Update app.py

app.py

@@ -9,12 +9,22 @@ BUNDLE_PATH = "spotify_recommender.pkl"
 with open(BUNDLE_PATH, "rb") as f:
     bundle = pickle.load(f)
 
+# Expected keys in the bundle:
+#   "nn_model"    : fitted NearestNeighbors model
+#   "features"    : feature matrix (N x D), e.g. float32
+#   "track_labels": list of "track_name – artist_name" strings
 nn_model: NearestNeighbors = bundle["nn_model"]
 features: np.ndarray = bundle["features"]
 track_labels = bundle["track_labels"]
 
+# Map label -> index for quick lookup (for ALL songs, not just dropdown)
 label_to_index = {label: i for i, label in enumerate(track_labels)}
 
+# Limit how many songs appear in the dropdown to avoid UI / memory issues
+MAX_DROPDOWN_SONGS = 1000  # you can reduce this to 500 if needed
+display_labels = track_labels[:MAX_DROPDOWN_SONGS]
+
+
 def _split_label(label: str):
     """
     label format: 'track_name – artist_name'
@@ -26,7 +36,6 @@ def _split_label(label: str):
         track_name, artist_name = label, ""
     return track_name, artist_name
 
-
 def recommend_tracks_ui(query_label: str, k: int):
     """
     Gradio-facing function that:
@@ -39,20 +48,17 @@ def recommend_tracks_ui(query_label: str, k: int):
         )
 
     idx = label_to_index[query_label]
-
     n_neighbors = min(len(features), int(k) + 1)
     distances, indices = nn_model.kneighbors(
         features[idx:idx + 1],
         n_neighbors=n_neighbors
     )
-
     distances = distances[0]
     indices = indices[0]
 
     mask = indices != idx
     indices = indices[mask][:int(k)]
     distances = distances[mask][:int(k)]
-
     similarities = 1.0 - distances
 
     rows = []
@@ -84,7 +90,7 @@ def evaluate_mean_similarity_ui(k: int, n_samples: int):
     if n == 0:
         return "No tracks in feature matrix."
 
-    n_samples = min(n_samples, n)
+    n_samples = min(n_samples, n, 200)
 
     rng = np.random.default_rng(42)
     sample_indices = rng.choice(n, size=n_samples, replace=False)
@@ -124,17 +130,22 @@ def evaluate_mean_similarity_ui(k: int, n_samples: int):
 with gr.Blocks(title="Spotify Content-Based Recommender") as demo:
     gr.Markdown("# 🎧 Spotify Content-Based Recommender")
     gr.Markdown(
-        "Select a song and get similar tracks using a trained Nearest Neighbors model."
+        "Select a song (subset of all tracks for performance) and get similar tracks "
+        "using a trained Nearest Neighbors model."
+    )
+    gr.Markdown(
+        f"Currently showing the first **{len(display_labels)}** songs in the dropdown "
+        "to keep the app responsive."
     )
 
     with gr.Tab("Recommender"):
         song_input = gr.Dropdown(
-            choices=track_labels,
+            choices=display_labels,
             label="Choose a track",
         )
         k_input = gr.Slider(
             minimum=1,
-            maximum=20,
+            maximum=15,
             value=10,
             step=1,
             label="Number of recommendations (k)",
@@ -154,10 +165,11 @@ with gr.Blocks(title="Spotify Content-Based Recommender") as demo:
     with gr.Tab("Evaluation"):
         gr.Markdown(
             "We measure quality using **mean cosine similarity** between query tracks "
-            "and their top-k recommendations."
+            "and their top-k recommendations. Evaluation is limited in size to keep "
+            "the app stable."
         )
-        k_eval = gr.Slider(1, 20, value=10, step=1, label="k (top-k neighbors)")
-        n_eval = gr.Slider(50, 500, value=200, step=50, label="Number of random tracks to sample")
+        k_eval = gr.Slider(1, 15, value=10, step=1, label="k (top-k neighbors)")
+        n_eval = gr.Slider(50, 200, value=100, step=25, label="Number of random tracks to sample")
         eval_button = gr.Button("Run evaluation")
         eval_output = gr.Textbox(label="Result")