HF Spaces Fix

.gitattributes

@@ -34,3 +34,7 @@ saved_model/**/* filter=lfs diff=lfs merge=lfs -text
 *.zst filter=lfs diff=lfs merge=lfs -text
 *tfevents* filter=lfs diff=lfs merge=lfs -text
 datasets/*.csv filter=lfs diff=lfs merge=lfs -text
+datasets filter=lfs diff=lfs merge=lfs -text
+datasets/Global Music Artists.csv filter=lfs diff=lfs merge=lfs -text
+datasets/Music.csv filter=lfs diff=lfs merge=lfs -text
+datasets/data.csv filter=lfs diff=lfs merge=lfs -text

README.md

@@ -11,3 +11,20 @@ short_description: A youtube and spotify song and playlist downloader
 ---
 
 Check out the configuration reference at https://huggingface.co/docs/hub/spaces-config-reference
+
+# SongPorter
+
+A music recommendation API that provides song recommendations and artist information.
+
+## Endpoints
+
+- `/` - API info and documentation
+- `/recommendations/` - Get song recommendations (POST)
+- `/artist-info/` - Get artist information (POST)
+
+## Technologies
+
+- FastAPI
+- Pandas & NumPy 
+- Scikit-learn
+- Content-based recommendation

app.py

@@ -1,34 +1,56 @@
 import logging
-from fastapi import FastAPI, HTTPException
+from fastapi import FastAPI, HTTPException, Request
 from pydantic import BaseModel, Field
 from typing import List, Dict, Optional
+import uvicorn
+from fastapi.middleware.cors import CORSMiddleware
+from fastapi.staticfiles import StaticFiles
+import os
+import time
 
-# If using recommender.py:
+# Import recommender and artist utils
 from recommendation import MusicRecommender, get_hardcoded_recommendations
-# Import artist utils
 from artist_utils import get_bulk_artist_info, load_artist_data
 
-logging.basicConfig(level=logging.INFO)
+# Configure logging
+logging.basicConfig(
+    level=logging.INFO,
+    format='%(asctime)s - %(name)s - %(levelname)s - %(message)s',
+)
 logger = logging.getLogger(__name__)
 
-app = FastAPI()
+# Initialize FastAPI app
+app = FastAPI(
+    title="SongPorter API",
+    description="Music recommendation and artist info API",
+    version="1.0.0",
+)
+
+# Add CORS middleware
+app.add_middleware(
+    CORSMiddleware,
+    allow_origins=["*"],  # Allows all origins
+    allow_credentials=True,
+    allow_methods=["*"],  # Allows all methods
+    allow_headers=["*"],  # Allows all headers
+)
 
 # --- Initialize Recommender ---
 recommender = None
 try:
+    logger.info("Initializing Music Recommender...")
     recommender = MusicRecommender()
     logger.info("Music Recommender loaded successfully.")
 except Exception as e:
     logger.error(f"Failed to load Music Recommender: {e}", exc_info=True)
 
 # --- Initialize Artist Data ---
-# Ensure artist data is loaded on startup
 if not load_artist_data():
      logger.error("CRITICAL: Failed to load artist data on startup.")
-     # Decide if the app should run without artist data. For now, it will, but log errors.
+     # App will run without artist data, but log errors.
 
 # --- API Input Models ---
-class RecommendationRequestData(BaseModel): # Renamed from UserData for clarity
+class RecommendationRequestData(BaseModel):
     recent_song_ids: List[str] = Field(..., example=["spotify_id_1", "song_name_2"])
     top_genres: List[str] = Field(..., example=["Pop", "Rock"])
     limit: int = 10
@@ -36,29 +58,37 @@ class RecommendationRequestData(BaseModel): # Renamed from UserData for clarity
 class ArtistInfoRequestData(BaseModel):
     artist_names: List[str] = Field(..., example=["Artist Name 1", "Another Artist"])
 
+# --- Request timing middleware ---
+@app.middleware("http")
+async def add_process_time_header(request: Request, call_next):
+    start_time = time.time()
+    response = await call_next(request)
+    process_time = time.time() - start_time
+    response.headers["X-Process-Time"] = str(process_time)
+    return response
+
 # --- API Endpoints ---
 
 @app.post("/recommendations/")
-async def get_recommendations_endpoint(request_data: RecommendationRequestData): # Use renamed model
+async def get_recommendations_endpoint(request_data: RecommendationRequestData):
     if recommender is None:
         logger.error("Recommender not available.")
-        return {"recommendations": get_hardcoded_recommendations(request_data.limit), "message": "Recommender unavailable, returning popular songs."}
+        return {"recommendations": get_hardcoded_recommendations(request_data.limit), 
+                "message": "Recommender unavailable, returning popular songs."}
 
     try:
         logger.info(f"Received recommendation request: {request_data.dict()}")
-        # (Keep the recommendation logic from the previous step here)
-        # ... existing recommendation logic using request_data ...
-
+        # Process recommendation logic
         all_recommendations = []
 
         # 1. Content-based from recent songs
-        for song_id in request_data.recent_song_ids[:5]: # Limit seed songs
+        for song_id in request_data.recent_song_ids[:5]:  # Limit seed songs
             song_recommendations = recommender.find_similar_songs(song_id, n=20)
             if song_recommendations:
                 all_recommendations.extend(song_recommendations)
 
         # 2. Genre-based
-        for genre in request_data.top_genres[:3]: # Limit seed genres
+        for genre in request_data.top_genres[:3]:  # Limit seed genres
              if genre and genre != 'Unknown':
                 genre_recommendations = recommender.get_recommendations_by_genre(genre, n=10)
                 if genre_recommendations:
@@ -103,7 +133,7 @@ async def get_artist_info_endpoint(request_data: ArtistInfoRequestData) -> Dict[
     try:
         logger.info(f"Received artist info request for {len(request_data.artist_names)} artists.")
         if not request_data.artist_names:
-            return {} # Return empty dict if no names provided
+            return {}  # Return empty dict if no names provided
 
         artist_info = get_bulk_artist_info(request_data.artist_names)
         logger.info(f"Returning info for {len(artist_info)} artists.")
@@ -116,4 +146,23 @@ async def get_artist_info_endpoint(request_data: ArtistInfoRequestData) -> Dict[
 
 @app.get("/")
 async def root():
-    return {"message": "TuneVault Recommender & Artist Info API"}
+    return {
+        "message": "SongPorter API",
+        "endpoints": [
+            {"path": "/", "method": "GET", "description": "This help message"},
+            {"path": "/recommendations/", "method": "POST", "description": "Get song recommendations"},
+            {"path": "/artist-info/", "method": "POST", "description": "Get artist information"}
+        ],
+        "version": "1.0.0"
+    }
+
+# --- Serve static files if they exist ---
+static_dir = os.path.join(os.path.dirname(os.path.abspath(__file__)), "static")
+if os.path.exists(static_dir):
+    app.mount("/static", StaticFiles(directory=static_dir), name="static")
+    logger.info(f"Mounted static files from {static_dir}")
+
+# --- For local development ---
+if __name__ == "__main__":
+    port = int(os.environ.get("PORT", 8000))
+    uvicorn.run("app:app", host="0.0.0.0", port=port, reload=True)

artist_utils.py

@@ -17,8 +17,14 @@ def load_artist_data():
         # Get the directory of the current script
         current_dir = os.path.dirname(os.path.abspath(__file__))
         datasets_path = os.path.join(current_dir, 'datasets')
-        # --- IMPORTANT: Adjust this filename if yours is different ---
-        csv_path = os.path.join(datasets_path, 'Global_Music_Artists.csv')
+        # --- IMPORTANT: Check for both filename variants ---
+        csv_path = os.path.join(datasets_path, 'Global Music Artists.csv')
+        if not os.path.exists(csv_path):
+            # Try alternative filename
+            csv_path = os.path.join(datasets_path, 'Global_Music_Artists.csv')
+            if not os.path.exists(csv_path):
+                logger.error(f"Artist CSV file not found at: {csv_path}")
+                raise FileNotFoundError(f"Artist CSV file not found in datasets directory")
         # --- ---
 
         if not os.path.exists(csv_path):

recommendation.py

@@ -2,15 +2,11 @@ import logging
 import numpy as np
 import pandas as pd
 import os
-from django.conf import settings
 from datetime import datetime
 from sklearn.metrics.pairwise import cosine_similarity
 from sklearn.preprocessing import StandardScaler
 from sklearn.pipeline import Pipeline
 from sklearn.cluster import KMeans
-from django.utils import timezone
-from django.db.models import Count
-from django.db.models.functions import TruncDate
 
 logger = logging.getLogger(__name__)
 
@@ -79,7 +75,9 @@ class MusicRecommender:
         
         try:
             # Load datasets
-            datasets_path = os.path.join(settings.BASE_DIR, 'songs', 'datasets')
+            # Get the directory of the current script for HF Spaces compatibility
+            current_dir = os.path.dirname(os.path.abspath(__file__))
+            datasets_path = os.path.join(current_dir, 'datasets')
             
             # Skip Music.csv as it has an invalid format with song titles as column names
             self.logger.info("Skipping Music.csv due to invalid format, using data.csv directly")
@@ -91,12 +89,16 @@ class MusicRecommender:
             self.logger.info("Loaded data.csv dataset")
             
             # Load genre and year data
-            self.genre_data = pd.read_csv(os.path.join(datasets_path, 'data_by_genres.csv'),
-                                        on_bad_lines='skip',
-                                        engine='python')
-            self.year_data = pd.read_csv(os.path.join(datasets_path, 'data_by_year.csv'),
-                                      on_bad_lines='skip',
-                                      engine='python')
+            try:
+                self.genre_data = pd.read_csv(os.path.join(datasets_path, 'data_by_genres.csv'),
+                                            on_bad_lines='skip',
+                                            engine='python')
+                self.year_data = pd.read_csv(os.path.join(datasets_path, 'data_by_year.csv'),
+                                          on_bad_lines='skip',
+                                          engine='python')
+                self.logger.info("Loaded genre and year datasets")
+            except FileNotFoundError:
+                self.logger.warning("Genre or year datasets not found, continuing with limited functionality")
             
             # Prepare data and build clusters
             self._prepare_data()
@@ -725,4 +727,4 @@ def update_user_recommendations(user):
         
     except Exception as e:
         logger.error(f"Error updating recommendations: {e}")
-        return False 
+        return False

requirements.txt

@@ -3,4 +3,8 @@ uvicorn[standard]
 pandas
 numpy
 scikit-learn
-python-dotenv # Optional, if you use environment variables
+python-dotenv # Optional, if you use environment variables
+matplotlib # For visualizations if needed
+requests # For external API calls
+pillow # For image handling
+plotly # Optional for interactive visualizations