Upload 4 files

.gitattributes

@@ -1,35 +1,35 @@
-*.7z filter=lfs diff=lfs merge=lfs -text
-*.arrow filter=lfs diff=lfs merge=lfs -text
-*.bin filter=lfs diff=lfs merge=lfs -text
-*.bz2 filter=lfs diff=lfs merge=lfs -text
-*.ckpt filter=lfs diff=lfs merge=lfs -text
-*.ftz filter=lfs diff=lfs merge=lfs -text
-*.gz filter=lfs diff=lfs merge=lfs -text
-*.h5 filter=lfs diff=lfs merge=lfs -text
-*.joblib filter=lfs diff=lfs merge=lfs -text
-*.lfs.* filter=lfs diff=lfs merge=lfs -text
-*.mlmodel filter=lfs diff=lfs merge=lfs -text
-*.model filter=lfs diff=lfs merge=lfs -text
-*.msgpack filter=lfs diff=lfs merge=lfs -text
-*.npy filter=lfs diff=lfs merge=lfs -text
-*.npz filter=lfs diff=lfs merge=lfs -text
-*.onnx filter=lfs diff=lfs merge=lfs -text
-*.ot filter=lfs diff=lfs merge=lfs -text
-*.parquet filter=lfs diff=lfs merge=lfs -text
-*.pb filter=lfs diff=lfs merge=lfs -text
-*.pickle filter=lfs diff=lfs merge=lfs -text
-*.pkl filter=lfs diff=lfs merge=lfs -text
-*.pt filter=lfs diff=lfs merge=lfs -text
-*.pth filter=lfs diff=lfs merge=lfs -text
-*.rar filter=lfs diff=lfs merge=lfs -text
-*.safetensors filter=lfs diff=lfs merge=lfs -text
-saved_model/**/* filter=lfs diff=lfs merge=lfs -text
-*.tar.* filter=lfs diff=lfs merge=lfs -text
-*.tar filter=lfs diff=lfs merge=lfs -text
-*.tflite filter=lfs diff=lfs merge=lfs -text
-*.tgz filter=lfs diff=lfs merge=lfs -text
-*.wasm filter=lfs diff=lfs merge=lfs -text
-*.xz filter=lfs diff=lfs merge=lfs -text
-*.zip filter=lfs diff=lfs merge=lfs -text
-*.zst filter=lfs diff=lfs merge=lfs -text
-*tfevents* filter=lfs diff=lfs merge=lfs -text
+*.7z filter=lfs diff=lfs merge=lfs -text
+*.arrow filter=lfs diff=lfs merge=lfs -text
+*.bin filter=lfs diff=lfs merge=lfs -text
+*.bz2 filter=lfs diff=lfs merge=lfs -text
+*.ckpt filter=lfs diff=lfs merge=lfs -text
+*.ftz filter=lfs diff=lfs merge=lfs -text
+*.gz filter=lfs diff=lfs merge=lfs -text
+*.h5 filter=lfs diff=lfs merge=lfs -text
+*.joblib filter=lfs diff=lfs merge=lfs -text
+*.lfs.* filter=lfs diff=lfs merge=lfs -text
+*.mlmodel filter=lfs diff=lfs merge=lfs -text
+*.model filter=lfs diff=lfs merge=lfs -text
+*.msgpack filter=lfs diff=lfs merge=lfs -text
+*.npy filter=lfs diff=lfs merge=lfs -text
+*.npz filter=lfs diff=lfs merge=lfs -text
+*.onnx filter=lfs diff=lfs merge=lfs -text
+*.ot filter=lfs diff=lfs merge=lfs -text
+*.parquet filter=lfs diff=lfs merge=lfs -text
+*.pb filter=lfs diff=lfs merge=lfs -text
+*.pickle filter=lfs diff=lfs merge=lfs -text
+*.pkl filter=lfs diff=lfs merge=lfs -text
+*.pt filter=lfs diff=lfs merge=lfs -text
+*.pth filter=lfs diff=lfs merge=lfs -text
+*.rar filter=lfs diff=lfs merge=lfs -text
+*.safetensors filter=lfs diff=lfs merge=lfs -text
+saved_model/**/* filter=lfs diff=lfs merge=lfs -text
+*.tar.* filter=lfs diff=lfs merge=lfs -text
+*.tar filter=lfs diff=lfs merge=lfs -text
+*.tflite filter=lfs diff=lfs merge=lfs -text
+*.tgz filter=lfs diff=lfs merge=lfs -text
+*.wasm filter=lfs diff=lfs merge=lfs -text
+*.xz filter=lfs diff=lfs merge=lfs -text
+*.zip filter=lfs diff=lfs merge=lfs -text
+*.zst filter=lfs diff=lfs merge=lfs -text
+*tfevents* filter=lfs diff=lfs merge=lfs -text

README.md

@@ -1,43 +1,10 @@
-# CLIP Image Embedding API
-
-Free CLIP embeddings - two deployment options.
-
-## Option 1: Hugging Face Spaces (Recommended - has working API!)
-
-1. Create a new Space at https://huggingface.co/spaces
-2. Choose "Gradio" as SDK
-3. Upload `gradio_app.py` as `app.py`
-4. Upload `requirements.txt`
-5. Your API will be at: `https://your-username-your-space.hf.space/api/predict`
-
-### API Usage (Gradio)
-```javascript
-// From Node.js backend
-const response = await axios.post('https://your-space.hf.space/api/predict', {
-  data: ['https://example.com/image.jpg']
-});
-const embedding = response.data.data[0].embedding; // 512 dimensions
-```
-
-### Environment Variable
-```
-HF_CLIP_URL=https://your-username-your-space.hf.space
-```
-
-## Option 2: Streamlit Cloud (Web UI only - no API)
-
-1. Push this folder to GitHub
-2. Go to https://share.streamlit.io
-3. Deploy `app.py`
-
-Note: Streamlit doesn't work as REST API, only for interactive web UI.
-
-## Local Testing
-```bash
-# Gradio (recommended)
-pip install -r requirements.txt
-python gradio_app.py
-
-# Streamlit
-streamlit run app.py
-```
+---
+title: alg
+emoji: 🖼️
+colorFrom: blue
+colorTo: purple
+sdk: gradio
+sdk_version: "4.44.0"
+app_file: app.py
+pinned: false
+---

app.py

@@ -1,123 +1,64 @@
 """
-CLIP Image Embedding API - Hosted on Streamlit Cloud
-Returns 512-dimensional embeddings for image similarity search
+CLIP Image Embedding API - Lightweight version for HF Spaces free tier
 """
 
-import streamlit as st
+import gradio as gr
 import torch
 from PIL import Image
 from transformers import CLIPProcessor, CLIPModel
 import requests
 from io import BytesIO
-import base64
-import json
 
-# Load model once (cached)
-@st.cache_resource
+# Use CPU and smaller memory footprint
+device = "cpu"
+model = None
+processor = None
+
 def load_model():
-    model = CLIPModel.from_pretrained("openai/clip-vit-base-patch32")
-    processor = CLIPProcessor.from_pretrained("openai/clip-vit-base-patch32")
+    global model, processor
+    if model is None:
+        model = CLIPModel.from_pretrained("openai/clip-vit-base-patch32")
+        processor = CLIPProcessor.from_pretrained("openai/clip-vit-base-patch32")
+        model.eval()
     return model, processor
 
-model, processor = load_model()
-
-def get_image_embedding(image):
-    """Get CLIP embedding for an image"""
-    inputs = processor(images=image, return_tensors="pt")
-    with torch.no_grad():
-        image_features = model.get_image_features(**inputs)
-    # Normalize
-    embedding = image_features / image_features.norm(dim=-1, keepdim=True)
-    return embedding[0].tolist()
-
-def load_image_from_source(source):
-    """Load image from URL or base64"""
-    if source.startswith('http'):
-        response = requests.get(source, timeout=30)
-        return Image.open(BytesIO(response.content)).convert('RGB')
-    elif source.startswith('data:image'):
-        # Base64 data URI
-        base64_data = source.split(',')[1]
-        image_data = base64.b64decode(base64_data)
-        return Image.open(BytesIO(image_data)).convert('RGB')
-    else:
-        raise ValueError("Invalid image source")
-
-# Streamlit UI
-st.title("🖼️ CLIP Image Embedding API")
-st.write("Get 512-dimensional embeddings for image similarity search")
-
-# API Mode - check query params
-query_params = st.query_params
-api_mode = query_params.get("api", "false") == "true"
-image_url = query_params.get("image", None)
-
-if api_mode and image_url:
-    # API mode - return JSON
+def get_embedding_from_url(image_url: str):
+    """Get CLIP embedding from image URL"""
     try:
-        image = load_image_from_source(image_url)
-        embedding = get_image_embedding(image)
-        result = {
-            "success": True,
-            "embedding": embedding,
-            "dimensions": len(embedding),
-            "model": "openai/clip-vit-base-patch32"
+        if not image_url or not image_url.startswith('http'):
+            return {"success": False, "error": "Please provide a valid image URL"}
+        
+        # Load model on first use
+        model, processor = load_model()
+        
+        # Download image
+        response = requests.get(image_url, timeout=30)
+        image = Image.open(BytesIO(response.content)).convert('RGB')
+        
+        # Get embedding
+        inputs = processor(images=image, return_tensors="pt")
+        with torch.no_grad():
+            features = model.get_image_features(**inputs)
+        
+        # Normalize
+        embedding = features / features.norm(dim=-1, keepdim=True)
+        
+        return {
+            "success": True, 
+            "embedding": embedding[0].tolist(), 
+            "dimensions": 512
         }
-        st.json(result)
     except Exception as e:
-        st.json({"success": False, "error": str(e)})
-else:
-    # Interactive UI mode
-    st.markdown("---")
-    
-    # Input options
-    tab1, tab2 = st.tabs(["🔗 Image URL", "📤 Upload Image"])
-    
-    with tab1:
-        url_input = st.text_input("Enter image URL:", placeholder="https://example.com/image.jpg")
-        if st.button("Get Embedding from URL", key="url_btn"):
-            if url_input:
-                with st.spinner("Processing..."):
-                    try:
-                        image = load_image_from_source(url_input)
-                        st.image(image, caption="Input Image", width=300)
-                        embedding = get_image_embedding(image)
-                        st.success(f"✅ Got {len(embedding)}-dimensional embedding!")
-                        st.json({
-                            "embedding": embedding[:10],  # Show first 10
-                            "dimensions": len(embedding),
-                            "note": "Showing first 10 values only"
-                        })
-                        # Full embedding in expander
-                        with st.expander("📋 Full Embedding (copy this)"):
-                            st.code(json.dumps(embedding), language="json")
-                    except Exception as e:
-                        st.error(f"Error: {e}")
-    
-    with tab2:
-        uploaded_file = st.file_uploader("Upload an image", type=['jpg', 'jpeg', 'png', 'webp'])
-        if uploaded_file:
-            image = Image.open(uploaded_file).convert('RGB')
-            st.image(image, caption="Uploaded Image", width=300)
-            if st.button("Get Embedding", key="upload_btn"):
-                with st.spinner("Processing..."):
-                    embedding = get_image_embedding(image)
-                    st.success(f"✅ Got {len(embedding)}-dimensional embedding!")
-                    with st.expander("📋 Full Embedding (copy this)"):
-                        st.code(json.dumps(embedding), language="json")
-    
-    # API Usage instructions
-    st.markdown("---")
-    st.subheader("🔌 API Usage")
-    st.code("""
-# Call from your backend:
-GET https://your-app.streamlit.app/?api=true&image=https://example.com/image.jpg
+        return {"success": False, "error": str(e)}
+
+# Gradio interface
+demo = gr.Interface(
+    fn=get_embedding_from_url,
+    inputs=gr.Textbox(label="Image URL", placeholder="https://example.com/image.jpg"),
+    outputs=gr.JSON(label="Result"),
+    title="CLIP Embedding API",
+    description="Get 512-dim CLIP embeddings for images"
+)
 
-# Response:
-{
-  "success": true,
-  "embedding": [0.0123, -0.0456, ...],
-  "dimensions": 512,
-  "model": "openai/clip-vit-base-patch32"
-}
-    """, language="python")
+if __name__ == "__main__":
+    demo.launch()

requirements.txt

@@ -1,6 +1,4 @@
-streamlit>=1.28.0
-gradio>=4.0.0
-torch>=2.0.0
-transformers>=4.35.0
-Pillow>=10.0.0
-requests>=2.31.0
+torch
+transformers
+Pillow
+requests