udaytag commited on
Commit
94dd0a4
1 Parent(s): 795a3a1

Upload 5 files

Browse files
Movie recommender system.jpg ADDED
app.py ADDED
@@ -0,0 +1,122 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ import pickle
2
+ import streamlit as st
3
+ from sklearn.metrics.pairwise import cosine_similarity
4
+ from PIL import Image
5
+ from scipy.sparse import csr_matrix
6
+
7
+
8
+ @st.cache_data
9
+ def get_recommendation(title):
10
+ idx = df1.index[df1['title'] == title][0]
11
+ poster = f'https://image.tmdb.org/t/p/w500/{df1.loc[idx, "poster_path"]}'
12
+
13
+ # Get the pairwise similarity scores of all movies with that movie
14
+ sim_scores = list(enumerate(
15
+ cosine_similarity(
16
+ tfidf_matrix,
17
+ tfidf_matrix[idx])))
18
+
19
+ # Sort the movies based on the similarity scores
20
+ sim_scores = sorted(sim_scores, key=lambda x: x[1], reverse=True)
21
+
22
+ # Get the scores of the 10 most similar movies
23
+ sim_scores = sim_scores[1:13]
24
+
25
+ # Get the movie indices
26
+ movie_indices = [i[0] for i in sim_scores]
27
+
28
+ # Return the top 10 most similar movies
29
+ result = df1.iloc[movie_indices]
30
+
31
+ recommended_movie_names = []
32
+ recommended_movie_posters = []
33
+ recommended_movie_overview = []
34
+
35
+ for i, j in enumerate(result.poster_path):
36
+ recommended_movie_names.append(result.iloc[i].title)
37
+ recommended_movie_posters.append(f'https://image.tmdb.org/t/p/w500/{j}')
38
+ recommended_movie_overview.append(result.iloc[i].overview)
39
+ return poster, recommended_movie_names, recommended_movie_posters, recommended_movie_overview
40
+
41
+
42
+ image = Image.open('Movie recommender system.jpg')
43
+ st.title("Movie Recommendation Engine")
44
+ st.image(image)
45
+
46
+ st.markdown('For this project, I developed a Content-Based Recommendation System that suggests movies based on attributes like genre, director, description, and actors. The system operates on the principle that users who enjoyed a specific movie or show are likely to appreciate other movies or shows with similar characteristics.')
47
+
48
+ df1 = pickle.load(open('movie_list.pkl', 'rb'))
49
+ tfidf_matrix = pickle.load(open('tfidf_matrix.pkl', 'rb'))
50
+
51
+ movies_list = df1['title'].values
52
+ selected_movie = st.sidebar.selectbox('Select a Movie', movies_list) # Move selectbox to the sidebar
53
+
54
+ if st.sidebar.button('Show Recommendation'): # Move button to the sidebar
55
+ poster, recommended_movie_names, recommended_movie_posters, recommended_movie_overview = get_recommendation(
56
+ selected_movie)
57
+ st.image(poster, width=160)
58
+ col1, col2, col3, col4 = st.columns(4)
59
+ with col1:
60
+ st.image(recommended_movie_posters[0])
61
+ st.markdown(recommended_movie_names[0])
62
+ with st.expander("OverView"):
63
+ st.write(recommended_movie_overview[0])
64
+
65
+ st.image(recommended_movie_posters[4])
66
+ st.markdown(recommended_movie_names[4])
67
+ with st.expander("OverView"):
68
+ st.write(recommended_movie_overview[4])
69
+
70
+ st.image(recommended_movie_posters[8])
71
+ st.markdown(recommended_movie_names[8])
72
+ with st.expander("OverView"):
73
+ st.write(recommended_movie_overview[8])
74
+
75
+ with col2:
76
+ st.image(recommended_movie_posters[1])
77
+ st.markdown(recommended_movie_names[1])
78
+ with st.expander("OverView"):
79
+ st.write(recommended_movie_overview[1])
80
+
81
+ st.image(recommended_movie_posters[5])
82
+ st.markdown(recommended_movie_names[5])
83
+ with st.expander("OverView"):
84
+ st.write(recommended_movie_overview[5])
85
+
86
+ st.image(recommended_movie_posters[9])
87
+ st.markdown(recommended_movie_names[9])
88
+ with st.expander("OverView"):
89
+ st.write(recommended_movie_overview[9])
90
+
91
+ with col3:
92
+ st.image(recommended_movie_posters[2])
93
+ st.markdown(recommended_movie_names[2])
94
+ with st.expander("OverView"):
95
+ st.write(recommended_movie_overview[2])
96
+
97
+ st.image(recommended_movie_posters[6])
98
+ st.markdown(recommended_movie_names[6])
99
+ with st.expander("OverView"):
100
+ st.write(recommended_movie_overview[6])
101
+
102
+ st.image(recommended_movie_posters[10])
103
+ st.markdown(recommended_movie_names[10])
104
+ with st.expander("OverView"):
105
+ st.write(recommended_movie_overview[10])
106
+
107
+ with col4:
108
+ st.image(recommended_movie_posters[3])
109
+ st.markdown(recommended_movie_names[3])
110
+ with st.expander("OverView"):
111
+ st.write(recommended_movie_overview[3])
112
+
113
+ st.image(recommended_movie_posters[7])
114
+ st.markdown(recommended_movie_names[7])
115
+ with st.expander("OverView"):
116
+ st.write(recommended_movie_overview[7])
117
+
118
+ st.image(recommended_movie_posters[11])
119
+ st.markdown(recommended_movie_names[11])
120
+ with st.expander("OverView"):
121
+ st.write(recommended_movie_overview[11])
122
+ st.sidebar.markdown("Made by Uday Singh")
movie_list.pkl ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:24cf160e196ce93c871caf36bcb5d1b1ee16ba5b6ad20f44a6ac40556a4bf605
3
+ size 53612081
requirements.txt ADDED
@@ -0,0 +1,6 @@
 
 
 
 
 
 
 
1
+ streamlit
2
+ pillow
3
+ scipy
4
+ scikit-learn==1.0.2
5
+ pandas==1.3.5
6
+ numpy==1.21.6
tfidf_matrix.pkl ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:8fdc5ad0bec25dad6f3c948c137a67b01506566f8efbaf0eebd44dbe30e514a0
3
+ size 20298550