feat: integrate HuggingFace Spaces GPU acceleration

- Added HuggingFace Spaces GPU decorator import with fallback
- Wrapped system initialization and query functions with

@GPU
decorator for GPU acceleration
- Updated query logic to support hybrid retrieval mode
- Added huggingface_hub[spaces] dependency to requirements.txt

This enables efficient GPU utilization on HuggingFace Spaces for improved performance in embedding and retrieval operations.

app.py

@@ -8,6 +8,13 @@ Provides a web UI for the FractalStat RAG system with GPU acceleration.
 import gradio as gr
 import time
 
+# Import HuggingFace Spaces GPU decorator
+try:
+    from spaces import GPU
+except ImportError:
+    # Fallback if not available
+    GPU = lambda func: func
+
 # Import Warbler CDA components
 from warbler_cda.retrieval_api import RetrievalAPI, RetrievalQuery, RetrievalMode
 from warbler_cda.embeddings import EmbeddingProviderFactory
@@ -18,26 +25,34 @@ from warbler_cda.pack_loader import PackLoader
 # Initialize the system
 print("🚀 Initializing Warbler CDA...")
 
-# Create embedding provider (will use sentence-transformers with GPU if available)
-embedding_provider = EmbeddingProviderFactory.get_default_provider()
-print(f"✅ Embedding provider: {embedding_provider.get_provider_info()['provider_id']}")
+@GPU
+def initialize_system():
+    """Initialize the Warbler CDA system with GPU support."""
+    # Create embedding provider (will use sentence-transformers with GPU if available)
+    embedding_provider = EmbeddingProviderFactory.get_default_provider()
+    print(f"✅ Embedding provider: {embedding_provider.get_provider_info()['provider_id']}")
+
+    # Create semantic anchors (required by RetrievalAPI)
+    semantic_anchors = SemanticAnchorGraph(embedding_provider=embedding_provider)
+    print("✅ Semantic anchors initialized")
 
-# Create semantic anchors (required by RetrievalAPI)
-semantic_anchors = SemanticAnchorGraph(embedding_provider=embedding_provider)
-print("✅ Semantic anchors initialized")
+    # Create FractalStat bridge
+    fractalstat_bridge = FractalStatRAGBridge()
+    print("✅ FractalStat bridge initialized")
 
-# Create FractalStat bridge
-fractalstat_bridge = FractalStatRAGBridge()
-print("✅ FractalStat bridge initialized")
+    # Create RetrievalAPI with proper components
+    api = RetrievalAPI(
+        semantic_anchors=semantic_anchors,
+        embedding_provider=embedding_provider,
+        fractalstat_bridge=fractalstat_bridge,
+        config={"enable_fractalstat_hybrid": True}
+    )
+    print("✅ RetrievalAPI initialized")
 
-# Create RetrievalAPI with proper components
-api = RetrievalAPI(
-    semantic_anchors=semantic_anchors,
-    embedding_provider=embedding_provider,
-    fractalstat_bridge=fractalstat_bridge,
-    config={"enable_fractalstat_hybrid": True}
-)
-print("✅ RetrievalAPI initialized")
+    return api
+
+# Initialize with GPU support
+api = initialize_system()
 
 # Load packs
 print("📚 Loading Warbler packs...")
@@ -109,6 +124,7 @@ else:
 print(f"🎉 Warbler CDA ready with {api.get_context_store_size()} documents!")
 
 
+@GPU
 def query_warbler(query_text: str, max_results: int = 5, use_hybrid: bool = True) -> str:
     """Query the Warbler CDA system."""
     if not query_text.strip():
@@ -116,10 +132,12 @@ def query_warbler(query_text: str, max_results: int = 5, use_hybrid: bool = True
     
     start_time = time.time()
     
-    # Create query
+    # Create query - use hybrid mode when requested
+    query_mode = RetrievalMode.HYBRID_SEMANTIC_FRACTALSTAT if use_hybrid else RetrievalMode.SEMANTIC_SIMILARITY
+
     query = RetrievalQuery(
         query_id=f"gradio_{int(time.time())}",
-        mode=RetrievalMode.SEMANTIC_SIMILARITY,
+        mode=query_mode,
         semantic_query=query_text,
         max_results=max_results,
         fractalstat_hybrid=use_hybrid

requirements.txt

@@ -16,6 +16,9 @@ python-multipart>=0.0.6
 # Gradio for HuggingFace Space
 gradio>=5.5.0
 
+# HuggingFace Spaces SDK
+huggingface_hub[spaces]>=0.25.0
+
 # CLI
 click>=8.1.0