Recommender.py for FastAPI

mini-projects/movie-rec-system/movie_rec_system/recommendation/recommender.py

@@ -0,0 +1,77 @@
+import json
+import pandas as pd
+from sklearn.feature_extraction.text import TfidfVectorizer
+from sklearn.metrics.pairwise import cosine_similarity
+from recommender_helper import (
+    content_movie_recommender,
+    get_popularity_rmse,
+    get_vote_avg_rmse,
+    get_vote_count_rmse,
+)
+
+
+def connect_duckdb():
+    """
+    Function that automatically connects
+    to duckdb as a GET call upon launch
+    of FastAPI
+
+    Returns a connection
+    """
+    # Will have to adjust this based on
+    # how we set up duckdb instance
+    pass
+
+
+def create_combined(df, weight=2):
+    df["combined"] = df["overview"] + " " + (df["genre_names"] + ", ") * weight
+    return df
+
+
+def get_recommendation(movie: str, num_rec: int = 10, stop_words="english"):
+    # conn = connect_duckdb()
+
+    # use sql/jupysql to query data
+    # convert data to pandas pd called df
+
+    # Create column with overview and genres
+    df = create_combined(1)
+
+    # Vectorize "combined"
+    tfidf = TfidfVectorizer(stop_words=stop_words)
+    tfidf_matrix = tfidf.fit_transform(df["combined"])
+
+    # Compute similarity
+    similarity = cosine_similarity(tfidf_matrix)
+
+    similarity_df = pd.DataFrame(
+        similarity, index=df.title.values, columns=df.title.values
+    )
+
+    movie_list = similarity_df.columns.values
+
+    # Get movie recommendations
+    recommendations = content_movie_recommender(
+        movie, similarity_df, movie_list, num_rec
+    )
+
+    # Compute metrics
+    popularity_rmse = get_popularity_rmse(df, movie, recommendations)
+
+    vote_avg_rmse = get_vote_avg_rmse(df, movie, recommendations)
+
+    vote_count_rmse = get_vote_count_rmse(df, movie, recommendations)
+
+    result = {
+        "movie": movie,
+        "recommendations": recommendations,
+        "metrics": {
+            "popularity": popularity_rmse,
+            "vote_avg": vote_avg_rmse,
+            "vote_count": vote_count_rmse,
+        },
+    }
+
+    result_json = json.dumps(result)
+
+    return result_json