{
 "cells": [
  {
   "cell_type": "code",
   "execution_count": 1,
   "metadata": {},
   "outputs": [],
   "source": [
    "import pandas as pd\n",
    "import numpy as np"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "# Fix \"-\" in \"Collected_by\" Columns in full_master and Heliconius subsets\n",
    "\n",
    "See [discussion](https://huggingface.co./datasets/imageomics/Heliconius-Collection_Cambridge-Butterfly/discussions/7). I then tested each column that showed up as \"null\" in type on the dataset viewer (searching for df[col] == \"-\")."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 2,
   "metadata": {},
   "outputs": [],
   "source": [
    "# while main is commit 3264ec40dbc9db025e027ddb2abca0914cd6500f\n",
    "df = pd.read_csv(\"https://huggingface.co./datasets/imageomics/Heliconius-Collection_Cambridge-Butterfly/resolve/main/img_master.csv\", low_memory= False)\n",
    "df_heli = pd.read_csv(\"https://huggingface.co./datasets/imageomics/Heliconius-Collection_Cambridge-Butterfly/resolve/main/Heliconius_img_master.csv\", low_memory=False)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 3,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "column Collected_by has 62 instances of '-'\n"
     ]
    }
   ],
   "source": [
    "for col in list(df.columns):\n",
    "    temp = df.loc[df[col] == \"-\"]\n",
    "    if temp.shape[0] > 0:\n",
    "        print(f\"column {col} has {temp.shape[0]} instances of '-'\")"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 4,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "column Collected_by has 62 instances of '-'\n"
     ]
    }
   ],
   "source": [
    "for col in list(df_heli.columns):\n",
    "    temp = df_heli.loc[df_heli[col] == \"-\"]\n",
    "    if temp.shape[0] > 0:\n",
    "        print(f\"column {col} has {temp.shape[0]} instances of '-'\")"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Looks like we just need to fix the `Collected_by` column for both CSVs. We'll replace these with `null` so that it doesn't break the dataset viewer."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 5,
   "metadata": {},
   "outputs": [],
   "source": [
    "df.loc[df[\"Collected_by\"] == \"-\", \"Collected_by\"] = np.nan\n",
    "df_heli.loc[df_heli[\"Collected_by\"] == \"-\", \"Collected_by\"] = np.nan"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## Now let's save the fix"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 6,
   "metadata": {},
   "outputs": [],
   "source": [
    "df.to_csv(\"../img_master.csv\", index = False)\n",
    "df_heli.to_csv(\"../Heliconius_img_master.csv\", index = False)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": []
  }
 ],
 "metadata": {
  "kernelspec": {
   "display_name": "std",
   "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": "3.11.3"
  }
 },
 "nbformat": 4,
 "nbformat_minor": 2
}