Update app.py

app.py

@@ -1,18 +1,54 @@
 import gradio as gr
-from transformers import pipeline
-
-def chat_with_model(user_input):
-    messages = [{"role": "user", "content": user_input}]
-    pipe = pipeline("text-generation", model="deepseek-ai/DeepSeek-R1", trust_remote_code=True)
-    response = pipe(messages)
-    return response[0]['generated_text']
-
-iface = gr.Interface(
-    fn=chat_with_model,
-    inputs=gr.Textbox(label="Enter your message"),
-    outputs=gr.Textbox(label="Model Response"),
-    title="Chat with DeepSeek-R1",
-    description="Enter a message to interact with the DeepSeek-R1 model."
-)
-
-iface.launch()
+from transformers import AutoModelForCausalLM, AutoTokenizer
+
+# Загрузка модели и токенизатора
+model_name = "deepseek/r1"
+model = AutoModelForCausalLM.from_pretrained(model_name)
+tokenizer = AutoTokenizer.from_pretrained(model_name)
+
+def inference(prompt, temperature, top_p, max_length):
+    try:
+        # Токенизация входного текста
+        inputs = tokenizer(prompt, return_tensors="np")
+
+        # Генерация ответа
+        outputs = model.generate(
+            inputs.input_ids,
+            max_length=max_length,
+            temperature=temperature,
+            top_p=top_p,
+            do_sample=True
+        )
+
+        # Декодирование ответа
+        response = tokenizer.decode(outputs[0], skip_special_tokens=True)
+        return response
+
+    except Exception as e:
+        return f"Ошибка: {str(e)}"
+
+# Создание интерфейса Gradio
+with gr.Blocks() as demo:
+    gr.Markdown("# DeepSeek-r1 Model")
+
+    with gr.Row():
+        input_text = gr.Textbox(label="Входной текст", placeholder="Введите ваш текст здесь...")
+        output_text = gr.Textbox(label="Выходной текст", placeholder="Ответ модели появится здесь...")
+
+    with gr.Row():
+        max_length = gr.Slider(label="Max Length", minimum=1, maximum=1000, value=500)
+        temperature = gr.Slider(label="Temperature", minimum=0.1, maximum=2.0, value=1.0)
+        top_p = gr.Slider(label="Top P", minimum=0.1, maximum=1.0, value=1.0)
+
+    run_button = gr.Button("Run")
+
+    # Подключение функции к кнопке
+    run_button.click(
+        fn=inference,
+        inputs=[input_text, temperature, top_p, max_length],
+        outputs=output_text
+    )
+
+# Запуск приложения
+if __name__ == "__main__":
+    demo.launch()