custom_field_code.ipynb

{
 "cells": [
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "# Code Editor inside Argilla with a CustomField\n",
    "\n",
    "This notebook is a demonstration of how to use the Argilla's custom field to create a code editor. The editor will allow you to write code and run it in the UI."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "!pip install argilla datasets"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 2,
   "metadata": {},
   "outputs": [],
   "source": [
    "import argilla as rg\n",
    "\n",
    "client = rg.Argilla()"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## Create an Argilla dataset with a custom field\n",
    "\n",
    "First, we need to create a dataset with a custom field. We can pass the `CustomField` class to the `fields` parameter of the `Settings` class."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 6,
   "metadata": {},
   "outputs": [
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "/Users/ben/code/argilla-cookbook/.venv/lib/python3.12/site-packages/argilla/datasets/_resource.py:264: UserWarning: Workspace not provided. Using default workspace: argilla id: 735cae0d-eb08-45c3-ad79-0a11ad4dd2c2\n",
      "  warnings.warn(f\"Workspace not provided. Using default workspace: {workspace.name} id: {workspace.id}\")\n"
     ]
    },
    {
     "data": {
      "text/plain": [
       "Dataset(id=UUID('2be4e1a5-5e5a-4035-b4eb-6d1d6f096245') inserted_at=datetime.datetime(2024, 10, 8, 12, 54, 33, 196235) updated_at=datetime.datetime(2024, 10, 8, 12, 54, 33, 344056) name='Codemirror-0aba08e0-c582-4921-bbee-48ece9305f94' status='ready' guidelines=None allow_extra_metadata=False distribution=OverlapTaskDistributionModel(strategy='overlap', min_submitted=1) workspace_id=UUID('735cae0d-eb08-45c3-ad79-0a11ad4dd2c2') last_activity_at=datetime.datetime(2024, 10, 8, 12, 54, 33, 344056))"
      ]
     },
     "execution_count": 6,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "settings = rg.Settings(\n",
    "    fields=[\n",
    "        rg.TextQuestion(\"instruction\"),\n",
    "        rg.TextQuestion(\"input\"),\n",
    "        rg.CustomField(\"code\", template=\"assets/codemirror.html\"),\n",
    "    ],\n",
    "    questions=[rg.RatingQuestion(\"rating\", [0, 1, 2, 3, 4, 5])],\n",
    ")\n",
    "dataset = rg.Dataset(\n",
    "    name=\"Codemirror-dataset\",\n",
    "    settings=settings,\n",
    ")\n",
    "dataset.create()"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## Load a code dataset\n",
    "\n",
    "We will use the `python_code_instructions_18k_alpaca` dataset because it contains python code, but you could extend this example to other programming languages by adapting the template in `assets/codemirror.html`."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 7,
   "metadata": {},
   "outputs": [],
   "source": [
    "from datasets import load_dataset\n",
    "\n",
    "ds = load_dataset(\"iamtarun/python_code_instructions_18k_alpaca\", split=\"train\")"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Now let's log the dataset's records."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 15,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/html": [
       "<pre style=\"white-space:pre;overflow-x:auto;line-height:normal;font-family:Menlo,'DejaVu Sans Mono',consolas,'Courier New',monospace\">DatasetRecords: The provided batch size <span style=\"color: #008080; text-decoration-color: #008080; font-weight: bold\">256</span> was normalized. Using value <span style=\"color: #008080; text-decoration-color: #008080; font-weight: bold\">1</span>.\n",
       "</pre>\n"
      ],
      "text/plain": [
       "DatasetRecords: The provided batch size \u001b[1;36m256\u001b[0m was normalized. Using value \u001b[1;36m1\u001b[0m.\n"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "Sending records...: 100%|██████████| 1/1 [00:00<00:00, 13.79batch/s]\n"
     ]
    },
    {
     "data": {
      "text/html": [
       "<pre style=\"white-space:pre;overflow-x:auto;line-height:normal;font-family:Menlo,'DejaVu Sans Mono',consolas,'Courier New',monospace\">DatasetRecords: The provided batch size <span style=\"color: #008080; text-decoration-color: #008080; font-weight: bold\">256</span> was normalized. Using value <span style=\"color: #008080; text-decoration-color: #008080; font-weight: bold\">1</span>.\n",
       "</pre>\n"
      ],
      "text/plain": [
       "DatasetRecords: The provided batch size \u001b[1;36m256\u001b[0m was normalized. Using value \u001b[1;36m1\u001b[0m.\n"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "Sending records...: 100%|██████████| 1/1 [00:00<00:00, 16.02batch/s]\n"
     ]
    }
   ],
   "source": [
    "max_samples = 2\n",
    "\n",
    "for sample in ds.select(range(max_samples)):\n",
    "    dataset.records.log(\n",
    "        records=[\n",
    "            {\n",
    "                \"code\": {\n",
    "                    \"value\": sample[\"output\"],\n",
    "                },\n",
    "                \"instruction\": sample[\"instruction\"],\n",
    "                \"input\": sample[\"input\"],\n",
    "                \"rating\": 5,\n",
    "            }\n",
    "        ]\n",
    "    )"
   ]
  }
 ],
 "metadata": {
  "kernelspec": {
   "display_name": ".venv",
   "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": "3.12.4"
  }
 },
 "nbformat": 4,
 "nbformat_minor": 2
}