Deployment_model.py

#!/usr/bin/env python
# coding: utf-8

# In[90]:


import pandas as pd
import numpy as np
import matplotlib.pyplot as plt 
import seaborn as sns

# Importing all the necessary data files.
links = pd.read_csv('links.csv')
movies = pd.read_csv('movies.csv')
ratings = pd.read_csv('ratings.csv')
tags = pd.read_csv('tags.csv')

display(links)
print()

display(movies)
print()

display(ratings)
print()

display(tags)
print()


# In[ ]:





# In[ ]:





# In[ ]:





# In[49]:


#Using regular expressions to find a year stored between parentheses
#We specify the parantheses so we don't conflict with movies that have years in their titles
movies['year'] = movies.title.str.extract('(\(\d\d\d\d\))',expand=False)
#Removing the parentheses
movies['year'] = movies.year.str.extract('(\d\d\d\d)',expand=False)
#Removing the years from the 'title' column
movies['title'] = movies.title.str.replace('(\(\d\d\d\d\))', '')
#Applying the strip function to get rid of any ending whitespace characters that may have appeared
movies['title'] = movies['title'].apply(lambda x: x.strip())
movies.head()


# In[50]:


#Every genre is separated by a | so we simply have to call the split function on |
movies['genres'] = movies.genres.str.split('|')
movies.head()


# In[51]:


#Copying the movie dataframe into a new one since we won't need to use the genre information in our first case.
moviesWithGenres_df = movies.copy()

#For every row in the dataframe, iterate through the list of genres and place a 1 into the corresponding column
for index, row in movies.iterrows():
    for genre in row['genres']:
        moviesWithGenres_df.at[index, genre] = 1
#Filling in the NaN values with 0 to show that a movie doesn't have that column's genre
moviesWithGenres_df = moviesWithGenres_df.fillna(0)
moviesWithGenres_df.head()


# In[52]:


#Drop removes a specified row or column from a dataframe
ratings = ratings.drop('timestamp', 1)
ratings.head()


# In[72]:


import pickle

pickle.dump(ratings, open('ratings.pkl', 'wb'))


# In[53]:


userInput = [
            {'title':'Breakfast Club, The', 'rating':5},
            {'title':'Toy Story', 'rating':3.5},
            {'title':'Jumanji', 'rating':2},
            {'title':"Pulp Fiction", 'rating':5},
            {'title':'Akira', 'rating':4.5}
         ] 
inputMovies = pd.DataFrame(userInput)
inputMovies


# In[54]:


#Filtering out the movies by title
inputId = movies[movies['title'].isin(inputMovies['title'].tolist())]
#Then merging it so we can get the movieId. It's implicitly merging it by title.
inputMovies = pd.merge(inputId, inputMovies)
#Dropping information we won't use from the input dataframe
inputMovies = inputMovies.drop('genres', 1).drop('year', 1)
#Final input dataframe
#If a movie you added in above isn't here, then it might not be in the original 
#dataframe or it might spelled differently, please check capitalisation.
inputMovies


# In[55]:


#Filtering out the movies from the input
userMovies = moviesWithGenres_df[moviesWithGenres_df['movieId'].isin(inputMovies['movieId'].tolist())]
userMovies


# In[56]:


#Resetting the index to avoid future issues
userMovies = userMovies.reset_index(drop=True)
#Dropping unnecessary issues due to save memory and to avoid issues
userGenreTable = userMovies.drop('movieId', 1).drop('title', 1).drop('genres', 1).drop('year', 1)
userGenreTable


# In[57]:


inputMovies['rating']


# In[58]:


#Dot produt to get weights
userProfile = userGenreTable.transpose().dot(inputMovies['rating'])
#The user profile
userProfile


# In[59]:


#Now let's get the genres of every movie in our original dataframe
genreTable = moviesWithGenres_df.set_index(moviesWithGenres_df['movieId'])
#And drop the unnecessary information
genreTable = genreTable.drop('movieId', 1).drop('title', 1).drop('genres', 1).drop('year', 1)
genreTable.head()


# In[60]:


genreTable.shape


# In[61]:


#Multiply the genres by the weights and then take the weighted average
recommendationTable_df = ((genreTable*userProfile).sum(axis=1))/(userProfile.sum())
recommendationTable_df.head()


# In[62]:


#Sort our recommendations in descending order
recommendationTable_df = recommendationTable_df.sort_values(ascending=False)
#Just a peek at the values
recommendationTable_df.head()


# In[63]:


#The final recommendation table
movies.loc[movies['movieId'].isin(recommendationTable_df.head(10).keys())]


# In[64]:


import pickle


# In[70]:


pickle.dump(movies, open('movies.pkl', 'wb'))


# In[91]:


links = links.drop('imdbId', 1)


# In[92]:


# links['tmdbId'].fillna(0)
# links.replace([np.inf, -np.inf], 0, inplace=True)

# links['tmdbId'] = links['tmdbId'].astype('int', errors = 'ignore')
# links.head()


# In[93]:


type(links['tmdbId'])


# In[96]:


links.tail()


# In[95]:



pickle.dump(links, open('links.pkl', 'wb'))


# In[ ]:





# In[ ]: