data_process.ipynb

{
 "cells": [
  {
   "cell_type": "code",
   "execution_count": 20,
   "metadata": {},
   "outputs": [],
   "source": [
    "from datasets import load_dataset, ClassLabel"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 45,
   "metadata": {},
   "outputs": [],
   "source": [
    "DATA_PATH = 'data/news'"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "# **SA Training Dataset**"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 42,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "DatasetDict({\n",
       "    train: Dataset({\n",
       "        features: ['summary', 'label'],\n",
       "        num_rows: 1835\n",
       "    })\n",
       "    train_upsample: Dataset({\n",
       "        features: ['summary', 'label'],\n",
       "        num_rows: 3000\n",
       "    })\n",
       "    test: Dataset({\n",
       "        features: ['summary', 'label'],\n",
       "        num_rows: 360\n",
       "    })\n",
       "})"
      ]
     },
     "execution_count": 42,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "# Load Dataset\n",
    "ds = load_dataset('json', data_files={'train': f'{DATA_PATH}/gpt4/news_train_gpt4.jsonl', \n",
    "                                      'train_upsample': f'{DATA_PATH}/gpt4/news_train_upsampled_gpt4.jsonl', \n",
    "                                      'test': f'{DATA_PATH}/gpt4/news_test_gpt4.jsonl'})\n",
    "#ds.update(split_hf(ds['train']))\n",
    "ds = ds.map(lambda ex: {'label': int(ex['label'] + 1)})\n",
    "ds = ds.select_columns(['summary', 'label'])\n",
    "ds"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 43,
   "metadata": {},
   "outputs": [
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "Creating parquet from Arrow format: 100%|██████████| 2/2 [00:00<00:00, 167.79ba/s]\n",
      "Uploading the dataset shards: 100%|██████████| 1/1 [00:01<00:00,  1.05s/it]\n",
      "Creating parquet from Arrow format: 100%|██████████| 3/3 [00:00<00:00, 309.31ba/s]\n",
      "Uploading the dataset shards: 100%|██████████| 1/1 [00:00<00:00,  1.54it/s]\n",
      "Creating parquet from Arrow format: 100%|██████████| 1/1 [00:00<00:00, 215.42ba/s]\n",
      "Uploading the dataset shards: 100%|██████████| 1/1 [00:00<00:00,  1.36it/s]\n"
     ]
    },
    {
     "data": {
      "text/plain": [
       "CommitInfo(commit_url='https://huggingface.co/datasets/MAPAi/Inflation_News_Sentiment_Analysis/commit/7f73129676ec301ed6467ccb8d18d6e497b85f55', commit_message='Upload dataset', commit_description='', oid='7f73129676ec301ed6467ccb8d18d6e497b85f55', pr_url=None, pr_revision=None, pr_num=None)"
      ]
     },
     "execution_count": 43,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "ds.push_to_hub(\"MAPAi/Inflation_News_Sentiment_Analysis\")"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "# **Index Dataset**"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 61,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "Dataset({\n",
       "    features: ['summary', 'date'],\n",
       "    num_rows: 12489\n",
       "})"
      ]
     },
     "execution_count": 61,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "# Load Dataset\n",
    "ds = load_dataset('parquet', data_files=f'{DATA_PATH}/nasdaq_news_index_2010_2023.parquet')['train']\n",
    "#ds.update(split_hf(ds['train']))\n",
    "#ds = ds.map(lambda ex: {'label': int(ex['label'] + 1)})\n",
    "ds = ds.rename_columns({'Date': 'date', 'Summary': 'summary'}).select_columns(['summary', 'date'])\n",
    "ds"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 62,
   "metadata": {},
   "outputs": [
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "Creating parquet from Arrow format: 100%|██████████| 13/13 [00:00<00:00, 335.50ba/s]\n",
      "Uploading the dataset shards: 100%|██████████| 1/1 [00:01<00:00,  1.65s/it]\n"
     ]
    },
    {
     "data": {
      "text/plain": [
       "CommitInfo(commit_url='https://huggingface.co/datasets/MAPAi/inflation_news/commit/aa812b30c9f72979c0e7189e8c7455a3bb8769bd', commit_message='Upload dataset', commit_description='', oid='aa812b30c9f72979c0e7189e8c7455a3bb8769bd', pr_url=None, pr_revision=None, pr_num=None)"
      ]
     },
     "execution_count": 62,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "ds.push_to_hub(\"MAPAi/inflation_news\")"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": []
  }
 ],
 "metadata": {
  "kernelspec": {
   "display_name": "modern_nlp",
   "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": "3.10.13"
  }
 },
 "nbformat": 4,
 "nbformat_minor": 2
}