Updating for faster training

02-autotrain.ipynb

@@ -99,7 +99,7 @@
     "    \"base_model\": \"tiiuae/Falcon3-7B-Instruct\",\n",
     "    \"project_name\": \"\",\n",
     "    \"log\": \"tensorboard\",\n",
-    "    \"backend\": \"spaces-l4x1\",\n",
+    "    \"backend\": \"spaces-a100-large\",\n",
     "    \"data\": {\n",
     "        \"path\": \"derek-thomas/labeled-multiple-choice-explained-falcon-tokenized\",\n",
     "        \"train_split\": \"train\",\n",
@@ -113,8 +113,8 @@
     "        \"block_size\": 512,\n",
     "        \"model_max_length\": 1500,\n",
     "        \"epochs\": 4,\n",
-    "        \"batch_size\": 1,\n",
-    "        \"lr\": 3e-7,\n",
+    "        \"batch_size\": 8,\n",
+    "        \"lr\": 1e-6,\n",
     "        \"peft\": True,\n",
     "        \"quantization\": \"int4\",\n",
     "        \"target_modules\": \"all-linear\",\n",
@@ -335,11 +335,10 @@
    ]
   },
   {
-   "cell_type": "code",
-   "execution_count": null,
-   "id": "f86ed8ad-4e38-454a-a2c1-b1f075399c37",
    "metadata": {},
+   "cell_type": "code",
    "outputs": [],
+   "execution_count": null,
    "source": [
     "for space in autotrain_spaces:\n",
     "    confirm = input(f\"Are you sure you want to delete the space '{space}'? (y/n): \")\n",
@@ -348,25 +347,25 @@
     "        print(f\"Deleted {space}\")\n",
     "    else:\n",
     "        print(f\"Skipped {space}\")\n"
-   ]
+   ],
+   "id": "de150f086f1c72fa"
   },
   {
-   "cell_type": "markdown",
-   "id": "2182f8fe-8504-4cb9-a0a6-4b143541158d",
    "metadata": {},
+   "cell_type": "markdown",
    "source": [
     "<span style=\"color:red; font-size:20px; font-weight:bold;\">\n",
     "ONLY RUN THIS IF YOU NEED TO RESTART FROM SCRATCH\n",
     "THIS WILL DELETE YOUR MODELS\n",
-    "</span>"
-   ]
+    "</span>\n"
+   ],
+   "id": "e6aa4788fe8b0297"
   },
   {
-   "cell_type": "code",
-   "execution_count": null,
-   "id": "12939405-a731-4a7c-ab4a-e1a4f1850bb6",
    "metadata": {},
+   "cell_type": "code",
    "outputs": [],
+   "execution_count": null,
    "source": [
     "# for model in autotrain_models:\n",
     "#     confirm = input(f\"Are you sure you want to delete the model '{model}'? (y/n): \")\n",
@@ -375,6 +374,98 @@
     "#         print(f\"Deleted {model}\")\n",
     "#     else:\n",
     "#         print(f\"Skipped {model}\")\n"
+   ],
+   "id": "c887e61558785e69"
+  },
+  {
+   "cell_type": "code",
+   "execution_count": 12,
+   "id": "a8a8d4c2-0fb5-438c-9fbf-436144d05543",
+   "metadata": {},
+   "outputs": [
+    {
+     "name": "stdin",
+     "output_type": "stream",
+     "text": [
+      "Are you sure you want to move the model from 'derek-thomas/falcon-v03-poe-RFA-gpt3-5' to derek-thomas/falcon-v03-poe-RFA-gpt3-5_backup? (y/n):  y\n"
+     ]
+    },
+    {
+     "name": "stdout",
+     "output_type": "stream",
+     "text": [
+      "Moved derek-thomas/falcon-v03-poe-RFA-gpt3-5 to derek-thomas/falcon-v03-poe-RFA-gpt3-5_backup\n"
+     ]
+    },
+    {
+     "name": "stdin",
+     "output_type": "stream",
+     "text": [
+      "Are you sure you want to move the model from 'derek-thomas/falcon-v03-poe-RFA-falcon' to derek-thomas/falcon-v03-poe-RFA-falcon_backup? (y/n):  y\n"
+     ]
+    },
+    {
+     "name": "stdout",
+     "output_type": "stream",
+     "text": [
+      "Moved derek-thomas/falcon-v03-poe-RFA-falcon to derek-thomas/falcon-v03-poe-RFA-falcon_backup\n"
+     ]
+    },
+    {
+     "name": "stdin",
+     "output_type": "stream",
+     "text": [
+      "Are you sure you want to move the model from 'derek-thomas/falcon-v03-poe-FAR-gpt3-5' to derek-thomas/falcon-v03-poe-FAR-gpt3-5_backup? (y/n):  y\n"
+     ]
+    },
+    {
+     "name": "stdout",
+     "output_type": "stream",
+     "text": [
+      "Moved derek-thomas/falcon-v03-poe-FAR-gpt3-5 to derek-thomas/falcon-v03-poe-FAR-gpt3-5_backup\n"
+     ]
+    },
+    {
+     "name": "stdin",
+     "output_type": "stream",
+     "text": [
+      "Are you sure you want to move the model from 'derek-thomas/falcon-v03-poe-FAR-falcon' to derek-thomas/falcon-v03-poe-FAR-falcon_backup? (y/n):  y\n"
+     ]
+    },
+    {
+     "name": "stdout",
+     "output_type": "stream",
+     "text": [
+      "Moved derek-thomas/falcon-v03-poe-FAR-falcon to derek-thomas/falcon-v03-poe-FAR-falcon_backup\n"
+     ]
+    },
+    {
+     "name": "stdin",
+     "output_type": "stream",
+     "text": [
+      "Are you sure you want to move the model from 'derek-thomas/falcon-v03-poe-FA' to derek-thomas/falcon-v03-poe-FA_backup? (y/n):  y\n"
+     ]
+    },
+    {
+     "name": "stdout",
+     "output_type": "stream",
+     "text": [
+      "Moved derek-thomas/falcon-v03-poe-FA to derek-thomas/falcon-v03-poe-FA_backup\n"
+     ]
+    }
+   ],
+   "source": [
+    "from huggingface_hub import move_repo\n",
+    "\n",
+    "for model in autotrain_models:\n",
+    "    user, model_tag = model.split('/')\n",
+    "    to = f\"{user}/{model_tag}_backup\"\n",
+    "    confirm = input(f\"Are you sure you want to move the model from '{model}' to {to}? (y/n): \")\n",
+    "    if confirm.lower() == 'y':\n",
+    "        move_repo(from_id=model, to_id=to)\n",
+    "        print(f\"Moved {model} to {to}\")\n",
+    "    else:\n",
+    "        print(f\"Skipped {model}\")"
    ]
   },
   {