made changes to fasten hf spaces start

.gitattributes

@@ -1,9 +1,10 @@
-\# filter=lfs diff=lfs merge=lfs -text
+# Audio files
 *.mp3 filter=lfs diff=lfs merge=lfs -text
 *.ogg filter=lfs diff=lfs merge=lfs -text
 *.wav filter=lfs diff=lfs merge=lfs -text
 *.flac filter=lfs diff=lfs merge=lfs -text
 *.m4a filter=lfs diff=lfs merge=lfs -text
+
 # Image files
 *.png filter=lfs diff=lfs merge=lfs -text
 *.jpg filter=lfs diff=lfs merge=lfs -text
@@ -12,14 +13,3 @@
 *.bmp filter=lfs diff=lfs merge=lfs -text
 *.webp filter=lfs diff=lfs merge=lfs -text
 *.tiff filter=lfs diff=lfs merge=lfs -text
-file filter=lfs diff=lfs merge=lfs -text
-types filter=lfs diff=lfs merge=lfs -text
-needed) filter=lfs diff=lfs merge=lfs -text
-Track filter=lfs diff=lfs merge=lfs -text
-PNG filter=lfs diff=lfs merge=lfs -text
-files filter=lfs diff=lfs merge=lfs -text
-can filter=lfs diff=lfs merge=lfs -text
-more filter=lfs diff=lfs merge=lfs -text
-as filter=lfs diff=lfs merge=lfs -text
-(you filter=lfs diff=lfs merge=lfs -text
-add filter=lfs diff=lfs merge=lfs -text

Dockerfile

@@ -1,5 +1,5 @@
 # Multi-stage Dockerfile for Dance Movement Analyzer
-# Optimized for production deployment
+# Optimized for Hugging Face Spaces
 
 # Stage 1: Base image with dependencies
 FROM python:3.10-slim as base
@@ -8,10 +8,11 @@ FROM python:3.10-slim as base
 ENV PYTHONUNBUFFERED=1 \
     PYTHONDONTWRITEBYTECODE=1 \
     PIP_NO_CACHE_DIR=1 \
-    PIP_DISABLE_PIP_VERSION_CHECK=1
+    PIP_DISABLE_PIP_VERSION_CHECK=1 \
+    DEBIAN_FRONTEND=noninteractive
 
 # Install system dependencies
-RUN apt-get update && apt-get install -y \
+RUN apt-get update && apt-get install -y --no-install-recommends \
     libgl1 \
     libglib2.0-0 \
     libsm6 \
@@ -19,6 +20,7 @@ RUN apt-get update && apt-get install -y \
     libxrender-dev \
     libgomp1 \
     ffmpeg \
+    curl \
     && rm -rf /var/lib/apt/lists/*
 
 # Create app directory
@@ -28,11 +30,14 @@ WORKDIR /app
 FROM base as dependencies
 
 # Copy requirements first for better caching
-COPY requirements.txt .
+COPY backend/requirements.txt .
 
 # Install Python dependencies
 RUN pip install --no-cache-dir -r requirements.txt
 
+# Pre-download MediaPipe models to speed up first run
+RUN python -c "import mediapipe as mp; pose = mp.solutions.pose.Pose(); pose.close()" || true
+
 # Stage 3: Production image
 FROM base as production
 
@@ -49,6 +54,12 @@ COPY backend/app /app/app
 # Copy frontend files
 COPY frontend /app/frontend
 
+# Copy startup script
+COPY startup.sh /app/startup.sh
+
+# Make startup script executable
+RUN chmod +x /app/startup.sh
+
 # Set permissions
 RUN chmod -R 755 /app
 
@@ -58,12 +69,16 @@ RUN useradd -m -u 1000 appuser && \
 
 USER appuser
 
-# Expose port
+# Expose port (7860 for Hugging Face, 8000 for local)
 EXPOSE 8000
+EXPOSE 7860
 
-# Health check
-HEALTHCHECK --interval=30s --timeout=10s --start-period=40s --retries=3 \
+# ⬇️ OPTION B: HEALTHCHECK GOES HERE ⬇️
+# Health check with 5 minute startup period (for model loading)
+# This allows the container up to 5 minutes to start before health checks begin
+HEALTHCHECK --interval=30s --timeout=10s --start-period=5m --retries=3 \
     CMD python -c "import requests; requests.get('http://localhost:8000/health')" || exit 1
+# ⬆️ HEALTHCHECK ENDS HERE ⬆️
 
-# Run the application
-CMD ["uvicorn", "app.main:app", "--host", "0.0.0.0", "--port", "8000"]
+# Run the application using startup script
+CMD ["/app/startup.sh"]

backend/app/__init__.py

@@ -4,7 +4,7 @@ Provides pose detection, movement classification, and rhythm analysis
 """
 
 __version__ = "1.0.0"
-__author__ = "Your Name"
+__author__ = "Prathamesh Vaidya"
 
 from .config import Config
 from .pose_analyzer import PoseAnalyzer, PoseKeypoints

backend/app/config.py

@@ -54,7 +54,8 @@ class Config:
     # ==================== MediaPipe Configuration ====================
     # Model complexity: 0 (Lite), 1 (Full), 2 (Heavy)
     MEDIAPIPE_MODEL_COMPLEXITY: int = int(
-        os.getenv("MEDIAPIPE_MODEL_COMPLEXITY", 1)
+        # os.getenv("MEDIAPIPE_MODEL_COMPLEXITY", 1)
+        os.getenv("MEDIAPIPE_MODEL_COMPLEXITY", 0)
     )
     
     # Confidence thresholds

backend/app/main.py

@@ -1,12 +1,12 @@
 """
-FastAPI Application - Phase 3
-REST API and WebSocket endpoints for Dance Movement Analysis
+FastAPI Application with Optimized Startup for Hugging Face Spaces
 """
 
 from fastapi import FastAPI, File, UploadFile, WebSocket, WebSocketDisconnect, HTTPException, Request
 from fastapi.responses import FileResponse, JSONResponse, HTMLResponse, StreamingResponse
 from fastapi.middleware.cors import CORSMiddleware
 from fastapi.staticfiles import StaticFiles
+from contextlib import asynccontextmanager
 from pathlib import Path
 import asyncio
 import json
@@ -29,33 +29,57 @@ logging.basicConfig(
 )
 logger = logging.getLogger(__name__)
 
-# Initialize FastAPI app
+# Global processor instance (initialized on startup)
+global_processor: Optional[VideoProcessor] = None
+
+@asynccontextmanager
+async def lifespan(app: FastAPI):
+    """Lifespan event handler for startup and shutdown"""
+    global global_processor
+    
+    # Startup
+    logger.info("🚀 Starting Dance Movement Analyzer...")
+    
+    # Initialize folders
+    Config.initialize_folders()
+    logger.info("✅ Folders initialized")
+    
+    # Pre-initialize VideoProcessor to load MediaPipe models
+    logger.info("📦 Loading MediaPipe models...")
+    global_processor = VideoProcessor()
+    logger.info("✅ MediaPipe models loaded successfully")
+    
+    logger.info("🎉 Application startup complete!")
+    
+    yield
+    
+    # Shutdown (cleanup if needed)
+    logger.info("👋 Shutting down application...")
+
+# Initialize FastAPI app with lifespan
 app = FastAPI(
     title="Dance Movement Analysis API",
     description="AI-powered dance movement analysis with pose detection and classification",
     version="1.0.0",
     docs_url="/api/docs",
-    redoc_url="/api/redoc"
+    redoc_url="/api/redoc",
+    lifespan=lifespan
 )
 
 # CORS Configuration
 app.add_middleware(
     CORSMiddleware,
-    allow_origins=["*"],  # In production, specify exact origins
+    allow_origins=["*"],
     allow_credentials=True,
     allow_methods=["*"],
     allow_headers=["*"],
 )
 
-# Initialize folders
-Config.initialize_folders()
-
-# Mount static files for frontend
+# Mount static files
 static_path = Path(__file__).parent.parent.parent / "frontend"
 if static_path.exists():
     app.mount("/static", StaticFiles(directory=str(static_path)), name="static")
 
-# Setup templates and static files
 templates = Jinja2Templates(directory=static_path)
 
 # Active WebSocket connections
@@ -64,32 +88,19 @@ active_connections: Dict[str, WebSocket] = {}
 # Processing sessions
 processing_sessions: Dict[str, Dict[str, Any]] = {}
 
-'''
 def convert_to_native_bool(obj):
-    if isinstance(obj, np.bool_):  # Check for numpy boolean
-        return bool(obj)
-    elif isinstance(obj, dict):  # If it's a dictionary, convert its values
-        return {k: convert_to_native_bool(v) for k, v in obj.items()}
-    elif isinstance(obj, list):  # If it's a list, convert each item
-        return [convert_to_native_bool(item) for item in obj]
-    else:
-        return obj
-'''
-
-def convert_to_native_bool(obj):
-    """Recursively convert numpy.bool_ and nested structures to native Python types."""
+    """Recursively convert numpy types to native Python types."""
     if isinstance(obj, np.bool_):
         return bool(obj)
     elif isinstance(obj, (np.integer, np.floating)):
-        return obj.item()  # Convert numpy numbers to Python int/float
+        return obj.item()
     elif isinstance(obj, dict):
         return {k: convert_to_native_bool(v) for k, v in obj.items()}
     elif isinstance(obj, (list, tuple)):
         return [convert_to_native_bool(v) for v in obj]
     else:
         return obj
-    
-    
+
 class ConnectionManager:
     """Manages WebSocket connections for real-time updates"""
     
@@ -126,23 +137,8 @@ class ConnectionManager:
         for session_id in disconnected:
             self.disconnect(session_id)
 
-
 manager = ConnectionManager()
 
-
-def progress_callback_factory(session_id: str):
-    """Create a progress callback function for a specific session"""
-    
-    async def callback(progress: float, message: str):
-        await manager.send_message(session_id, {
-            "type": "progress",
-            "progress": progress,
-            "message": message,
-            "timestamp": datetime.now().isoformat()
-        })
-    
-    return callback
-
 @app.get("/", response_class=HTMLResponse)
 async def home(request: Request):
     """Home page."""
@@ -150,11 +146,12 @@ async def home(request: Request):
 
 @app.get("/info")
 async def root():
-    """Root endpoint - serves frontend or API info"""
+    """Root endpoint - serves API info"""
     return {
         "name": "Dance Movement Analysis API",
         "version": "1.0.0",
         "status": "online",
+        "models_loaded": global_processor is not None,
         "endpoints": {
             "upload": "/api/upload",
             "analyze": "/api/analyze/{session_id}",
@@ -164,32 +161,24 @@ async def root():
         }
     }
 
-
 @app.get("/health")
 async def health_check():
     """Health check endpoint"""
     return {
         "status": "healthy",
+        "models_loaded": global_processor is not None,
         "timestamp": datetime.now().isoformat(),
         "active_sessions": len(processing_sessions),
-        "active_connections": len(active_connections)
+        "active_connections": len(manager.active_connections)
     }
 
-
 @app.post("/api/upload")
 async def upload_video(file: UploadFile = File(...)):
-    """
-    Upload a video file for processing
-    
-    Returns:
-        JSON with session_id and file info
-    """
-
+    """Upload a video file for processing"""
     from typing import List
     allowed_extensions: List[str] = [".mp4", ".avi", ".mov", ".mkv", ".webm"]
 
     try:
-        # Generate unique session ID
         session_id = str(uuid.uuid4())
 
         # Validate file
@@ -203,10 +192,9 @@ async def upload_video(file: UploadFile = File(...)):
         with open(upload_path, "wb") as buffer:
             shutil.copyfileobj(file.file, buffer)
         
-        # Get video info
-        processor = VideoProcessor()
+        # Use pre-initialized processor
+        processor = global_processor or VideoProcessor()
         video_info = processor.load_video(upload_path)
-        print("get video info")
         
         # Store session info
         processing_sessions[session_id] = {

spaces.yaml

@@ -5,4 +5,6 @@ colorTo: indigo
 sdk: docker
 app_port: 8000
 pinned: false
-short_description: AI-powered tool for real-time dance movement analysis.
+short_description: AI-powered tool for real-time dance movement analysis.
+startup_duration_timeout: 10m
+python_version: "3.10"

startup.sh

@@ -0,0 +1,34 @@
+#!/bin/bash
+set -e
+
+echo "🚀 Starting Dance Movement Analyzer..."
+
+# Pre-warm MediaPipe models
+echo "📦 Pre-loading MediaPipe models..."
+python3 -c "
+import mediapipe as mp
+import logging
+logging.basicConfig(level=logging.INFO)
+logger = logging.getLogger(__name__)
+
+try:
+    logger.info('Loading Pose model...')
+    pose = mp.solutions.pose.Pose(
+        static_image_mode=False,
+        model_complexity=1,
+        min_detection_confidence=0.5,
+        min_tracking_confidence=0.5
+    )
+    pose.close()
+    logger.info('✅ Models loaded successfully')
+except Exception as e:
+    logger.error(f'❌ Error loading models: {e}')
+    # Continue anyway - models will load on first use
+" || echo "⚠️ Model pre-loading failed, will load on first use"
+
+# Detect port (Hugging Face uses 7860, local uses 8000)
+PORT=${PORT:-8000}
+
+# Start the application
+echo "🎬 Starting Uvicorn server on port $PORT..."
+exec uvicorn app.main:app --host 0.0.0.0 --port $PORT --workers 1