Upload app.py

app.py

@@ -1,5 +1,6 @@
 """
 Visual Search API - HuggingFace Space
+Returns embedding vector for external Pinecone queries
 """
 
 import os
@@ -7,12 +8,7 @@ import gradio as gr
 import torch
 import numpy as np
 from PIL import Image
-
-# Pinecone config from HF Secrets
-PINECONE_API_KEY = os.environ.get('PINECONE_API_KEY')
-PINECONE_HOST = os.environ.get('PINECONE_HOST')
-
-print(f"Pinecone configured: Key={'yes' if PINECONE_API_KEY else 'NO'}, Host={'yes' if PINECONE_HOST else 'NO'}")
+import json
 
 # Model (loaded on first use)
 model = None
@@ -48,79 +44,36 @@ def get_embedding(image: Image.Image) -> list:
         return emb.tolist()
 
 
-def query_pinecone(embedding: list, top_k: int = 12) -> list:
-    """Query Pinecone for similar products."""
-    if not PINECONE_API_KEY or not PINECONE_HOST:
-        return []
-
-    import requests
-
-    resp = requests.post(
-        f"https://{PINECONE_HOST}/query",
-        headers={
-            "Api-Key": PINECONE_API_KEY,
-            "Content-Type": "application/json"
-        },
-        json={
-            "vector": embedding,
-            "topK": top_k,
-            "includeMetadata": True
-        },
-        timeout=15
-    )
-
-    if resp.status_code != 200:
-        return []
-
-    matches = resp.json().get('matches', [])
-    return [
-        {
-            'handle': m.get('metadata', {}).get('handle', m.get('id')),
-            'title': m.get('metadata', {}).get('title', ''),
-            'score': m.get('score', 0),
-        }
-        for m in matches
-    ]
-
-
 def search(image):
-    """Main search function."""
+    """Return embedding vector as JSON."""
     if image is None:
-        return "No image provided"
+        return json.dumps({"error": "No image provided"})
 
     try:
-        print("Getting embedding...")
+        print("Generating embedding...")
         embedding = get_embedding(image)
-        print(f"Embedding length: {len(embedding)}")
+        print(f"Embedding generated: {len(embedding)} dimensions")
 
-        if not PINECONE_API_KEY or not PINECONE_HOST:
-            return f"Pinecone not configured. Key: {'set' if PINECONE_API_KEY else 'missing'}, Host: {'set' if PINECONE_HOST else 'missing'}"
-
-        print("Querying Pinecone...")
-        products = query_pinecone(embedding)
-        print(f"Found {len(products)} products")
-
-        if not products:
-            return "No similar products found (check if index has data)"
+        # Return embedding as JSON
+        result = {
+            "embedding": embedding,
+            "dimensions": len(embedding)
+        }
+        return json.dumps(result, indent=2)
 
-        result = "\n".join([
-            f"{i+1}. {p['title']} ({p['handle']}) - score: {p['score']:.3f}"
-            for i, p in enumerate(products)
-        ])
-        return result
     except Exception as e:
         import traceback
         traceback.print_exc()
-        return f"Error: {str(e)}"
+        return json.dumps({"error": str(e)})
 
 
-# Simple Gradio interface
+# Gradio interface - returns embedding as JSON
 demo = gr.Interface(
     fn=search,
     inputs=gr.Image(type="pil", label="Upload Image"),
-    outputs=gr.Textbox(label="Similar Products", lines=15),
-    title="Visual Product Search",
-    description="Upload an image to find similar products."
+    outputs=gr.Textbox(label="Embedding Vector (JSON)", lines=15),
+    title="Visual Search - Embedding Generator",
+    description="Upload an image to get its 512-dimensional CLIP embedding as JSON."
 )
 
 if __name__ == "__main__":