Update app.py

app.py

@@ -4,8 +4,8 @@ import librosa
 import numpy as np
 from transformers import ASTFeatureExtractor, ASTForAudioClassification
 
-# ── CONFIG ──────────────────────────────────────────────────────────────────
-HF_REPO    = "kashishvijayvergiya/music-genre-ast"   # your HF repo
+# CONFIG───────
+HF_REPO    = "vectorverse/Messy_Mashup_Genre_Classifier"
 SAMPLE_RATE = 16000
 DURATION    = 20
 MAX_LENGTH  = SAMPLE_RATE * DURATION
@@ -21,7 +21,7 @@ GENRE_EMOJI = {
     "reggae": "🌴", "rock": "🔥"
 }
 
-# ── LOAD MODEL (once at startup) ────────────────────────────────────────────
+#LOAD MODEL (once at startup) 
 print("Loading model...")
 feature_extractor = ASTFeatureExtractor.from_pretrained(HF_REPO)
 model = ASTForAudioClassification.from_pretrained(HF_REPO)
@@ -30,7 +30,7 @@ DEVICE = "cuda" if torch.cuda.is_available() else "cpu"
 model.to(DEVICE)
 print(f"Model ready on {DEVICE}!")
 
-# ── AUDIO HELPERS ───────────────────────────────────────────────────────────
+# AUDIO HELPERS 
 def load_audio(path):
     y, _ = librosa.load(path, sr=SAMPLE_RATE, mono=True)
     return y.astype(np.float32)
@@ -50,7 +50,7 @@ def center_crop(y):
         return y[start:start + MAX_LENGTH]
     return np.pad(y, (0, MAX_LENGTH - len(y)))
 
-# ── PREDICTION WITH TTA ─────────────────────────────────────────────────────
+# PREDICTION WITH TTA
 def predict(audio_path):
     if audio_path is None:
         return "Please upload an audio file.", None
@@ -91,7 +91,7 @@ def predict(audio_path):
     result = f"## {GENRE_EMOJI.get(pred_genre, '')} {pred_genre.capitalize()}\n**Confidence: {confidence:.1f}%**"
     return result, label_probs
 
-# ── GRADIO UI ────────────────────────────────────────────────────────────────
+# GRADIO UI 
 with gr.Blocks(title="🎵 Music Genre Classifier") as demo:
     gr.Markdown(
         """
@@ -129,4 +129,4 @@ with gr.Blocks(title="🎵 Music Genre Classifier") as demo:
     )
 
 if __name__ == "__main__":
-    demo.launch()
+    demo.launch()