Update src/streamlit_app.py

src/streamlit_app.py

@@ -5,21 +5,25 @@ import tensorflow as tf
 import joblib
 import os
 import zipfile
+import tempfile
 
-# Define paths
+# Paths
 BASE_DIR = os.path.dirname(__file__)
 ZIP_MODEL_PATH = os.path.join(BASE_DIR, "recommender_model.zip")
-EXTRACTED_MODEL_DIR = os.path.join(BASE_DIR, "recommender_model")
 MOVIES_PATH = os.path.join(BASE_DIR, "movies.csv")
 ENCODINGS_PATH = os.path.join(BASE_DIR, "encodings.pkl")
 
 @st.cache_resource
 def load_model():
     try:
-        if not os.path.exists(EXTRACTED_MODEL_DIR):
+        extract_dir = os.path.join(tempfile.gettempdir(), "recommender_model")
+
+        # Extract only if not already extracted
+        if not os.path.exists(extract_dir):
             with zipfile.ZipFile(ZIP_MODEL_PATH, "r") as zip_ref:
-                zip_ref.extractall(EXTRACTED_MODEL_DIR)
-        return tf.keras.models.load_model(EXTRACTED_MODEL_DIR)
+                zip_ref.extractall(extract_dir)
+
+        return tf.keras.models.load_model(extract_dir)
     except Exception as e:
         st.error(f"Failed to load model: {e}")
         st.stop()
@@ -29,25 +33,29 @@ def load_assets():
     try:
         df_movies = pd.read_csv(MOVIES_PATH)
     except FileNotFoundError:
-        st.error("movies.csv not found.")
+        st.error("❌ Error: movies.csv not found.")
         st.stop()
+
     try:
         user_map, movie_map = joblib.load(ENCODINGS_PATH)
     except FileNotFoundError:
-        st.error("encodings.pkl not found.")
+        st.error("❌ Error: encodings.pkl not found.")
         st.stop()
+
     return df_movies, user_map, movie_map
 
+# Load model and assets
 model = load_model()
 movies_df, user2idx, movie2idx = load_assets()
 reverse_movie_map = {v: k for k, v in movie2idx.items()}
 
+# UI
 st.title("🎬 TensorFlow Movie Recommender")
-st.write("Select some movies you've liked to get recommendations:")
+st.write("Select some movies you've liked to get personalized recommendations:")
 
 movie_titles = movies_df.set_index("movieId")["title"].to_dict()
 movie_choices = [movie_titles[mid] for mid in movie2idx.keys() if mid in movie_titles]
-selected_titles = st.multiselect("Liked movies", sorted(movie_choices))
+selected_titles = st.multiselect("🎞️ Liked movies", sorted(movie_choices))
 
 user_ratings = {}
 for title in selected_titles:
@@ -55,13 +63,13 @@ for title in selected_titles:
     if movie_id:
         user_ratings[movie_id] = 5.0
 
-if st.button("Get Recommendations"):
+if st.button("🎯 Get Recommendations"):
     if not user_ratings:
         st.warning("Please select at least one movie.")
     else:
         liked_indices = [movie2idx[m] for m in user_ratings if m in movie2idx]
         if not liked_indices:
-            st.error("No valid movie encodings found.")
+            st.error("⚠️ No valid movie encodings found.")
             st.stop()
 
         avg_embedding = tf.reduce_mean(model.layers[2](tf.constant(liked_indices)), axis=0, keepdims=True)
@@ -78,6 +86,6 @@ if st.button("Get Recommendations"):
             if len(recommended) >= 10:
                 break
 
-        st.subheader("🎯 Top Recommendations")
+        st.subheader("🍿 Top 10 Recommendations")
         for title, score in recommended:
-            st.write(f"{title} — Score: {score:.3f}")
+            st.write(f"**{title}** — Score: `{score:.3f}`")