Update space, changed language.

app.py

@@ -1,80 +1,58 @@
 import gradio as gr
 from transformers import AutoModelForCausalLM, AutoTokenizer
 import torch
-import os
 
-# ЗАМЕНИТЕ ЭТИ ЗНАЧЕНИЯ НА СВОИ!
-SPACE_NAME = "livecoder"  # Имя вашего Space
-USERNAME = "sarekuwa"  # Ваш username на Hugging Face
-
-# Выводим endpoint сразу при запуске
-print("=" * 50)
-print(f"Space name: {SPACE_NAME}")
-print(f"Username: {USERNAME}")
-print(f"Local URL: http://0.0.0.0:7860")
-print(f"Public API endpoint: https://{USERNAME}-{SPACE_NAME}.hf.space/run/predict")
-print("=" * 50)
-
-# Загрузка модели
+# Load model and tokenizer
 model_name = "microsoft/phi-2"
-print("Loading tokenizer...")
 tokenizer = AutoTokenizer.from_pretrained(model_name, trust_remote_code=True)
-if tokenizer.pad_token is None:
-    tokenizer.pad_token = tokenizer.eos_token
+tokenizer.pad_token = tokenizer.eos_token
 
-print("Loading model...")
 model = AutoModelForCausalLM.from_pretrained(
     model_name,
     torch_dtype=torch.float32,
     device_map="cpu",
     trust_remote_code=True
 )
-print("Model loaded successfully!")
 
-def predict(message, history):
-    """Обработчик сообщений"""
-    try:
-        # Формируем промпт
-        prompt = ""
-        if history:
-            for human_msg, ai_msg in history:
-                prompt += f"### Human: {human_msg}\n### Assistant: {ai_msg}\n"
-        prompt += f"### Human: {message}\n### Assistant:"
-        
-        # Токенизация и генерация
-        inputs = tokenizer(prompt, return_tensors="pt", truncation=True, max_length=1024)
-        
-        with torch.no_grad():
-            outputs = model.generate(
-                inputs.input_ids,
-                max_new_tokens=256,
-                temperature=0.7,
-                do_sample=True,
-                top_p=0.9,
-                pad_token_id=tokenizer.pad_token_id,
-                eos_token_id=tokenizer.eos_token_id
-            )
-        
-        response = tokenizer.decode(outputs[0][inputs.input_ids.shape[1]:], skip_special_tokens=True)
-        response = response.split("### Human:")[0].strip()
-        return response
-        
-    except Exception as e:
-        return f"Ошибка: {str(e)}"
+def generate_response(message):
+    """API function - takes text, returns response"""
+    prompt = f"### Instruction: {message}\n### Response:"
+    
+    inputs = tokenizer(prompt, return_tensors="pt", truncation=True, max_length=512)
+    attention_mask = inputs.get('attention_mask', None)
+    
+    with torch.no_grad():
+        outputs = model.generate(
+            inputs.input_ids,
+            attention_mask=attention_mask,
+            max_new_tokens=256,
+            temperature=0.7,
+            do_sample=True,
+            top_p=0.9,
+            pad_token_id=tokenizer.pad_token_id,
+            eos_token_id=tokenizer.eos_token_id
+        )
+    
+    response = tokenizer.decode(outputs[0][inputs.input_ids.shape[1]:], skip_special_tokens=True)
+    return response.strip()
 
-# Исправляем предупреждение: добавляем type='messages'
-demo = gr.ChatInterface(
-    fn=predict,
+# Create Gradio Interface
+interface = gr.Interface(
+    fn=generate_response,
+    inputs=gr.Textbox(
+        label="Programming question", 
+        placeholder="Example: Write a binary search function in C++...", 
+        lines=3
+    ),
+    outputs=gr.Textbox(label="Model response", lines=10),
     title="LiveCoder LLM API",
-    description="Модель Phi-2 для помощи в написании кода",
-    chatbot=gr.Chatbot(height=400, type='messages'),  # ИСПРАВЛЕНО ЗДЕСЬ
-    textbox=gr.Textbox(placeholder="Введите вопрос по программированию...", lines=3),
+    description="Phi-2 model for programming assistance",
     examples=[
-        ["Напиши hello world на C++"],
-        ["Объясни принцип ООП"],
-        ["Как работает указатель в C++?"]
+        ["Write a hello world program in C++"],
+        ["Explain the OOP principle"],
+        ["How does a pointer work in C++?"]
     ]
 )
 
-if __name__ == "__main__":
-    demo.launch(server_name="0.0.0.0", server_port=7860, share=False)
+# Launch the application
+interface.launch(server_name="0.0.0.0", server_port=7860, share=False)