Update app.py

app.py

@@ -0,0 +1,126 @@
+# app.py
+import gradio as gr
+from sentence_transformers import SentenceTransformer
+import numpy as np
+import os
+from pathlib import Path
+import tempfile
+
+# Initialize the embedding model
+model = SentenceTransformer('all-MiniLM-L6-v2')
+
+# In-memory storage for documents and embeddings
+documents = []
+embeddings = []
+file_names = []
+
+def process_file(file):
+    """Process uploaded file and store its embedding"""
+    if file is None:
+        return "❌ No file uploaded"
+    
+    try:
+        # Read file content
+        with open(file.name, 'r', encoding='utf-8') as f:
+            content = f.read()
+        
+        # Generate embedding
+        embedding = model.encode(content)
+        
+        # Store document
+        documents.append(content)
+        embeddings.append(embedding)
+        file_names.append(os.path.basename(file.name))
+        
+        return f"✅ Successfully processed: {os.path.basename(file.name)}\nTotal documents: {len(documents)}"
+    
+    except Exception as e:
+        return f"❌ Error processing file: {str(e)}"
+
+def semantic_search(query, top_k=3):
+    """Perform semantic search across uploaded documents"""
+    if not query:
+        return "⚠️ Please enter a search query"
+    
+    if not documents:
+        return "⚠️ No documents uploaded yet. Please upload some files first."
+    
+    try:
+        # Generate query embedding
+        query_embedding = model.encode(query)
+        
+        # Calculate cosine similarities
+        similarities = []
+        for i, doc_embedding in enumerate(embeddings):
+            similarity = np.dot(query_embedding, doc_embedding) / (
+                np.linalg.norm(query_embedding) * np.linalg.norm(doc_embedding)
+            )
+            similarities.append((similarity, i))
+        
+        # Sort by similarity (descending)
+        similarities.sort(reverse=True)
+        
+        # Build results
+        results = []
+        for score, idx in similarities[:top_k]:
+            doc_content = documents[idx][:500]  # Show first 500 chars
+            results.append(f"**File:** {file_names[idx]}\n"
+                          f"**Similarity Score:** {score:.3f}\n"
+                          f"**Content Preview:**\n{doc_content}...\n")
+        
+        return "\n---\n".join(results)
+    
+    except Exception as e:
+        return f"❌ Search error: {str(e)}"
+
+def clear_documents():
+    """Clear all uploaded documents"""
+    documents.clear()
+    embeddings.clear()
+    file_names.clear()
+    return "🗑️ All documents cleared"
+
+# Create the Gradio interface
+with gr.Blocks(title="AI Semantic File Search", theme=gr.themes.Soft()) as app:
+    gr.Markdown("# 🔍 AI Semantic File Search")
+    gr.Markdown("Upload documents and search through them using AI-powered semantic search!")
+    
+    with gr.Row():
+        with gr.Column(scale=1):
+            gr.Markdown("### Upload Documents")
+            file_input = gr.File(label="Upload Text File", file_types=[".txt", ".md", ".py", ".json"])
+            process_btn = gr.Button("📤 Process File", variant="primary")
+            status_output = gr.Textbox(label="Status", interactive=False, lines=2)
+            clear_btn = gr.Button("🗑️ Clear All", variant="secondary")
+        
+        with gr.Column(scale=2):
+            gr.Markdown("### Search Documents")
+            query_input = gr.Textbox(
+                label="Search Query",
+                placeholder="Enter your search query...",
+                lines=2
+            )
+            top_k_slider = gr.Slider(1, 10, value=3, step=1, label="Number of Results")
+            search_btn = gr.Button("🔎 Search", variant="primary")
+            results_output = gr.Markdown(label="Search Results")
+    
+    # Event handlers
+    process_btn.click(
+        fn=process_file,
+        inputs=[file_input],
+        outputs=[status_output]
+    )
+    
+    search_btn.click(
+        fn=semantic_search,
+        inputs=[query_input, top_k_slider],
+        outputs=[results_output]
+    )
+    
+    clear_btn.click(
+        fn=clear_documents,
+        outputs=[status_output]
+    )
+
+if __name__ == "__main__":
+    app.launch()