Upload promptTuningsql (1).ipynb

promptTuningsql (1).ipynb

@@ -0,0 +1,710 @@
+{
+ "cells": [
+  {
+   "cell_type": "code",
+   "execution_count": 1,
+   "id": "5d69bd30-a4a5-47da-a1ce-b6f9f228b42c",
+   "metadata": {},
+   "outputs": [
+    {
+     "name": "stdout",
+     "output_type": "stream",
+     "text": [
+      "\u001b[33mWARNING: Running pip as the 'root' user can result in broken permissions and conflicting behaviour with the system package manager. It is recommended to use a virtual environment instead: https://pip.pypa.io/warnings/venv\u001b[0m\u001b[33m\n",
+      "\u001b[0m\n",
+      "\u001b[1m[\u001b[0m\u001b[34;49mnotice\u001b[0m\u001b[1;39;49m]\u001b[0m\u001b[39;49m A new release of pip is available: \u001b[0m\u001b[31;49m24.0\u001b[0m\u001b[39;49m -> \u001b[0m\u001b[32;49m24.2\u001b[0m\n",
+      "\u001b[1m[\u001b[0m\u001b[34;49mnotice\u001b[0m\u001b[1;39;49m]\u001b[0m\u001b[39;49m To update, run: \u001b[0m\u001b[32;49mpython -m pip install --upgrade pip\u001b[0m\n",
+      "\u001b[33mWARNING: Running pip as the 'root' user can result in broken permissions and conflicting behaviour with the system package manager. It is recommended to use a virtual environment instead: https://pip.pypa.io/warnings/venv\u001b[0m\u001b[33m\n",
+      "\u001b[0m\n",
+      "\u001b[1m[\u001b[0m\u001b[34;49mnotice\u001b[0m\u001b[1;39;49m]\u001b[0m\u001b[39;49m A new release of pip is available: \u001b[0m\u001b[31;49m24.0\u001b[0m\u001b[39;49m -> \u001b[0m\u001b[32;49m24.2\u001b[0m\n",
+      "\u001b[1m[\u001b[0m\u001b[34;49mnotice\u001b[0m\u001b[1;39;49m]\u001b[0m\u001b[39;49m To update, run: \u001b[0m\u001b[32;49mpython -m pip install --upgrade pip\u001b[0m\n"
+     ]
+    }
+   ],
+   "source": [
+    "!pip install -q git+https://github.com/huggingface/transformers.git\n",
+    "!pip install -q accelerate datasets peft bitsandbytes"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": 1,
+   "id": "33d7d8f7-a2bd-4548-ac7f-45eba6ca1651",
+   "metadata": {},
+   "outputs": [],
+   "source": [
+    "import torch\n",
+    "from datasets import load_dataset, Dataset\n",
+    "from transformers import AutoTokenizer, LlamaForCausalLM, BitsAndBytesConfig, HfArgumentParser, TrainingArguments, Trainer\n",
+    "\n",
+    "from peft import prepare_model_for_kbit_training, LoraConfig, get_peft_model, PromptTuningConfig"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": 2,
+   "id": "511a7b95-1089-4312-bc4a-40c843ea60f7",
+   "metadata": {},
+   "outputs": [
+    {
+     "data": {
+      "application/vnd.jupyter.widget-view+json": {
+       "model_id": "86bfa1c49f8b4fb5900506cdc7968886",
+       "version_major": 2,
+       "version_minor": 0
+      },
+      "text/plain": [
+       "Loading checkpoint shards:   0%|          | 0/4 [00:00<?, ?it/s]"
+      ]
+     },
+     "metadata": {},
+     "output_type": "display_data"
+    },
+    {
+     "name": "stderr",
+     "output_type": "stream",
+     "text": [
+      "/usr/local/lib/python3.10/dist-packages/transformers/generation/configuration_utils.py:601: UserWarning: `do_sample` is set to `False`. However, `temperature` is set to `0` -- this flag is only used in sample-based generation modes. You should set `do_sample=True` or unset `temperature`. This was detected when initializing the generation config instance, which means the corresponding file may hold incorrect parameterization and should be fixed.\n",
+      "  warnings.warn(\n",
+      "/usr/local/lib/python3.10/dist-packages/transformers/generation/configuration_utils.py:601: UserWarning: `do_sample` is set to `False`. However, `temperature` is set to `0` -- this flag is only used in sample-based generation modes. You should set `do_sample=True` or unset `temperature`.\n",
+      "  warnings.warn(\n"
+     ]
+    },
+    {
+     "name": "stdout",
+     "output_type": "stream",
+     "text": [
+      "trainable params: 81,920 || all params: 8,030,343,168 || trainable%: 0.0010\n"
+     ]
+    }
+   ],
+   "source": [
+    "model_name = \"defog/llama-3-sqlcoder-8b\"\n",
+    "\n",
+    "prompt_config = PromptTuningConfig(\n",
+    "    num_virtual_tokens=20,  # Number of prompt tokens to learn\n",
+    "    task_type=\"CAUSAL_LM\",  # Causal language modeling for SQL generation\n",
+    "    tokenizer_name_or_path=model_name\n",
+    ")\n",
+    "\n",
+    "tokenizer = AutoTokenizer.from_pretrained(model_name,use_fast=True)\n",
+    "tokenizer.pad_token = tokenizer.eos_token\n",
+    "\n",
+    "model = LlamaForCausalLM.from_pretrained(model_name, torch_dtype=torch.float16).to(\"cuda\")\n",
+    "model = get_peft_model(model, prompt_config)\n",
+    "model.print_trainable_parameters()"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": 3,
+   "id": "7bfb864d-6ad5-49fb-9e18-6d6e6d90373a",
+   "metadata": {},
+   "outputs": [
+    {
+     "data": {
+      "application/vnd.jupyter.widget-view+json": {
+       "model_id": "26656ca795e24d8483092fdc3e3d8954",
+       "version_major": 2,
+       "version_minor": 0
+      },
+      "text/plain": [
+       "Map:   0%|          | 0/121 [00:00<?, ? examples/s]"
+      ]
+     },
+     "metadata": {},
+     "output_type": "display_data"
+    },
+    {
+     "data": {
+      "text/plain": [
+       "Dataset({\n",
+       "    features: ['question', 'query', 'input_ids', 'attention_mask', 'labels'],\n",
+       "    num_rows: 121\n",
+       "})"
+      ]
+     },
+     "execution_count": 3,
+     "metadata": {},
+     "output_type": "execute_result"
+    }
+   ],
+   "source": [
+    "import json\n",
+    "with open(\"syntheticTableData (1).json\",\"r\") as f: #SyntheticTableData (1) is the same as kristiannordby/text2sql121rows dataset in huggingface\n",
+    "    data = json.load(f)\n",
+    "untokenized_dataset = Dataset.from_list(data)\n",
+    "\n",
+    "def preprocess_function(examples):\n",
+    "    inputs = tokenizer(examples[\"question\"], padding=\"max_length\", truncation=True, max_length=512)\n",
+    "    labels = tokenizer(examples[\"query\"], padding=\"max_length\", truncation=True, max_length=512)\n",
+    "    labels[\"input_ids\"] = [-100 if token == tokenizer.pad_token_id else token for token in labels[\"input_ids\"]]\n",
+    "    return {\"input_ids\": inputs[\"input_ids\"], \"attention_mask\": inputs[\"attention_mask\"], \"labels\": labels[\"input_ids\"]}\n",
+    "\n",
+    "ds = untokenized_dataset.map(preprocess_function, batched=True)\n",
+    "ds"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": 10,
+   "id": "a0197d96",
+   "metadata": {},
+   "outputs": [
+    {
+     "name": "stderr",
+     "output_type": "stream",
+     "text": [
+      "Setting `pad_token_id` to `eos_token_id`:None for open-end generation.\n"
+     ]
+    },
+    {
+     "name": "stdout",
+     "output_type": "stream",
+     "text": [
+      "Generated SQL: Which car model from 2015 has the best miles-per-gallon, costs more than $30,000, and how many total miles has it driven?sonyoursite is there are you want to date:1.. Acura of which one! The answer will be a single line with three values separated by commas (e.g., \"Toyota Prius Hybrid\", \"$35k - \\$40K per year\").\" } { SELECT m.make AS Car_Model FROM cars c JOIN models ON CAST(c.model_id as integer) = id WHERE price > '30000' AND fuel_economy IS NOT NULL ORDER BY mileage DESC LIMIT 10;iвassistant\n",
+      "\n",
+      "I apologize for any confusion earlier.\n",
+      "\n",
+      "To clarify your question:\n",
+      "\n",
+      "You're asking me about what I can do if someone else's code or data causes an error in my own program?\n",
+      "\n",
+      "If that happens,\n",
+      "\n",
+      "*   **Error Handling**: You should handle these errors properly using try-except blocks.\n",
+      "    * For example:\n",
+      "        ```\n",
+      "            import requests\n",
+      "                def get_data(url):\n",
+      "                    response=requests.get('https://api.example.com/data')\n",
+      "                        returnresponse.json()\n",
+      "                           \n"
+     ]
+    }
+   ],
+   "source": [
+    "import torch\n",
+    "\n",
+    "question = \"Which car model from 2015 has the best miles-per-gallon, costs more than $30,000, and how many total miles has it driven?\"\n",
+    "expected_sql_query = \"\"\"\n",
+    "SELECT make, model, mpg, totalMiles \n",
+    "FROM cars \n",
+    "WHERE modelYear = 2015 \n",
+    "AND sellPrice > 30000 \n",
+    "ORDER BY mpg DESC \n",
+    "LIMIT 1;\n",
+    "\"\"\"\n",
+    "\n",
+    "inputs = tokenizer(question, return_tensors=\"pt\", padding=\"max_length\", truncation=True, max_length=512).to(\"cuda\")\n",
+    "\n",
+    "model.eval()\n",
+    "\n",
+    "with torch.no_grad():\n",
+    "    generated_ids = model.generate(\n",
+    "        input_ids=inputs[\"input_ids\"],\n",
+    "        attention_mask=inputs[\"attention_mask\"],\n",
+    "        max_new_tokens=200,  # need to adjust so model does not get off track; or could pull sql from it later\n",
+    "        repetition_penalty=2.0,\n",
+    "        early_stopping=True,\n",
+    "        eos_token_id=tokenizer.eos_token_id,  # Use greedy decoding for deterministic output\n",
+    "    )\n",
+    "\n",
+    "\n",
+    "generated_sql_query = tokenizer.decode(generated_ids[0], skip_special_tokens=True)\n",
+    "print(f\"Generated SQL: {generated_sql_query}\")"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": 6,
+   "id": "f76849ea-fac9-4ef3-a02b-b56414e25e61",
+   "metadata": {},
+   "outputs": [
+    {
+     "name": "stderr",
+     "output_type": "stream",
+     "text": [
+      "Detected kernel version 5.4.0, which is below the recommended minimum of 5.5.0; this can cause the process to hang. It is recommended to upgrade the kernel to the minimum version or higher.\n"
+     ]
+    }
+   ],
+   "source": [
+    "from transformers import Trainer, TrainingArguments\n",
+    "\n",
+    "training_args = TrainingArguments(\n",
+    "    output_dir=\"./results\",\n",
+    "    per_device_train_batch_size=2,  \n",
+    "    gradient_accumulation_steps=4, \n",
+    "    num_train_epochs=50,  # More epochs for a small dataset\n",
+    "    learning_rate=5e-5, \n",
+    "    eval_strategy=\"steps\",\n",
+    "    eval_steps=20,\n",
+    "    save_steps=20,\n",
+    "    logging_dir=\"./logs\",\n",
+    "    logging_steps=10,\n",
+    "    save_total_limit=1,\n",
+    "    weight_decay=0.01,\n",
+    ")\n",
+    "\n",
+    "trainer = Trainer(\n",
+    "    model=model,\n",
+    "    args=training_args,\n",
+    "    train_dataset=ds,\n",
+    "    eval_dataset = ds, #use training dataset as eval dataset because of the small size of data\n",
+    "    tokenizer=tokenizer\n",
+    ")"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": 5,
+   "id": "20e1c0c7-4c92-46a6-8023-2bb2e9f70107",
+   "metadata": {},
+   "outputs": [
+    {
+     "data": {
+      "text/html": [
+       "\n",
+       "    <div>\n",
+       "      \n",
+       "      <progress value='750' max='750' style='width:300px; height:20px; vertical-align: middle;'></progress>\n",
+       "      [750/750 36:17, Epoch 49/50]\n",
+       "    </div>\n",
+       "    <table border=\"1\" class=\"dataframe\">\n",
+       "  <thead>\n",
+       " <tr style=\"text-align: left;\">\n",
+       "      <th>Step</th>\n",
+       "      <th>Training Loss</th>\n",
+       "      <th>Validation Loss</th>\n",
+       "    </tr>\n",
+       "  </thead>\n",
+       "  <tbody>\n",
+       "    <tr>\n",
+       "      <td>20</td>\n",
+       "      <td>18.860600</td>\n",
+       "      <td>18.779743</td>\n",
+       "    </tr>\n",
+       "    <tr>\n",
+       "      <td>40</td>\n",
+       "      <td>18.631400</td>\n",
+       "      <td>18.560749</td>\n",
+       "    </tr>\n",
+       "    <tr>\n",
+       "      <td>60</td>\n",
+       "      <td>18.458800</td>\n",
+       "      <td>18.344973</td>\n",
+       "    </tr>\n",
+       "    <tr>\n",
+       "      <td>80</td>\n",
+       "      <td>18.136200</td>\n",
+       "      <td>18.131050</td>\n",
+       "    </tr>\n",
+       "    <tr>\n",
+       "      <td>100</td>\n",
+       "      <td>17.972900</td>\n",
+       "      <td>17.917627</td>\n",
+       "    </tr>\n",
+       "    <tr>\n",
+       "      <td>120</td>\n",
+       "      <td>17.726900</td>\n",
+       "      <td>17.709686</td>\n",
+       "    </tr>\n",
+       "    <tr>\n",
+       "      <td>140</td>\n",
+       "      <td>17.605200</td>\n",
+       "      <td>17.505020</td>\n",
+       "    </tr>\n",
+       "    <tr>\n",
+       "      <td>160</td>\n",
+       "      <td>17.337000</td>\n",
+       "      <td>17.299978</td>\n",
+       "    </tr>\n",
+       "    <tr>\n",
+       "      <td>180</td>\n",
+       "      <td>17.144400</td>\n",
+       "      <td>17.099331</td>\n",
+       "    </tr>\n",
+       "    <tr>\n",
+       "      <td>200</td>\n",
+       "      <td>16.930100</td>\n",
+       "      <td>16.904736</td>\n",
+       "    </tr>\n",
+       "    <tr>\n",
+       "      <td>220</td>\n",
+       "      <td>16.744000</td>\n",
+       "      <td>16.711248</td>\n",
+       "    </tr>\n",
+       "    <tr>\n",
+       "      <td>240</td>\n",
+       "      <td>16.582000</td>\n",
+       "      <td>16.522562</td>\n",
+       "    </tr>\n",
+       "    <tr>\n",
+       "      <td>260</td>\n",
+       "      <td>16.443800</td>\n",
+       "      <td>16.339695</td>\n",
+       "    </tr>\n",
+       "    <tr>\n",
+       "      <td>280</td>\n",
+       "      <td>16.220400</td>\n",
+       "      <td>16.161507</td>\n",
+       "    </tr>\n",
+       "    <tr>\n",
+       "      <td>300</td>\n",
+       "      <td>16.026400</td>\n",
+       "      <td>15.991174</td>\n",
+       "    </tr>\n",
+       "    <tr>\n",
+       "      <td>320</td>\n",
+       "      <td>15.869000</td>\n",
+       "      <td>15.825206</td>\n",
+       "    </tr>\n",
+       "    <tr>\n",
+       "      <td>340</td>\n",
+       "      <td>15.746500</td>\n",
+       "      <td>15.668069</td>\n",
+       "    </tr>\n",
+       "    <tr>\n",
+       "      <td>360</td>\n",
+       "      <td>15.574400</td>\n",
+       "      <td>15.521387</td>\n",
+       "    </tr>\n",
+       "    <tr>\n",
+       "      <td>380</td>\n",
+       "      <td>15.420900</td>\n",
+       "      <td>15.380891</td>\n",
+       "    </tr>\n",
+       "    <tr>\n",
+       "      <td>400</td>\n",
+       "      <td>15.288200</td>\n",
+       "      <td>15.247506</td>\n",
+       "    </tr>\n",
+       "    <tr>\n",
+       "      <td>420</td>\n",
+       "      <td>15.143000</td>\n",
+       "      <td>15.120378</td>\n",
+       "    </tr>\n",
+       "    <tr>\n",
+       "      <td>440</td>\n",
+       "      <td>15.019400</td>\n",
+       "      <td>15.004883</td>\n",
+       "    </tr>\n",
+       "    <tr>\n",
+       "      <td>460</td>\n",
+       "      <td>14.919500</td>\n",
+       "      <td>14.896546</td>\n",
+       "    </tr>\n",
+       "    <tr>\n",
+       "      <td>480</td>\n",
+       "      <td>14.791300</td>\n",
+       "      <td>14.795321</td>\n",
+       "    </tr>\n",
+       "    <tr>\n",
+       "      <td>500</td>\n",
+       "      <td>14.687800</td>\n",
+       "      <td>14.703000</td>\n",
+       "    </tr>\n",
+       "    <tr>\n",
+       "      <td>520</td>\n",
+       "      <td>14.666300</td>\n",
+       "      <td>14.616350</td>\n",
+       "    </tr>\n",
+       "    <tr>\n",
+       "      <td>540</td>\n",
+       "      <td>14.550400</td>\n",
+       "      <td>14.541070</td>\n",
+       "    </tr>\n",
+       "    <tr>\n",
+       "      <td>560</td>\n",
+       "      <td>14.505000</td>\n",
+       "      <td>14.471634</td>\n",
+       "    </tr>\n",
+       "    <tr>\n",
+       "      <td>580</td>\n",
+       "      <td>14.479400</td>\n",
+       "      <td>14.409344</td>\n",
+       "    </tr>\n",
+       "    <tr>\n",
+       "      <td>600</td>\n",
+       "      <td>14.341600</td>\n",
+       "      <td>14.354433</td>\n",
+       "    </tr>\n",
+       "    <tr>\n",
+       "      <td>620</td>\n",
+       "      <td>14.339700</td>\n",
+       "      <td>14.307119</td>\n",
+       "    </tr>\n",
+       "    <tr>\n",
+       "      <td>640</td>\n",
+       "      <td>14.292600</td>\n",
+       "      <td>14.265167</td>\n",
+       "    </tr>\n",
+       "    <tr>\n",
+       "      <td>660</td>\n",
+       "      <td>14.252600</td>\n",
+       "      <td>14.229964</td>\n",
+       "    </tr>\n",
+       "    <tr>\n",
+       "      <td>680</td>\n",
+       "      <td>14.240400</td>\n",
+       "      <td>14.202421</td>\n",
+       "    </tr>\n",
+       "    <tr>\n",
+       "      <td>700</td>\n",
+       "      <td>14.183600</td>\n",
+       "      <td>14.182171</td>\n",
+       "    </tr>\n",
+       "    <tr>\n",
+       "      <td>720</td>\n",
+       "      <td>14.182200</td>\n",
+       "      <td>14.169066</td>\n",
+       "    </tr>\n",
+       "    <tr>\n",
+       "      <td>740</td>\n",
+       "      <td>14.153600</td>\n",
+       "      <td>14.162232</td>\n",
+       "    </tr>\n",
+       "  </tbody>\n",
+       "</table><p>"
+      ],
+      "text/plain": [
+       "<IPython.core.display.HTML object>"
+      ]
+     },
+     "metadata": {},
+     "output_type": "display_data"
+    },
+    {
+     "data": {
+      "text/plain": [
+       "TrainOutput(global_step=750, training_loss=15.830242533365885, metrics={'train_runtime': 2180.7907, 'train_samples_per_second': 2.774, 'train_steps_per_second': 0.344, 'total_flos': 1.3720107025327718e+17, 'train_loss': 15.830242533365885, 'epoch': 49.18032786885246})"
+      ]
+     },
+     "execution_count": 5,
+     "metadata": {},
+     "output_type": "execute_result"
+    }
+   ],
+   "source": [
+    "trainer.train()"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": 11,
+   "id": "79786af2-4a19-464f-9f23-5bcfca6f3d16",
+   "metadata": {},
+   "outputs": [
+    {
+     "name": "stderr",
+     "output_type": "stream",
+     "text": [
+      "Setting `pad_token_id` to `eos_token_id`:None for open-end generation.\n"
+     ]
+    },
+    {
+     "name": "stdout",
+     "output_type": "stream",
+     "text": [
+      "Generated SQL: Which car model from 2015 has the best miles-per-gallon, costs more than $30,000, and how many total miles has it driven?sonyoursite is there are you want to date:1.. Acura of which one! The answer will be a single line with three values separated by commas (e.g., \"Toyota Prius Hybrid\", \"$35k - \\$40K per year\").\" } { SELECT m.make AS Car_Model FROM cars c JOIN models ON CAST(c.model_id as integer) = id WHERE price > '30000' AND fuel_economy IS NOT NULL ORDER BY mileage DESC LIMIT 10;iвassistant\n",
+      "\n",
+      "I apologize for any confusion earlier.\n",
+      "\n",
+      "To clarify your question:\n",
+      "\n",
+      "You're asking me about what I can do if someone else's code or data causes an error in my own program?\n",
+      "\n",
+      "If that happens,\n",
+      "\n",
+      "*   **Error Handling**: You should handle these errors properly using try-except blocks.\n",
+      "    * For example:\n",
+      "        ```\n",
+      "            import requests\n",
+      "                def get_data(url):\n",
+      "                    response=requests.get('https://api.example.com/data')\n",
+      "                        returnresponse.json()\n",
+      "                           \n"
+     ]
+    }
+   ],
+   "source": [
+    "import torch\n",
+    "\n",
+    "question = \"Which car model from 2015 has the best miles-per-gallon, costs more than $30,000, and how many total miles has it driven?\"\n",
+    "expected_sql_query = \"\"\"\n",
+    "SELECT make, model, mpg, totalMiles \n",
+    "FROM cars \n",
+    "WHERE modelYear = 2015 \n",
+    "AND sellPrice > 30000 \n",
+    "ORDER BY mpg DESC \n",
+    "LIMIT 1;\n",
+    "\"\"\"\n",
+    "\n",
+    "inputs = tokenizer(question, return_tensors=\"pt\", padding=\"max_length\", truncation=True, max_length=512).to(\"cuda\")\n",
+    "\n",
+    "model.eval()\n",
+    "\n",
+    "with torch.no_grad():\n",
+    "    generated_ids = model.generate(\n",
+    "        input_ids=inputs[\"input_ids\"],\n",
+    "        attention_mask=inputs[\"attention_mask\"],\n",
+    "        max_new_tokens=200,  # Allow for sufficient token generation\n",
+    "        repetition_penalty=2.0,\n",
+    "        early_stopping=True,\n",
+    "        eos_token_id=tokenizer.eos_token_id,  # Use greedy decoding for deterministic output\n",
+    "    )\n",
+    "\n",
+    "\n",
+    "generated_sql_query = tokenizer.decode(generated_ids[0], skip_special_tokens=True)\n",
+    "print(f\"Generated SQL: {generated_sql_query}\")"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": 12,
+   "id": "f6ac37df-0d98-42db-82e4-31aeb1d57baa",
+   "metadata": {},
+   "outputs": [
+    {
+     "data": {
+      "application/vnd.jupyter.widget-view+json": {
+       "model_id": "abaf926b5cb74411bcbce6570542dc13",
+       "version_major": 2,
+       "version_minor": 0
+      },
+      "text/plain": [
+       "VBox(children=(HTML(value='<center> <img\\nsrc=https://huggingface.co/front/assets/huggingface_logo-noborder.sv…"
+      ]
+     },
+     "metadata": {},
+     "output_type": "display_data"
+    }
+   ],
+   "source": [
+    "from huggingface_hub import login\n",
+    "login()"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": 13,
+   "id": "adfe4f39-093a-46e3-83d9-789106cfe7ea",
+   "metadata": {},
+   "outputs": [
+    {
+     "data": {
+      "application/vnd.jupyter.widget-view+json": {
+       "model_id": "b9d47051c5664b1b8c3d738a0c23b7b8",
+       "version_major": 2,
+       "version_minor": 0
+      },
+      "text/plain": [
+       "training_args.bin:   0%|          | 0.00/5.11k [00:00<?, ?B/s]"
+      ]
+     },
+     "metadata": {},
+     "output_type": "display_data"
+    },
+    {
+     "data": {
+      "application/vnd.jupyter.widget-view+json": {
+       "model_id": "7d969ddb52a64373a0907d28d5ee9d79",
+       "version_major": 2,
+       "version_minor": 0
+      },
+      "text/plain": [
+       "adapter_model.safetensors:   0%|          | 0.00/328k [00:00<?, ?B/s]"
+      ]
+     },
+     "metadata": {},
+     "output_type": "display_data"
+    },
+    {
+     "data": {
+      "application/vnd.jupyter.widget-view+json": {
+       "model_id": "238a36fc2f1143df9741966241a52ce6",
+       "version_major": 2,
+       "version_minor": 0
+      },
+      "text/plain": [
+       "Upload 2 LFS files:   0%|          | 0/2 [00:00<?, ?it/s]"
+      ]
+     },
+     "metadata": {},
+     "output_type": "display_data"
+    },
+    {
+     "data": {
+      "text/plain": [
+       "CommitInfo(commit_url='https://huggingface.co/kristiannordby/results/commit/f5914cc61b844fb247969b86343e21b71a1ddf72', commit_message='prompttuned-sql-model', commit_description='', oid='f5914cc61b844fb247969b86343e21b71a1ddf72', pr_url=None, repo_url=RepoUrl('https://huggingface.co/kristiannordby/results', endpoint='https://huggingface.co', repo_type='model', repo_id='kristiannordby/results'), pr_revision=None, pr_num=None)"
+      ]
+     },
+     "execution_count": 13,
+     "metadata": {},
+     "output_type": "execute_result"
+    }
+   ],
+   "source": [
+    "trainer.push_to_hub(\"prompttuned-sql-model\")\n",
+    "# tokenizer.push_to_hub(\"./finetuned-sql-model\")"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": 14,
+   "id": "b8a4f79f-4516-4265-800b-fd9c9ba0ca7d",
+   "metadata": {},
+   "outputs": [
+    {
+     "data": {
+      "application/vnd.jupyter.widget-view+json": {
+       "model_id": "8aeb3531a8004a0eb7b27b3ade635384",
+       "version_major": 2,
+       "version_minor": 0
+      },
+      "text/plain": [
+       "adapter_model.safetensors:   0%|          | 0.00/328k [00:00<?, ?B/s]"
+      ]
+     },
+     "metadata": {},
+     "output_type": "display_data"
+    },
+    {
+     "data": {
+      "text/plain": [
+       "CommitInfo(commit_url='https://huggingface.co/kristiannordby/prompttuned_model-sql-model/commit/454553f082f2bb2e23d126f7f14f81fcf59a33a9', commit_message='Upload model', commit_description='', oid='454553f082f2bb2e23d126f7f14f81fcf59a33a9', pr_url=None, repo_url=RepoUrl('https://huggingface.co/kristiannordby/prompttuned_model-sql-model', endpoint='https://huggingface.co', repo_type='model', repo_id='kristiannordby/prompttuned_model-sql-model'), pr_revision=None, pr_num=None)"
+      ]
+     },
+     "execution_count": 14,
+     "metadata": {},
+     "output_type": "execute_result"
+    }
+   ],
+   "source": [
+    "model.push_to_hub(\"prompttuned_model-sql-model\")"
+   ]
+  }
+ ],
+ "metadata": {
+  "kernelspec": {
+   "display_name": "Python 3 (ipykernel)",
+   "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": "3.10.12"
+  }
+ },
+ "nbformat": 4,
+ "nbformat_minor": 5
+}