creating app.py

app.py

@@ -0,0 +1,126 @@
+import gradio as gr
+from sentence_transformers import SentenceTransformer
+import numpy as np
+from typing import List, Union
+import spaces  # ZeroGPU decorator
+
+# Load model once at startup
+MODEL_NAME = "sentence-transformers/all-MiniLM-L6-v2"
+model = SentenceTransformer(MODEL_NAME)
+
+@spaces.GPU(duration=60)  # ZeroGPU: allocate GPU for 60 seconds
+def generate_embeddings(texts: Union[str, List[str]]) -> List[List[float]]:
+    """
+    Generate embeddings for text(s)
+    
+    Args:
+        texts: Single string or list of strings
+        
+    Returns:
+        List of embedding vectors (384 dimensions)
+    """
+    # Handle single string
+    if isinstance(texts, str):
+        texts = [texts]
+    
+    # Generate embeddings
+    embeddings = model.encode(
+        texts,
+        convert_to_numpy=True,
+        normalize_embeddings=True,  # Normalize for cosine similarity
+        show_progress_bar=False
+    )
+    
+    # Convert to list for JSON serialization
+    return embeddings.tolist()
+
+def batch_generate(texts_input: str) -> str:
+    """
+    Gradio interface for batch embedding generation
+    Expects newline-separated texts
+    """
+    if not texts_input.strip():
+        return "Error: Please provide at least one text"
+    
+    texts = [t.strip() for t in texts_input.split('\n') if t.strip()]
+    
+    try:
+        embeddings = generate_embeddings(texts)
+        
+        result = f"Generated {len(embeddings)} embeddings\n"
+        result += f"Dimensions: {len(embeddings[0])}\n\n"
+        result += "First embedding preview:\n"
+        result += str(embeddings[0][:10]) + "...\n"
+        
+        return result
+    except Exception as e:
+        return f"Error: {str(e)}"
+
+# Create Gradio interface
+with gr.Blocks(title="FairFate Embeddings API") as demo:
+    gr.Markdown("""
+    # FairFate Embeddings API
+    
+    Generate semantic embeddings using sentence-transformers/all-MiniLM-L6-v2
+    - **Dimensions:** 384
+    - **Max Tokens:** 256
+    - **Normalized:** Yes (ready for cosine similarity)
+    
+    ## API Usage
+    
+    **Endpoint:** `https://your-username-fairfate-embeddings.hf.space/api/predict`
+    
+    **Request:**
+    ```json
+    {
+      "data": ["Your text here", "Another text"]
+    }
+    ```
+    
+    **Response:**
+    ```json
+    {
+      "data": [[[0.123, -0.456, ...]], [[0.789, -0.012, ...]]]
+    }
+    ```
+    """)
+    
+    with gr.Tab("Test Embeddings"):
+        input_text = gr.Textbox(
+            label="Input Texts (one per line)",
+            placeholder="Enter texts, one per line:\nDungeons and Dragons adventure\nSci-fi space opera campaign",
+            lines=5
+        )
+        output_text = gr.Textbox(label="Results", lines=10)
+        submit_btn = gr.Button("Generate Embeddings")
+        submit_btn.click(batch_generate, inputs=input_text, outputs=output_text)
+    
+    with gr.Tab("API Documentation"):
+        gr.Markdown("""
+        ### Python Example
+        ```python
+        import requests
+        
+        response = requests.post(
+            "https://your-username-fairfate-embeddings.hf.space/api/predict",
+            json={"data": ["Your text here"]}
+        )
+        embeddings = response.json()["data"][0]
+        ```
+        
+        ### JavaScript/TypeScript Example
+        ```typescript
+        const response = await fetch(
+            'https://your-username-fairfate-embeddings.hf.space/api/predict',
+            {
+                method: 'POST',
+                headers: { 'Content-Type': 'application/json' },
+                body: JSON.stringify({ data: ["Your text here"] })
+            }
+        );
+        const result = await response.json();
+        const embeddings = result.data[0];
+        ```
+        """)
+# Launch with API enabled
+demo.launch()