.self.key.bias', 'encoder.block.2.1.mlp.dense2.weight', 'decode_head.linear_c.0.proj.bias', 'encoder.block.3.1.attention.self.value.bias', 'encoder.block.0.1.attention.self.query.bias', 'encoder.block.3.0.attention.self.key.bias', 'encoder.patch_embeddings.2.proj.bias', 'encoder.layer_norm.1.bias', 'encoder.block.2.0.attention.self.query.weight', 'encoder.layer_norm.0.weight', 'encoder.block.0.0.attention.self.query.weight', 'encoder.block.0.1.attention.output.dense.weight', 'encoder.block.1.0.attention.self.query.bias', 'encoder.block.2.1.attention.self.key.weight', 'encoder.block.1.0.attention.self.layer_norm.weight', 'encoder.block.3.0.layer_norm_2.weight', 'encoder.block.0.0.attention.self.value.weight', 'encoder.block.0.0.mlp.dense1.bias', 'encoder.block.2.0.layer_norm_2.weight', 'encoder.patch_embeddings.3.proj.weight', 'encoder.block.1.0.mlp.dense2.bias', 'encoder.block.1.1.mlp.dense1.bias', 'encoder.block.3.1.attention.output.dense.weight', 'encoder.block.0.1.layer_norm_1.bias', 'encoder.block.0.0.attention.output.dense.bias', 'encoder.block.3.1.layer_norm_1.weight', 'decode_head.batch_norm.bias', 'encoder.block.2.0.mlp.dwconv.dwconv.weight', 'encoder.block.3.0.mlp.dwconv.dwconv.bias', 'encoder.block.2.1.layer_norm_1.bias', 'encoder.patch_embeddings.2.layer_norm.weight', 'encoder.block.2.0.mlp.dense1.bias', 'encoder.block.3.1.attention.self.key.weight', 'encoder.block.0.0.mlp.dense1.weight', 'encoder.block.1.1.attention.self.query.bias', 'encoder.block.3.0.attention.self.query.weight', 'decode_head.linear_c.2.proj.bias', 'encoder.layer_norm.2.bias', 'encoder.block.0.0.attention.self.sr.weight', 'encoder.block.3.0.layer_norm_2.bias', 'encoder.block.1.0.layer_norm_1.bias', 'encoder.layer_norm.3.weight', 'encoder.block.0.1.mlp.dense2.weight', 'encoder.block.2.0.mlp.dense2.weight', 'encoder.block.0.1.attention.self.key.bias', 'encoder.block.1.1.attention.self.value.bias', 'encoder.block.2.1.attention.self.value.weight', 'encoder.block.1.1.layer_norm_2.bias', 'encoder.block.3.0.attention.self.value.bias', 'encoder.patch_embeddings.2.layer_norm.bias', 'encoder.block.2.0.attention.output.dense.weight', 'encoder.block.3.1.mlp.dense2.bias', 'encoder.block.3.0.attention.self.value.weight', 'encoder.patch_embeddings.0.proj.weight', 'encoder.block.2.0.mlp.dense2.bias']\n",
+      "You should probably TRAIN this model on a down-stream task to be able to use it for predictions and inference.\n",
+      "/home/chainyo/code/segformer-sidewalk/. is already a clone of https://huggingface.co/ChainYo/segformer-sidewalk. Make sure you pull the latest changes with `repo.git_pull()`.\n"
      ]
     },
     {
-     "ename": "TypeError",
-     "evalue": "__init__() got an unexpected keyword argument 'num_labels'",
-     "output_type": "error",
-     "traceback": [
-      "\u001b[0;31m---------------------------------------------------------------------------\u001b[0m",
-      "\u001b[0;31mTypeError\u001b[0m                                 Traceback (most recent call last)",
-      "\u001b[1;32m/home/chainyo/code/segformer-sidewalk/finetuning.ipynb Cell 23'\u001b[0m in \u001b[0;36m<cell line: 11>\u001b[0;34m()\u001b[0m\n\u001b[1;32m      <a href='vscode-notebook-cell:/home/chainyo/code/segformer-sidewalk/finetuning.ipynb#ch0000016?line=7'>8</a>\u001b[0m config \u001b[39m=\u001b[39m AutoConfig\u001b[39m.\u001b[39mfrom_pretrained(\u001b[39mf\u001b[39m\u001b[39m\"\u001b[39m\u001b[39mnvidia/mit-b0\u001b[39m\u001b[39m\"\u001b[39m)\n\u001b[1;32m      <a href='vscode-notebook-cell:/home/chainyo/code/segformer-sidewalk/finetuning.ipynb#ch0000016?line=8'>9</a>\u001b[0m config\u001b[39m.\u001b[39mpush_to_hub(\u001b[39m\"\u001b[39m\u001b[39m.\u001b[39m\u001b[39m\"\u001b[39m, repo_url\u001b[39m=\u001b[39m\u001b[39m\"\u001b[39m\u001b[39mhttps://huggingface.co/ChainYo/segformer-sidewalk\u001b[39m\u001b[39m\"\u001b[39m)\n\u001b[0;32m---> <a href='vscode-notebook-cell:/home/chainyo/code/segformer-sidewalk/finetuning.ipynb#ch0000016?line=10'>11</a>\u001b[0m model \u001b[39m=\u001b[39m SegformerForSemanticSegmentation\u001b[39m.\u001b[39;49mfrom_pretrained(\n\u001b[1;32m     <a href='vscode-notebook-cell:/home/chainyo/code/segformer-sidewalk/finetuning.ipynb#ch0000016?line=11'>12</a>\u001b[0m     \u001b[39m\"\u001b[39;49m\u001b[39m/home/chainyo/code/segformer-sidewalk/checkpoints/epoch=44-step=1125.ckpt\u001b[39;49m\u001b[39m\"\u001b[39;49m, \n\u001b[1;32m     <a href='vscode-notebook-cell:/home/chainyo/code/segformer-sidewalk/finetuning.ipynb#ch0000016?line=12'>13</a>\u001b[0m     num_labels\u001b[39m=\u001b[39;49mnum_labels, \n\u001b[1;32m     <a href='vscode-notebook-cell:/home/chainyo/code/segformer-sidewalk/finetuning.ipynb#ch0000016?line=13'>14</a>\u001b[0m     id2label\u001b[39m=\u001b[39;49mid2label, \n\u001b[1;32m     <a href='vscode-notebook-cell:/home/chainyo/code/segformer-sidewalk/finetuning.ipynb#ch0000016?line=14'>15</a>\u001b[0m     label2id\u001b[39m=\u001b[39;49mid2label,\n\u001b[1;32m     <a href='vscode-notebook-cell:/home/chainyo/code/segformer-sidewalk/finetuning.ipynb#ch0000016?line=15'>16</a>\u001b[0m     config\u001b[39m=\u001b[39;49mconfig,\n\u001b[1;32m     <a href='vscode-notebook-cell:/home/chainyo/code/segformer-sidewalk/finetuning.ipynb#ch0000016?line=16'>17</a>\u001b[0m )\n\u001b[1;32m     <a href='vscode-notebook-cell:/home/chainyo/code/segformer-sidewalk/finetuning.ipynb#ch0000016?line=17'>18</a>\u001b[0m model\u001b[39m.\u001b[39mpush_to_hub(\u001b[39m\"\u001b[39m\u001b[39m.\u001b[39m\u001b[39m\"\u001b[39m, repo_url\u001b[39m=\u001b[39m\u001b[39m\"\u001b[39m\u001b[39mhttps://huggingface.co/ChainYo/segformer-sidewalk\u001b[39m\u001b[39m\"\u001b[39m)\n",
-      "File \u001b[0;32m~/miniconda3/envs/segformer/lib/python3.8/site-packages/transformers/modeling_utils.py:2024\u001b[0m, in \u001b[0;36mPreTrainedModel.from_pretrained\u001b[0;34m(cls, pretrained_model_name_or_path, *model_args, **kwargs)\u001b[0m\n\u001b[1;32m   <a href='file:///home/chainyo/miniconda3/envs/segformer/lib/python3.8/site-packages/transformers/modeling_utils.py?line=2021'>2022</a>\u001b[0m \u001b[39melse\u001b[39;00m:\n\u001b[1;32m   <a href='file:///home/chainyo/miniconda3/envs/segformer/lib/python3.8/site-packages/transformers/modeling_utils.py?line=2022'>2023</a>\u001b[0m     \u001b[39mwith\u001b[39;00m no_init_weights(_enable\u001b[39m=\u001b[39m_fast_init):\n\u001b[0;32m-> <a href='file:///home/chainyo/miniconda3/envs/segformer/lib/python3.8/site-packages/transformers/modeling_utils.py?line=2023'>2024</a>\u001b[0m         model \u001b[39m=\u001b[39m \u001b[39mcls\u001b[39;49m(config, \u001b[39m*\u001b[39;49mmodel_args, \u001b[39m*\u001b[39;49m\u001b[39m*\u001b[39;49mmodel_kwargs)\n\u001b[1;32m   <a href='file:///home/chainyo/miniconda3/envs/segformer/lib/python3.8/site-packages/transformers/modeling_utils.py?line=2025'>2026</a>\u001b[0m \u001b[39mif\u001b[39;00m from_tf:\n\u001b[1;32m   <a href='file:///home/chainyo/miniconda3/envs/segformer/lib/python3.8/site-packages/transformers/modeling_utils.py?line=2026'>2027</a>\u001b[0m     \u001b[39mif\u001b[39;00m resolved_archive_file\u001b[39m.\u001b[39mendswith(\u001b[39m\"\u001b[39m\u001b[39m.index\u001b[39m\u001b[39m\"\u001b[39m):\n\u001b[1;32m   <a href='file:///home/chainyo/miniconda3/envs/segformer/lib/python3.8/site-packages/transformers/modeling_utils.py?line=2027'>2028</a>\u001b[0m         \u001b[39m# Load from a TensorFlow 1.X checkpoint - provided by original authors\u001b[39;00m\n",
-      "\u001b[0;31mTypeError\u001b[0m: __init__() got an unexpected keyword argument 'num_labels'"
+     "data": {
+      "application/vnd.jupyter.widget-view+json": {
+       "model_id": "99e53578044149a1ac2cb3a53bc93015",
+       "version_major": 2,
+       "version_minor": 0
+      },
+      "text/plain": [
+       "Upload file pytorch_model.bin:   0%|          | 32.0k/14.3M [00:00<?, ?B/s]"
+      ]
+     },
+     "metadata": {},
+     "output_type": "display_data"
+    },
+    {
+     "name": "stderr",
+     "output_type": "stream",
+     "text": [
+      "remote: Enforcing permissions...        \n",
+      "remote: Allowed refs: all        \n",
+      "To https://huggingface.co/ChainYo/segformer-sidewalk\n",
+      "   fcb528d..8370452  main -> main\n",
+      "\n"
      ]
+    },
+    {
+     "data": {
+      "text/plain": [
+       "'https://huggingface.co/ChainYo/segformer-sidewalk/commit/83704520224204a5932e7a7e174051010a99fe83'"
+      ]
+     },
+     "execution_count": 33,
+     "metadata": {},
+     "output_type": "execute_result"
     }
    ],
    "source": [