{
 "cells": [
  {
   "attachments": {},
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Recuperate sentences from HuggingFace\n",
    "----------------------"
   ]
  },
  {
   "attachments": {},
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "- French sentences"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 2,
   "metadata": {},
   "outputs": [],
   "source": [
    "import requests\n",
    "import pandas as pd"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 19,
   "metadata": {},
   "outputs": [],
   "source": [
    "curl = \"https://datasets-server.huggingface.co/rows?dataset=Harsit%2Fxnli2.0_french&config=Harsit--xnli2.0_french&split=test&offset=0&limit=100\"\n",
    "\n",
    "def query():\n",
    "    \n",
    "    response = requests.get(curl)\n",
    "    return response.json()\n",
    "\n",
    "data = query()"
   ]
  },
  {
   "attachments": {},
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Get the hypothesis in a data frame."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 22,
   "metadata": {},
   "outputs": [],
   "source": [
    "sentences = data['rows']\n",
    "\n",
    "french_sents = []\n",
    "\n",
    "for i in range(len(sentences)):\n",
    "    french_sents.append(sentences[i]['row']['hypothesis'])"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 23,
   "metadata": {},
   "outputs": [],
   "source": [
    "# save the sentences\n",
    "pd.DataFrame({'sentences': french_sents}).to_csv('french.csv', index=False)"
   ]
  },
  {
   "attachments": {},
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "- Wolof sentences"
   ]
  },
  {
   "attachments": {},
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "1. First one hundred"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 2,
   "metadata": {},
   "outputs": [],
   "source": [
    "curl = \"https://datasets-server.huggingface.co/rows?dataset=perrynelson%2Fwaxal-wolof&config=perrynelson--waxal-wolof&split=test&offset=0&limit=100\"\n",
    "\n",
    "def query():\n",
    "    \n",
    "    response = requests.get(curl)\n",
    "    return response.json()\n",
    "\n",
    "data = query()"
   ]
  },
  {
   "attachments": {},
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Get the transcriptions in a data frame."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 5,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "100"
      ]
     },
     "execution_count": 5,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "len(data['rows'])"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 3,
   "metadata": {},
   "outputs": [],
   "source": [
    "sentences = data['rows']\n",
    "\n",
    "wolof_sents = []\n",
    "\n",
    "for i in range(len(sentences)):\n",
    "    \n",
    "    wolof_sents.append(sentences[i]['row']['transcription'])"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 28,
   "metadata": {},
   "outputs": [],
   "source": [
    "# save the sentences\n",
    "pd.DataFrame({'sentences': wolof_sents}).to_csv('wolof_1.csv', index=False)"
   ]
  },
  {
   "attachments": {},
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "2. second ..."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 3,
   "metadata": {},
   "outputs": [],
   "source": [
    "curl = \"https://datasets-server.huggingface.co/rows?dataset=perrynelson%2Fwaxal-wolof&config=perrynelson--waxal-wolof&split=test&offset=100&limit=100\"\n",
    "\n",
    "def query():\n",
    "    \n",
    "    response = requests.get(curl)\n",
    "    return response.json()\n",
    "\n",
    "data = query()"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Get the transcriptions in a data frame."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 4,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "100"
      ]
     },
     "execution_count": 4,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "len(data['rows'])"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 5,
   "metadata": {},
   "outputs": [],
   "source": [
    "sentences = data['rows']\n",
    "\n",
    "wolof_sents = []\n",
    "\n",
    "for i in range(len(sentences)):\n",
    "    \n",
    "    wolof_sents.append(sentences[i]['row']['transcription'])"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 6,
   "metadata": {},
   "outputs": [],
   "source": [
    "# save the sentences\n",
    "pd.DataFrame({'sentences': wolof_sents}).to_csv('wolof_2.csv', index=False)"
   ]
  }
 ],
 "metadata": {
  "kernelspec": {
   "display_name": "pytorch1-HleOW5am-py3.10",
   "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": "3.10.8"
  },
  "orig_nbformat": 4
 },
 "nbformat": 4,
 "nbformat_minor": 2
}