dradecic

values = defaultdict(list)
for ind, row in data.iterrows():
    for genre in row['genres'].split('|'):
        values[genre].append(row['rating'])
        
        
genre_lst, rating_lst = [], []
for key, item in values.items():
    if key not in [0, 1]:
        genre_lst.append(key)

dradecic

def make_bar_chart(dataset, attribute, bar_color='#3498db', edge_color='#2980b9', title='Title', xlab='X', ylab='Y', sort_index=False):
    if sort_index == False:
        xs = dataset[attribute].value_counts().index
        ys = dataset[attribute].value_counts().values
    else:
        xs = dataset[attribute].value_counts().sort_index().index
        ys = dataset[attribute].value_counts().sort_index().values
        
    
    fig, ax = plt.subplots(figsize=(14, 7))

dradecic

genre_df = pd.DataFrame(data['genres'].str.split('|').tolist(), index=data['movieId']).stack()
genre_df = genre_df.reset_index([0, 'movieId'])
genre_df.columns = ['movieId', 'Genre']

dradecic

def make_histogram(dataset, attribute, bins=25, bar_color='#3498db', edge_color='#2980b9', title='Title', xlab='X', ylab='Y', sort_index=False):
    if attribute == 'moviePubYear':
        dataset = dataset[dataset['moviePubYear'] != 9999]
        
    fig, ax = plt.subplots(figsize=(14, 7))
    ax.spines['top'].set_visible(False)
    ax.spines['right'].set_visible(False)
    ax.set_title(title, fontsize=24, pad=20)
    ax.set_xlabel(xlab, fontsize=16, labelpad=20)
    ax.set_ylabel(ylab, fontsize=16, labelpad=20)

dradecic

years = []

for title in data['title']:
    year_subset = title[-5:-1]
    try: years.append(int(year_subset))
    except: years.append(9999)
        
data['moviePubYear'] = years
print(len(data[data['moviePubYear'] == 9999]))

dradecic

%matplotlib inline

import numpy as np
import pandas as pd
import matplotlib.pyplot as plt
from collections import defaultdict

# read CSVs
movies = pd.read_csv('data/movies.csv')
ratings = pd.read_csv('data/ratings.csv')

dradecic

def get_similar_movies(movie_title, n_ratings_filter=100, n_recommendations=5):
    similar = matrix.corrwith(matrix[movie_title])
    corr_similar = pd.DataFrame(similar, columns=['correlation'])
    corr_similar.dropna(inplace=True)
    
    orig = data.copy()
    
    corr_with_movie = pd.merge(
        left=corr_similar, 
        right=orig, 

dradecic

matrix = data.pivot_table(
    index='userId',
    columns='title',
    values='rating'
)

dradecic

data['name_length'] = data['name'].apply(lambda x: len(str(x)))
data['name_length_group'] = pd.qcut(data['name_length'], 10)
pledged_per_length_agg = data[['pledged', 'name_length_group']].groupby('name_length_group').median()

indexes = []
for val in pledged_per_length_agg.index.tolist():
    indexes.append(str(val))
    
xs = indexes
ys = np.ravel(pledged_per_length_agg.values)

dradecic

goal_df = data[['goal', 'pledged', 'main_category']]
goal_df_grouped = goal_df.groupby(by='main_category').median()

xs = np.array(goal_df_grouped.index)
ys_1 = goal_df_grouped['goal'].values
ys_2 = goal_df_grouped['pledged'].values

fig, ax = plt.subplots(figsize=(14, 7))
ax.spines['top'].set_visible(False)
ax.spines['right'].set_visible(False)