{
 "cells": [
  {
   "attachments": {},
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## Fetch [FIMI](http://fimi.uantwerpen.be/data/) datasets\n",
    "`FIMI` is a popular repository referencing standard datasets and algorithms in pattern mining\n",
    "\n",
    "### Load individual datasets from the repository"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 1,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "This tutorial was tested with the following version of skmine : 1.0.0\n"
     ]
    }
   ],
   "source": [
    "import skmine\n",
    "\n",
    "print(\"This tutorial was tested with the following version of skmine :\", skmine.__version__)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 2,
   "metadata": {},
   "outputs": [],
   "source": [
    "from skmine.datasets.fimi import fetch_chess\n",
    "from skmine.datasets.fimi import fetch_accidents\n",
    "from skmine.datasets.fimi import fetch_kosarak\n",
    "from skmine.datasets.fimi import fetch_iris"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 3,
   "metadata": {
    "scrolled": true
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "100% [............................................................................] 342294 / 342294"
     ]
    },
    {
     "data": {
      "text/plain": [
       "0    [1, 3, 5, 7, 9, 11, 13, 15, 17, 19, 21, 23, 25...\n",
       "1    [1, 3, 5, 7, 9, 12, 13, 15, 17, 19, 21, 23, 25...\n",
       "2    [1, 3, 5, 7, 9, 12, 13, 16, 17, 19, 21, 23, 25...\n",
       "3    [1, 3, 5, 7, 9, 11, 13, 15, 17, 20, 21, 23, 25...\n",
       "4    [1, 3, 5, 7, 9, 11, 13, 15, 17, 19, 21, 23, 25...\n",
       "Name: chess, dtype: object"
      ]
     },
     "execution_count": 3,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "chess = fetch_chess()\n",
    "chess.head()\n",
    "# .str accessor allows horizontal slicing"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 4,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "100% [........................................................................] 35509823 / 35509823"
     ]
    },
    {
     "data": {
      "text/plain": [
       "0    [1, 2, 3, 4, 5, 6, 7, 8, 9, 10, 11, 12, 13, 14...\n",
       "1    [2, 5, 7, 8, 9, 10, 12, 13, 14, 15, 16, 17, 18...\n",
       "2    [7, 10, 12, 13, 14, 15, 16, 17, 18, 20, 25, 28...\n",
       "3    [1, 5, 8, 10, 12, 14, 15, 16, 17, 18, 19, 20, ...\n",
       "4    [5, 8, 10, 12, 14, 15, 16, 17, 18, 21, 22, 24,...\n",
       "Name: accidents, dtype: object"
      ]
     },
     "execution_count": 4,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "accidents = fetch_accidents()\n",
    "accidents.head()"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "You also have access to other known datasets from https://cgi.csc.liv.ac.uk/~frans/KDD/Software/LUCS-KDD-DN/DataSets/dataSets.html that have been **discretized** as the **iris** dataset."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 5,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "\r\n",
      "  0% [                                                                                  ]   0 / 318\r\n",
      "100% [..................................................................................] 318 / 318"
     ]
    },
    {
     "data": {
      "text/plain": [
       "0     [2, 9, 12, 15, 18]\n",
       "1    [1, 10, 11, 14, 17]\n",
       "2    [5, 10, 13, 16, 19]\n",
       "3     [2, 6, 12, 15, 18]\n",
       "4     [1, 8, 11, 14, 17]\n",
       "Name: iris.D19.N150.C3.num, dtype: object"
      ]
     },
     "execution_count": 5,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "iris = fetch_iris()\n",
    "iris.head()"
   ]
  },
  {
   "attachments": {},
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Some datasets like iris also allow **classification and have a target column**.To get X and y separately, use the *return_y* parameter"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 6,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "0     [2, 9, 12, 15]\n",
      "1    [1, 10, 11, 14]\n",
      "2    [5, 10, 13, 16]\n",
      "3     [2, 6, 12, 15]\n",
      "4     [1, 8, 11, 14]\n",
      "Name: iris.D19.N150.C3.num, dtype: object 0    18\n",
      "1    17\n",
      "2    19\n",
      "3    18\n",
      "4    17\n",
      "Name: iris.D19.N150.C3.num, dtype: int64\n"
     ]
    }
   ],
   "source": [
    "X, y = fetch_iris(return_y=True)\n",
    "print(X.head(), y.head())"
   ]
  },
  {
   "attachments": {},
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### Load your own files in FIMI format\n",
    "The **fetch_file** method lets you load your own dataset in FIMI format. You can indicate whether the values in the file, the item identifiers, are **integers or not** (e.g. strings). The performance of the algorithms is better with integers. You can also **specify the separator** between the different items with 'separator'. By default, it's a blank space. (Example FIMI format [here](http://fimi.uantwerpen.be/data/chess.dat))"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 7,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "0    [0, 1, 2]\n",
       "1       [0, 1]\n",
       "2    [1, 2, 3]\n",
       "Name: data, dtype: object"
      ]
     },
     "execution_count": 7,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "from skmine.datasets.fimi import fetch_file\n",
    "db = fetch_file('data.dat', int_values=True)\n",
    "db"
   ]
  },
  {
   "attachments": {},
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### Print basic statistics about a dataset"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 8,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "{'n_items': 75,\n",
       " 'avg_transaction_size': 37.0,\n",
       " 'n_transactions': 3196,\n",
       " 'density': 0.49333333333333335}"
      ]
     },
     "execution_count": 8,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "from skmine.datasets.utils import describe\n",
    "describe(chess)"
   ]
  }
 ],
 "metadata": {
  "kernelspec": {
   "display_name": "Python 3 (ipykernel)",
   "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": "3.10.9"
  },
  "toc": {
   "base_numbering": 1,
   "nav_menu": {},
   "number_sections": true,
   "sideBar": true,
   "skip_h1_title": false,
   "title_cell": "Table of Contents",
   "title_sidebar": "Contents",
   "toc_cell": false,
   "toc_position": {
    "height": "calc(100% - 180px)",
    "left": "10px",
    "top": "150px",
    "width": "261.517px"
   },
   "toc_section_display": true,
   "toc_window_display": true
  }
 },
 "nbformat": 4,
 "nbformat_minor": 4
}