from flask import Flask, jsonify, render_template, request
import tmdbsimple as tmdb
from yamr.dataset import EnhancedDataset
from yamr.recommender import RecommendationEngine
import tmdb_util
import util
app = Flask(__name__)
tmdb.API_KEY = tmdb._get_env_key('TMDB_API_KEY')
dataset_path = 'datasets/ml-latest-enhanced'
ds = EnhancedDataset(dataset_path)
recommender = RecommendationEngine(ds)
recommender.load_model('item_sim_model')
@app.route('/')
def index():
return render_template('index.html')
@app.route('/api/movies/<int:movieId>')
def movie_detail(movieId):
movie = ds.find_movie_by_id(movieId)
movie = tmdb_util.add_trailer(movie)
return jsonify(movie)
@app.route('/api/movies/top_rated')
"""
before running this script, do
activate dato-env
ipcluster start -n 4
"""
def chunks(l, n):
"""Yield successive n-sized chunks from l."""
for i in xrange(0, len(l), n):
yield l[i : i + n]
tmdbsimple.API_KEY = tmdbsimple._get_env_key("TMDB_API_KEY")
original = OriginalDataset("datasets/ml-latest")
try:
enhanced = gl.SFrame(data="datasets/ml-latest-enhanced/movies.csv")
except:
import unicodedata
init_tmdb_info = tmdb_util.get_tmdb_info(original.movies[0])
sf_tmdb_info = gl.SFrame(init_tmdb_info)
enhanced = original.movies.join(sf_tmdb_info, on="movieId")
enhanced.export_csv("datasets\\ml-latest-enhanced\\movies.csv")
not_yet_enhanced_movies = original.movies.filter_by(enhanced["movieId"], "movieId", exclude=True)
not_yet_enhanced_movies, dropped_movies = not_yet_enhanced_movies.dropna_split(columns="tmdbId")
