Archivo de pruebas, se logro transferir información a Mongo con archivos...

Archivo de pruebas, se logro transferir información a Mongo con archivos .json, sin embargo para el análisis se utilizaron los archivos .csv

Proyecto_Periodico_Inteligente/MongoD.ipynb

+{
+ "cells": [
+  {
+   "cell_type": "code",
+   "execution_count": null,
+   "metadata": {},
+   "outputs": [],
+   "source": [
+    "from pymongo import MongoClient\n",
+    "import csv\n",
+    "import json\n",
+    "\n",
+    "#Se crea un cliente para el servidor de mongodb\n",
+    "client = MongoClient('mongodb://localhost:27017/')\n",
+    "\n",
+    "#Se crea una base de datos\n",
+    "db = client['PeriodicoInteligente']\n",
+    "\n",
+    "#crear la colección\n",
+    "collection_train = db['noticias_train']\n",
+    "collection_test = db['noticias_test']\n",
+    "\n"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": null,
+   "metadata": {},
+   "outputs": [],
+   "source": [
+    "# Convierte un archivo CSV a JSON\n",
+    "import csv\n",
+    "import json\n",
+    "\n",
+    "csvfile = open('noticias_test.csv', 'r')\n",
+    "jsonfile = open('noticias_test.json', 'w')\n",
+    "\n",
+    "reader = csv.DictReader( csvfile)\n",
+    "count=0\n",
+    "for row in reader:\n",
+    "\tjson.dump(row, jsonfile)\n",
+    "\tjsonfile.write('\\n')\n",
+    "\tcount+=1\n",
+    "print('Se convirtieron ', count, 'registros, archivo .JSON creado ')"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": null,
+   "metadata": {},
+   "outputs": [],
+   "source": [
+    "########Se insertan registros de TEST a la colección#########\n",
+    "file_name = './noticias_test.json'\n",
+    "cont = 0\n",
+    "with open(file_name) as f:\n",
+    "\tfor line in f:\n",
+    "\t\tj_content = json.loads(line)\n",
+    "\t\t#Agregar un dato a la colección\n",
+    "\t\tmydict = {\"FECHA\": j_content['FECHA'], \"PERIODICO\": j_content['PERIODICO'], \"TITULO\": j_content['TITULO'], \"NOTA\": j_content['NOTA'], \"URL\": j_content['URL']}\n",
+    "\t\tx = collection.insert_one(mydict)\n",
+    "\t\tcont = cont+1\n",
+    "\n",
+    "f.close()\n",
+    "print(\"Se ingresaron \"+str(cont)+\" registros\")"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": null,
+   "metadata": {},
+   "outputs": [],
+   "source": [
+    "########Se insertan registros de TRAIN a la colección#########\n",
+    "\n",
+    "file_name = './noticias_train.json'\n",
+    "cont = 0\n",
+    "with open(file_name) as f:\n",
+    "\tfor line in f:\n",
+    "\t\tj_content = json.loads(line)\n",
+    "\t\t#Agregar un dato a la colección\n",
+    "\t\tmydict = {\"data_news\": j_content['data_news'], \"target_news\": j_content['target_news'], \"target_names_news\": j_content['target_names_news']}\n",
+    "\t\tx = collection.insert_one(mydict)\n",
+    "\t\tcont = cont+1\n",
+    "\n",
+    "f.close()\n",
+    "print(\"Se ingresaron \"+str(cont)+\" registros\")"
+   ]
+  }
+ ],
+ "metadata": {
+  "kernelspec": {
+   "display_name": "Python 3",
+   "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": "3.5.3"
+  }
+ },
+ "nbformat": 4,
+ "nbformat_minor": 2
+}