new

app.py

@@ -1,47 +1,198 @@
-import gradio as gr
+#!/usr/bin/env python3
+
+import os
+import sys
+import asyncio
+import warnings
+import signal
 import torch
 from transformers import AutoTokenizer, AutoModelForCausalLM
+import gradio as gr
 
-MODEL_ID = "abdelac/Mistral_Test"   # your model repo
+# =================== ASYNCIO FIX ===================
+# Fix for the asyncio cleanup error
+if sys.version_info >= (3, 8) and sys.platform.startswith('linux'):
+    # This prevents the error on Linux with Python 3.8+
+    try:
+        import uvloop
+        uvloop.install()
+    except ImportError:
+        pass
 
-print("🔁 Loading tokenizer...")
-tokenizer = AutoTokenizer.from_pretrained(MODEL_ID)
+# Suppress warnings
+warnings.filterwarnings("ignore")
+os.environ["PYTHONWARNINGS"] = "ignore"
 
-print("🔁 Loading model...")
-model = AutoModelForCausalLM.from_pretrained(
-    MODEL_ID,
-    torch_dtype=torch.float16 if torch.cuda.is_available() else torch.float32,
-    device_map="auto"
-)
+# =================== MODEL LOADING ===================
+@gr.cache_resource
+def load_model():
+    """Load the TinyLlama model"""
+    print("🚀 Loading TinyLlama model...")
+    MODEL_ID = "abdelac/tinyllama"
+    
+    tokenizer = AutoTokenizer.from_pretrained(MODEL_ID)
+    model = AutoModelForCausalLM.from_pretrained(
+        MODEL_ID,
+        torch_dtype=torch.float32,
+        device_map="auto",
+        low_cpu_mem_usage=True
+    )
+    
+    print("✅ Model loaded successfully!")
+    return tokenizer, model
 
-model.eval()
-print("✅ Model loaded")
+# Load model once
+tokenizer, model = load_model()
 
-# ---------- Inference function ----------
-def generate(prompt):
-    inputs = tokenizer(prompt, return_tensors="pt").to(model.device)
+# =================== GENERATION FUNCTION ===================
+def generate_text(prompt, max_tokens=150, temperature=0.7):
+    """Generate text based on prompt"""
+    try:
+        # Tokenize
+        inputs = tokenizer(prompt, return_tensors="pt")
+        
+        # Generate
+        with torch.no_grad():
+            outputs = model.generate(
+                **inputs,
+                max_new_tokens=max_tokens,
+                temperature=temperature,
+                do_sample=True,
+                pad_token_id=tokenizer.eos_token_id
+            )
+        
+        # Decode
+        result = tokenizer.decode(outputs[0], skip_special_tokens=True)
+        return result
+        
+    except Exception as e:
+        return f"❌ Error: {str(e)}"
 
-    with torch.no_grad():
-        output_ids = model.generate(
-            **inputs,
-            max_new_tokens=128,
-            temperature=0.7,
-            do_sample=True
+# =================== GRADIO INTERFACE ===================
+def create_interface():
+    """Create the Gradio interface"""
+    with gr.Blocks(
+        title="🦙 TinyLlama Demo",
+        theme=gr.themes.Soft(),
+        css=".gradio-container {max-width: 800px !important}"
+    ) as demo:
+        
+        gr.Markdown("""
+        # 🦙 TinyLlama Text Generator
+        
+        Generate text using the TinyLlama model (1.1B parameters)
+        
+        **Model**: [abdelac/tinyllama](https://huggingface.co/abdelac/tinyllama)
+        """)
+        
+        with gr.Row():
+            with gr.Column(scale=2):
+                prompt = gr.Textbox(
+                    label="📝 Input Prompt",
+                    placeholder="Type your text here...",
+                    lines=5,
+                    value="Once upon a time in a magical forest,"
+                )
+                
+                with gr.Row():
+                    max_tokens = gr.Slider(
+                        50, 500, value=150,
+                        label="📏 Max Tokens",
+                        info="Maximum length of generated text"
+                    )
+                    temperature = gr.Slider(
+                        0.1, 2.0, value=0.7,
+                        label="🌡️ Temperature",
+                        info="Higher = more creative, Lower = more focused"
+                    )
+                
+                with gr.Row():
+                    generate_btn = gr.Button(
+                        "✨ Generate",
+                        variant="primary",
+                        size="lg"
+                    )
+                    clear_btn = gr.Button(
+                        "🗑️ Clear",
+                        variant="secondary"
+                    )
+            
+            with gr.Column(scale=3):
+                output = gr.Textbox(
+                    label="📄 Generated Text",
+                    lines=12,
+                    interactive=False
+                )
+        
+        # Examples
+        gr.Examples(
+            examples=[
+                ["Write a short story about a robot learning to paint"],
+                ["Explain quantum computing in simple terms"],
+                ["Python function to calculate fibonacci sequence:"],
+                ["The benefits of renewable energy include"],
+                ["Write a poem about artificial intelligence"]
+            ],
+            inputs=prompt,
+            label="💡 Try these examples"
         )
+        
+        # Functions
+        generate_btn.click(
+            fn=generate_text,
+            inputs=[prompt, max_tokens, temperature],
+            outputs=output,
+            api_name="generate"
+        )
+        
+        clear_btn.click(
+            fn=lambda: ("", ""),
+            inputs=[],
+            outputs=[prompt, output]
+        )
+        
+        # Status
+        gr.Markdown("---")
+        gr.Markdown("""
+        <div style='text-align: center; color: #666; font-size: 0.9em;'>
+        ✅ Model loaded successfully | 🚀 Ready to generate text
+        </div>
+        """)
+    
+    return demo
 
-    return tokenizer.decode(output_ids[0], skip_special_tokens=True)
-
-# ---------- Gradio UI ----------
-demo = gr.Interface(
-    fn=generate,
-    inputs=gr.Textbox(lines=4, label="Prompt"),
-    outputs=gr.Textbox(lines=8, label="Output"),
-    title="Mistral Test – Space Inference"
-)
-
-demo.launch(
-    server_name="0.0.0.0",
-    server_port=7860,
-    ssr_mode=False,
-)
+# =================== MAIN ENTRY POINT ===================
+def main():
+    """Main function with proper cleanup"""
+    demo = create_interface()
+    
+    # Clean launch configuration
+    try:
+        demo.launch(
+            server_name="0.0.0.0",
+            server_port=7860,
+            share=False,
+            quiet=True,          # Reduce console output
+            debug=False,         # Disable debug mode
+            show_error=True,     # Show errors in UI
+            favicon_path=None,
+            ssl_verify=True,
+            max_file_size="2MB",
+            allowed_paths=["./"],
+            blocked_paths=[],
+            show_api=True
+        )
+    except KeyboardInterrupt:
+        print("\n👋 Shutting down gracefully...")
+        sys.exit(0)
+    except Exception as e:
+        print(f"❌ Error: {e}")
+        sys.exit(1)
 
+if __name__ == "__main__":
+    # Set up signal handlers for clean shutdown
+    signal.signal(signal.SIGINT, lambda s, f: sys.exit(0))
+    signal.signal(signal.SIGTERM, lambda s, f: sys.exit(0))
+    
+    # Run the app
+    main()