Criando uma cara para contar relações de quem fez X, também fez X

Readme.md

Relacionamento de quem fez X, também fez X

O objetivo aqui é pegar uma entrada composta por um csv com: produto, sessão. E considerando que a ação X foi aplicada a ele criar um cara que pega as relações por sessão.

Exemplo de entrada:

1231312124,cb808cb9-3bfd-4c0c-ad20-2e181b9b3a93
1231212124,cb808cb9-3bfd-4c0c-ad20-2e181b9b3a93
1231112124,cb808cb9-3bfd-4c0c-ad20-2e181b9b3a93
1231762124,cb808cb9-3bfd-4c0c-ad20-2e181b9b3a93

A saida vai mapear:

Produto: [products: numero_de_vezes_que_ocorreu]

Exemplo de saída:

{"5767613": {"3096140": 2, "2068884": 2, "3252281": 2, "2333267": 2, "3654870": 2 (...)}

Tudo foi feito usando apenas os built-in do python. Ainda tem muito a ser melhorado, mas é um exemplo bem didático

wxax.ipynb

{
 "cells": [
  {
   "cell_type": "code",
   "execution_count": 1,
   "metadata": {
    "collapsed": true
   },
   "outputs": [],
   "source": [
    "import uuid\n",
    "import random\n",
    "\n",
    "from collections import Counter\n",
    "from functools import reduce\n",
    "\n",
    "\n",
    "def _entries_generator():\n",
    "    entry = ''\n",
    "    products = [str(random.randrange(1000000, 9000000)) for _ in range(201)]\n",
    "    for _ in range(random.randrange(50, 100)):\n",
    "        entry += _entry_generator(products)\n",
    "    return entry\n",
    "\n",
    "\n",
    "def _entry_generator(products):\n",
    "    session = str(uuid.uuid4())\n",
    "    number_of_interactions = random.randrange(1, 30)\n",
    "    entry = ''\n",
    "    for _ in range(number_of_interactions):\n",
    "        entry += '{product},{session}\\n'.format(\n",
    "            product=products[random.randrange(0, 200)],\n",
    "            session=session\n",
    "        )\n",
    "\n",
    "    return entry\n",
    "\n",
    "\n",
    "def _format_entry(entry, data):\n",
    "    if data:\n",
    "        product_id, session_id = data.split(',')\n",
    "\n",
    "        data = {\n",
    "            'product_id': product_id,\n",
    "        }\n",
    "\n",
    "        if session_id in entry:\n",
    "            entry[session_id].append(data)\n",
    "        else:\n",
    "            entry[session_id] = [data]\n",
    "\n",
    "    return entry\n",
    "\n",
    "\n",
    "def _wxax(entry, data):\n",
    "    if data:\n",
    "        products = [d['product_id'] for d in data]\n",
    "\n",
    "        for product in products:\n",
    "            if product in entry:\n",
    "                entry[product].extend([p for p in products if p != product])\n",
    "            else:\n",
    "                entry[product] = [p for p in products if p != product]\n",
    "    return entry\n",
    "\n",
    "\n",
    "# def _score(rec):\n",
    "#     scored = {}\n",
    "#     for k, value in rec.items():\n",
    "#         model = collections.defaultdict(lambda: 1)\n",
    "#         for v in value:\n",
    "#             if v != k:\n",
    "#                 model[v] += 1\n",
    "#         scored[k] = model\n",
    "#     return scored\n",
    "\n",
    "\n",
    "def run():\n",
    "    views = _entries_generator().split('\\n')\n",
    "    formated = reduce(_format_entry, views, {})\n",
    "    wxax = {\n",
    "        k: v for k, v in reduce(_wxax, formated.values(), {}).items()\n",
    "        if len(v) > 0\n",
    "        }\n",
    "\n",
    "    return {k: Counter(v) for k, v in wxax.items()}\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    ""
   ]
  }
 ],
 "metadata": {
  "kernelspec": {
   "display_name": "Python 3",
   "language": "python",
   "name": "python3"
  },
  "language_info": {
   "codemirror_mode": {
    "name": "ipython",
    "version": 3.0
   },
   "file_extension": ".py",
   "mimetype": "text/x-python",
   "name": "python",
   "nbconvert_exporter": "python",
   "pygments_lexer": "ipython3",
   "version": "3.5.2"
  }
 },
 "nbformat": 4,
 "nbformat_minor": 0
}