course/videos/save_load_dataset.ipynb

{
 "cells": [
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "This notebook regroups the code sample of the video below, which is a part of the [Hugging Face course](https://huggingface.co/course)."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {
    "cellView": "form"
   },
   "outputs": [
    {
     "data": {
      "text/html": [
       "<iframe width=\"560\" height=\"315\" src=\"https://www.youtube.com/embed/blF9uxYcKHo?rel=0&amp;controls=0&amp;showinfo=0\" frameborder=\"0\" allowfullscreen></iframe>"
      ],
      "text/plain": [
       "<IPython.core.display.HTML object>"
      ]
     },
     "execution_count": null,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "#@title\n",
    "from IPython.display import HTML\n",
    "\n",
    "HTML('<iframe width=\"560\" height=\"315\" src=\"https://www.youtube.com/embed/blF9uxYcKHo?rel=0&amp;controls=0&amp;showinfo=0\" frameborder=\"0\" allowfullscreen></iframe>')"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Install the Transformers and Datasets libraries to run this notebook."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "! pip install datasets transformers[sentencepiece]"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "from datasets import load_dataset\n",
    "\n",
    "raw_datasets = load_dataset(\"allocine\")\n",
    "raw_datasets.cache_files"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "raw_datasets.save_to_disk(\"my-arrow-datasets\")"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "from datasets import load_from_disk\n",
    "\n",
    "arrow_datasets_reloaded = load_from_disk(\"my-arrow-datasets\")\n",
    "arrow_datasets_reloaded"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "for split, dataset in raw_datasets.items():\n",
    "    dataset.to_csv(f\"my-dataset-{split}.csv\", index=None)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "data_files = {\n",
    "    \"train\": \"my-dataset-train.csv\",\n",
    "    \"validation\": \"my-dataset-validation.csv\",\n",
    "    \"test\": \"my-dataset-test.csv\",\n",
    "}\n",
    "\n",
    "csv_datasets_reloaded = load_dataset(\"csv\", data_files=data_files)\n",
    "csv_datasets_reloaded"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "# Save in JSON Lines format\n",
    "for split, dataset in raw_datasets.items():\n",
    "    dataset.to_json(f\"my-dataset-{split}.jsonl\")\n",
    "\n",
    "# Save in Parquet format\n",
    "for split, dataset in raw_datasets.items():\n",
    "    dataset.to_parquet(f\"my-dataset-{split}.parquet\")"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "json_data_files = {\n",
    "    \"train\": \"my-dataset-train.jsonl\",\n",
    "    \"validation\": \"my-dataset-validation.jsonl\",\n",
    "    \"test\": \"my-dataset-test.jsonl\",\n",
    "}\n",
    "\n",
    "parquet_data_files = {\n",
    "    \"train\": \"my-dataset-train.parquet\",\n",
    "    \"validation\": \"my-dataset-validation.parquet\",\n",
    "    \"test\": \"my-dataset-test.parquet\",\n",
    "}\n",
    "\n",
    "# Reload with the `json` script\n",
    "json_datasets_reloaded = load_dataset(\"json\", data_files=json_data_files)\n",
    "# Reload with the `parquet` script\n",
    "parquet_datasets_reloaded = load_dataset(\"parquet\", data_files=parquet_data_files)"
   ]
  }
 ],
 "metadata": {
  "colab": {
   "name": "Saving and reloading a dataset",
   "provenance": []
  }
 },
 "nbformat": 4,
 "nbformat_minor": 4
}