Training in progress, epoch 2

pytorch_model.bin

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:a8d0b025faa8c33ff9e377a0ecb3a8066849218e66d79dabdb9c9e1289d3a014
+oid sha256:7d9fe1fd2b8ff6706c8464c30b35de48c417dea60cb57eee31281e0106c6005a
 size 274752173

train_factual_consistency.ipynb

@@ -9,7 +9,7 @@
     {
      "data": {
       "application/vnd.jupyter.widget-view+json": {
-       "model_id": "0d95da7d7c36430cb42eb766dd2e4088",
+       "model_id": "eab00695e2b240ffb58ab998c85c0e7d",
        "version_major": 2,
        "version_minor": 0
       },
@@ -81,7 +81,7 @@
     {
      "data": {
       "application/vnd.jupyter.widget-view+json": {
-       "model_id": "9eb3a7a65ef5432c8837e53a8f2653e4",
+       "model_id": "f20c88d0f96c4c06a9a0ddf835e544e3",
        "version_major": 2,
        "version_minor": 0
       },
@@ -103,7 +103,7 @@
     {
      "data": {
       "application/vnd.jupyter.widget-view+json": {
-       "model_id": "cda91dc75e7e4a2fbdffa8216279d83b",
+       "model_id": "97591288ca9d40fd91e7737b41828f63",
        "version_major": 2,
        "version_minor": 0
       },
@@ -125,7 +125,7 @@
   },
   {
    "cell_type": "code",
-   "execution_count": 4,
+   "execution_count": null,
    "id": "6bc83d4c-378c-4313-b641-8ead0c02f715",
    "metadata": {},
    "outputs": [
@@ -144,8 +144,8 @@
        "\n",
        "    <div>\n",
        "      \n",
-       "      <progress value='1245' max='30600' style='width:300px; height:20px; vertical-align: middle;'></progress>\n",
-       "      [ 1245/30600 02:31 < 59:40, 8.20 it/s, Epoch 4.07/100]\n",
+       "      <progress value='576' max='30600' style='width:300px; height:20px; vertical-align: middle;'></progress>\n",
+       "      [  576/30600 01:07 < 58:30, 8.55 it/s, Epoch 1.88/100]\n",
        "    </div>\n",
        "    <table border=\"1\" class=\"dataframe\">\n",
        "  <thead>\n",
@@ -159,22 +159,7 @@
        "    <tr>\n",
        "      <td>1</td>\n",
        "      <td>No log</td>\n",
-       "      <td>-3.357920</td>\n",
-       "    </tr>\n",
-       "    <tr>\n",
-       "      <td>2</td>\n",
-       "      <td>-2.673600</td>\n",
-       "      <td>-7.069220</td>\n",
-       "    </tr>\n",
-       "    <tr>\n",
-       "      <td>3</td>\n",
-       "      <td>-2.673600</td>\n",
-       "      <td>-11.083688</td>\n",
-       "    </tr>\n",
-       "    <tr>\n",
-       "      <td>4</td>\n",
-       "      <td>-8.789900</td>\n",
-       "      <td>-15.529228</td>\n",
+       "      <td>0.085583</td>\n",
        "    </tr>\n",
        "  </tbody>\n",
        "</table><p>"
@@ -185,21 +170,6 @@
      },
      "metadata": {},
      "output_type": "display_data"
-    },
-    {
-     "ename": "KeyboardInterrupt",
-     "evalue": "",
-     "output_type": "error",
-     "traceback": [
-      "\u001b[0;31m---------------------------------------------------------------------------\u001b[0m",
-      "\u001b[0;31mKeyboardInterrupt\u001b[0m                         Traceback (most recent call last)",
-      "Cell \u001b[0;32mIn[4], line 27\u001b[0m\n\u001b[1;32m     17\u001b[0m data_collator \u001b[38;5;241m=\u001b[39m DataCollatorWithPadding(tokenizer\u001b[38;5;241m=\u001b[39mtokenizer)\n\u001b[1;32m     18\u001b[0m trainer \u001b[38;5;241m=\u001b[39m Trainer(\n\u001b[1;32m     19\u001b[0m     model\u001b[38;5;241m=\u001b[39mmodel,\n\u001b[1;32m     20\u001b[0m     args\u001b[38;5;241m=\u001b[39mtraining_args,\n\u001b[0;32m   (...)\u001b[0m\n\u001b[1;32m     24\u001b[0m     data_collator\u001b[38;5;241m=\u001b[39mdata_collator,\n\u001b[1;32m     25\u001b[0m )\n\u001b[0;32m---> 27\u001b[0m \u001b[43mtrainer\u001b[49m\u001b[38;5;241;43m.\u001b[39;49m\u001b[43mtrain\u001b[49m\u001b[43m(\u001b[49m\u001b[43m)\u001b[49m\n\u001b[1;32m     28\u001b[0m trainer\u001b[38;5;241m.\u001b[39mpush_to_hub(\u001b[38;5;124m'\u001b[39m\u001b[38;5;124mfactual-consistency-regression-ja\u001b[39m\u001b[38;5;124m'\u001b[39m)\n",
-      "File \u001b[0;32m/opt/conda/lib/python3.10/site-packages/transformers/trainer.py:1582\u001b[0m, in \u001b[0;36mTrainer.train\u001b[0;34m(self, resume_from_checkpoint, trial, ignore_keys_for_eval, **kwargs)\u001b[0m\n\u001b[1;32m   1579\u001b[0m \u001b[38;5;28;01mtry\u001b[39;00m:\n\u001b[1;32m   1580\u001b[0m     \u001b[38;5;66;03m# Disable progress bars when uploading models during checkpoints to avoid polluting stdout\u001b[39;00m\n\u001b[1;32m   1581\u001b[0m     hf_hub_utils\u001b[38;5;241m.\u001b[39mdisable_progress_bars()\n\u001b[0;32m-> 1582\u001b[0m     \u001b[38;5;28;01mreturn\u001b[39;00m \u001b[43minner_training_loop\u001b[49m\u001b[43m(\u001b[49m\n\u001b[1;32m   1583\u001b[0m \u001b[43m        \u001b[49m\u001b[43margs\u001b[49m\u001b[38;5;241;43m=\u001b[39;49m\u001b[43margs\u001b[49m\u001b[43m,\u001b[49m\n\u001b[1;32m   1584\u001b[0m \u001b[43m        \u001b[49m\u001b[43mresume_from_checkpoint\u001b[49m\u001b[38;5;241;43m=\u001b[39;49m\u001b[43mresume_from_checkpoint\u001b[49m\u001b[43m,\u001b[49m\n\u001b[1;32m   1585\u001b[0m \u001b[43m        \u001b[49m\u001b[43mtrial\u001b[49m\u001b[38;5;241;43m=\u001b[39;49m\u001b[43mtrial\u001b[49m\u001b[43m,\u001b[49m\n\u001b[1;32m   1586\u001b[0m \u001b[43m        \u001b[49m\u001b[43mignore_keys_for_eval\u001b[49m\u001b[38;5;241;43m=\u001b[39;49m\u001b[43mignore_keys_for_eval\u001b[49m\u001b[43m,\u001b[49m\n\u001b[1;32m   1587\u001b[0m \u001b[43m    \u001b[49m\u001b[43m)\u001b[49m\n\u001b[1;32m   1588\u001b[0m \u001b[38;5;28;01mfinally\u001b[39;00m:\n\u001b[1;32m   1589\u001b[0m     hf_hub_utils\u001b[38;5;241m.\u001b[39menable_progress_bars()\n",
-      "File \u001b[0;32m/opt/conda/lib/python3.10/site-packages/transformers/trainer.py:1950\u001b[0m, in \u001b[0;36mTrainer._inner_training_loop\u001b[0;34m(self, batch_size, args, resume_from_checkpoint, trial, ignore_keys_for_eval)\u001b[0m\n\u001b[1;32m   1945\u001b[0m         nn\u001b[38;5;241m.\u001b[39mutils\u001b[38;5;241m.\u001b[39mclip_grad_norm_(\n\u001b[1;32m   1946\u001b[0m             amp\u001b[38;5;241m.\u001b[39mmaster_params(\u001b[38;5;28mself\u001b[39m\u001b[38;5;241m.\u001b[39moptimizer),\n\u001b[1;32m   1947\u001b[0m             args\u001b[38;5;241m.\u001b[39mmax_grad_norm,\n\u001b[1;32m   1948\u001b[0m         )\n\u001b[1;32m   1949\u001b[0m     \u001b[38;5;28;01melse\u001b[39;00m:\n\u001b[0;32m-> 1950\u001b[0m         \u001b[38;5;28;43mself\u001b[39;49m\u001b[38;5;241;43m.\u001b[39;49m\u001b[43maccelerator\u001b[49m\u001b[38;5;241;43m.\u001b[39;49m\u001b[43mclip_grad_norm_\u001b[49m\u001b[43m(\u001b[49m\n\u001b[1;32m   1951\u001b[0m \u001b[43m            \u001b[49m\u001b[43mmodel\u001b[49m\u001b[38;5;241;43m.\u001b[39;49m\u001b[43mparameters\u001b[49m\u001b[43m(\u001b[49m\u001b[43m)\u001b[49m\u001b[43m,\u001b[49m\n\u001b[1;32m   1952\u001b[0m \u001b[43m            \u001b[49m\u001b[43margs\u001b[49m\u001b[38;5;241;43m.\u001b[39;49m\u001b[43mmax_grad_norm\u001b[49m\u001b[43m,\u001b[49m\n\u001b[1;32m   1953\u001b[0m \u001b[43m        \u001b[49m\u001b[43m)\u001b[49m\n\u001b[1;32m   1955\u001b[0m \u001b[38;5;66;03m# Optimizer step\u001b[39;00m\n\u001b[1;32m   1956\u001b[0m optimizer_was_run \u001b[38;5;241m=\u001b[39m \u001b[38;5;28;01mTrue\u001b[39;00m\n",
-      "File \u001b[0;32m/opt/conda/lib/python3.10/site-packages/accelerate/accelerator.py:2121\u001b[0m, in \u001b[0;36mAccelerator.clip_grad_norm_\u001b[0;34m(self, parameters, max_norm, norm_type)\u001b[0m\n\u001b[1;32m   2119\u001b[0m     \u001b[38;5;28;01mreturn\u001b[39;00m \u001b[38;5;28;01mNone\u001b[39;00m\n\u001b[1;32m   2120\u001b[0m \u001b[38;5;28mself\u001b[39m\u001b[38;5;241m.\u001b[39munscale_gradients()\n\u001b[0;32m-> 2121\u001b[0m \u001b[38;5;28;01mreturn\u001b[39;00m \u001b[43mtorch\u001b[49m\u001b[38;5;241;43m.\u001b[39;49m\u001b[43mnn\u001b[49m\u001b[38;5;241;43m.\u001b[39;49m\u001b[43mutils\u001b[49m\u001b[38;5;241;43m.\u001b[39;49m\u001b[43mclip_grad_norm_\u001b[49m\u001b[43m(\u001b[49m\u001b[43mparameters\u001b[49m\u001b[43m,\u001b[49m\u001b[43m \u001b[49m\u001b[43mmax_norm\u001b[49m\u001b[43m,\u001b[49m\u001b[43m \u001b[49m\u001b[43mnorm_type\u001b[49m\u001b[38;5;241;43m=\u001b[39;49m\u001b[43mnorm_type\u001b[49m\u001b[43m)\u001b[49m\n",
-      "File \u001b[0;32m/opt/conda/lib/python3.10/site-packages/torch_xla/_patched_functions.py:49\u001b[0m, in \u001b[0;36mclip_grad_norm_\u001b[0;34m(parameters, max_norm, norm_type, error_if_nonfinite, foreach)\u001b[0m\n\u001b[1;32m     44\u001b[0m \u001b[38;5;28;01mif\u001b[39;00m error_if_nonfinite \u001b[38;5;129;01mand\u001b[39;00m (total_norm\u001b[38;5;241m.\u001b[39misnan() \u001b[38;5;129;01mor\u001b[39;00m total_norm\u001b[38;5;241m.\u001b[39misinf()):\n\u001b[1;32m     45\u001b[0m   \u001b[38;5;28;01mraise\u001b[39;00m \u001b[38;5;167;01mRuntimeError\u001b[39;00m(\n\u001b[1;32m     46\u001b[0m       \u001b[38;5;124mf\u001b[39m\u001b[38;5;124m'\u001b[39m\u001b[38;5;124mThe norm of order \u001b[39m\u001b[38;5;132;01m{\u001b[39;00mnorm_type\u001b[38;5;132;01m}\u001b[39;00m\u001b[38;5;124m for a gradient from `parameters` \u001b[39m\u001b[38;5;124m'\u001b[39m\n\u001b[1;32m     47\u001b[0m       \u001b[38;5;124m'\u001b[39m\u001b[38;5;124mis non-finite, so it cannot be clipped. This error can be \u001b[39m\u001b[38;5;124m'\u001b[39m\n\u001b[1;32m     48\u001b[0m       \u001b[38;5;124m'\u001b[39m\u001b[38;5;124mdisabled with `error_if_nonfinite=False`\u001b[39m\u001b[38;5;124m'\u001b[39m)\n\u001b[0;32m---> 49\u001b[0m clip_coef \u001b[38;5;241m=\u001b[39m \u001b[43mtorch\u001b[49m\u001b[38;5;241;43m.\u001b[39;49m\u001b[43mtensor\u001b[49m\u001b[43m(\u001b[49m\u001b[43mmax_norm\u001b[49m\u001b[43m,\u001b[49m\u001b[43m \u001b[49m\u001b[43mdevice\u001b[49m\u001b[38;5;241;43m=\u001b[39;49m\u001b[43mdevice\u001b[49m\u001b[43m)\u001b[49m \u001b[38;5;241m/\u001b[39m (total_norm \u001b[38;5;241m+\u001b[39m \u001b[38;5;241m1e-6\u001b[39m)\n\u001b[1;32m     50\u001b[0m clip_value \u001b[38;5;241m=\u001b[39m torch\u001b[38;5;241m.\u001b[39mwhere(clip_coef \u001b[38;5;241m<\u001b[39m \u001b[38;5;241m1\u001b[39m, clip_coef,\n\u001b[1;32m     51\u001b[0m                          torch\u001b[38;5;241m.\u001b[39mtensor(\u001b[38;5;241m1.\u001b[39m, device\u001b[38;5;241m=\u001b[39mdevice))\n\u001b[1;32m     52\u001b[0m \u001b[38;5;28;01mfor\u001b[39;00m p \u001b[38;5;129;01min\u001b[39;00m parameters:\n",
-      "\u001b[0;31mKeyboardInterrupt\u001b[0m: "
-     ]
     }
    ],
    "source": [