Upload 4 files

movie_recommendation.py

@@ -0,0 +1,86 @@
+# -*- coding: utf-8 -*-
+"""Movie Recommendation.ipynb
+
+Automatically generated by Colaboratory.
+
+Original file is located at
+    https://colab.research.google.com/drive/16mb8GFViCsAzCEZxBKLbV12h3pQEoU_l
+"""
+
+# Commented out IPython magic to ensure Python compatibility.
+# %pip install gradio
+
+import pandas as pd
+import requests
+
+movies_df = pd.read_csv('./movies.csv')
+links_df = pd.read_csv('./links.csv')
+combined_df = pd.concat([movies_df, links_df[['imdbId','tmdbId']]], axis=1)
+combined_df = combined_df.set_index('title')
+
+combined_df.head()
+
+df = movies_df[['title','genres']]
+df.head()
+
+print(df.isnull().sum())
+
+from sklearn.feature_extraction.text import TfidfVectorizer
+
+tf = TfidfVectorizer(analyzer='word', ngram_range=(1, 3), min_df=0, stop_words='english')
+
+matrix = tf.fit_transform(df['genres'])
+
+from sklearn.metrics.pairwise import linear_kernel
+
+cosine_similarities = linear_kernel(matrix,matrix)
+
+movie_title = df['title']
+
+indices = pd.Series(df.index, index=df['title'])
+
+def movie_recommend(original_title):
+
+    id = 'tt'+str(combined_df.loc[[original_title]].imdbId.values).replace('[','').replace(']','').zfill(7)
+
+    URL = f"http://www.omdbapi.com/?i={id}&apikey=3bd2165d"
+  
+    # sending get request and saving the response as response object
+    r = requests.get(url = URL)
+      
+    # extracting data in json format
+    data = r.json()
+
+    poster_url = data['Poster']
+
+    idx = indices[original_title]
+
+    sim_scores = list(enumerate(cosine_similarities[idx]))
+
+    sim_scores = sorted(sim_scores, key=lambda x: x[1], reverse=True)
+
+    sim_scores = sim_scores[2:12]
+
+    movie_indices = [i[0] for i in sim_scores]
+
+    results = pd.DataFrame(list(data.items()), columns=['Key','Value']).head(20)
+
+    movies = pd.DataFrame(movie_title.iloc[movie_indices].reset_index(drop=True))
+
+    return results, movies, poster_url
+
+import gradio as gr
+
+with gr.Blocks(title='Movie Recommendation') as Intf:
+  gr.Markdown(value='Content Based Recommendation System')
+  with gr.Row():
+    with gr.Column():
+      inp = gr.Dropdown(choices=list(df['title']), label="Choose Movie")
+      btn = gr.Button("Run")
+      gr.Markdown(value='Movie Details')
+      results = gr.DataFrame()
+    poster = gr.Image(label="Poster Image")
+  recomms = gr.DataFrame(label='Recommended Content, Similar to this Movie')
+  btn.click(fn=movie_recommend, inputs=inp, outputs=[results,recomms,poster])
+
+Intf.launch(debug=False)

requirements.txt

@@ -0,0 +1,4 @@
+gradio
+pandas
+requests
+scikit_learn