def recommend():
user_input = dict(request.args)
"""Here is where you take the user input dictionary and pre-process it
for your recommendation function"""
# movies = nmf_recommend(user_input)
movies = random_recommend(MOVIES, 5)
return render_template('recommendation.html', movies=movies)
def recommend():
user_response = dict(request.args)
print(user_response)
#movies = random_recommend(user_input)
movies = random_recommend(user_response)
return render_template('recommendation.html', movies=movies)
def __repr__(self):
return '{} - {}'.format(self.iso, self.name)
def as_dict(self):
return {'name': self.name}
def recommender():
#by default, request.args is an ImmutableMultiDict
#where keys --> names from HTML form, and vals ---> user input
user_input = dict(request.args)
user_input_movies = list(user_input.values())[::2]
user_input_ratings = list(user_input.values())[1::2]
user_input_movies2 = []
"""YOUR TASK:"""
#Write a python function (or multiple) that does the following:
# 1. Take in user input from the HTML form (dictionary)
# 2. Map the movies to the database / dataframe
# 3. Convert the input into a New User Array (601st), with length of ~9000
# 4. Then use the components of the pre-trained NMF to transform this array (predict)
# 5. Map the resulting movie "recommendations" to real names.
# 6. Return the names (e.g. list)
MOVIES = [
x[:-7] for x in (pd.read_csv('movies.csv'))['title'].values.tolist()
]
#check = all(item in user_input_movies for item in MOVIES)
#[True if x in MOVIES else False for x in user_input_movies]
userprintout = [
x if x in MOVIES else 'sorry: This movie does not exist'
for x in user_input_movies
]
user_input2 = zip(userprintout, user_input_ratings)
result_list = random_recommend(user_input_movies, user_input_ratings)
# if check==True:
# elif check==False:
# user_input2 = dictionary
# yield user_input2
return render_template('recommender.html',
result_html=result_list,
user_input_html=user_input2)
def recommender():
user_input = request.args
result_list = random_recommend(8)
return render_template('recommender.html',
result_html=result_list,
user_input=user_input)