Update app.py

app.py

@@ -1,64 +1,67 @@
 import gradio as gr
-from huggingface_hub import InferenceClient
+from transformers import AutoModelForCausalLM, AutoTokenizer
+import torch
 
-"""
-For more information on `huggingface_hub` Inference API support, please check the docs: https://huggingface.co/docs/huggingface_hub/v0.22.2/en/guides/inference
-"""
-client = InferenceClient("ewhk9887/deepseek-cokoder")
+# 모델 로드
+MODEL_NAME = "ewhk9887/deepseek-cokoder"
 
+device = "cuda" if torch.cuda.is_available() else "cpu"
+model = AutoModelForCausalLM.from_pretrained(MODEL_NAME, torch_dtype=torch.float16 if device == "cuda" else torch.float32)
+model.to(device)
+tokenizer = AutoTokenizer.from_pretrained(MODEL_NAME)
 
-def respond(
-    message,
-    history: list[tuple[str, str]],
-    system_message,
-    max_tokens,
-    temperature,
-    top_p,
-):
-    messages = [{"role": "system", "content": system_message}]
-
-    for val in history:
-        if val[0]:
-            messages.append({"role": "user", "content": val[0]})
-        if val[1]:
-            messages.append({"role": "assistant", "content": val[1]})
-
-    messages.append({"role": "user", "content": message})
-
-    response = ""
-
-    for message in client.chat_completion(
-        messages,
-        max_tokens=max_tokens,
-        stream=True,
+# 코드 리뷰 생성 함수
+def generate_code_review(user_input, system_message, max_tokens, temperature, top_p):
+    """
+    사용자 입력과 시스템 메시지를 기반으로 AI 코드 리뷰 생성.
+    """
+    # 메시지 포맷 생성
+    prompt = f"{system_message}\n\nCode:\n{user_input}\n\nReview:"
+    inputs = tokenizer(prompt, return_tensors="pt").to(device)
+    
+    # 모델 출력 생성
+    outputs = model.generate(
+        inputs.input_ids,
+        max_new_tokens=max_tokens,
         temperature=temperature,
         top_p=top_p,
-    ):
-        token = message.choices[0].delta.content
+        repetition_penalty=1.1,
+        pad_token_id=tokenizer.eos_token_id
+    )
+    review = tokenizer.decode(outputs[0], skip_special_tokens=True)
+    return review
+
+# Gradio 인터페이스 생성
+with gr.Blocks() as demo:
+    gr.Markdown("# DeepSeek Code Review Assistant")
+    gr.Markdown("AI가 코드에 대한 리뷰를 제공합니다. 코드를 입력하고 리뷰를 확인하세요!")
 
-        response += token
-        yield response
+    with gr.Row():
+        with gr.Column():
+            code_input = gr.Textbox(label="코드 입력", placeholder="리뷰를 원하는 코드를 입력하세요...", lines=10)
+            system_message = gr.Textbox(
+                label="시스템 메시지",
+                value="You are a helpful assistant providing detailed and constructive code reviews.",
+                lines=3,
+            )
+        with gr.Column():
+            review_output = gr.Textbox(label="코드 리뷰 결과", lines=10)
 
+    # 추가 옵션
+    max_tokens = gr.Slider(label="Max Tokens", minimum=10, maximum=512, value=256, step=10)
+    temperature = gr.Slider(label="Temperature", minimum=0.1, maximum=1.0, value=0.7, step=0.1)
+    top_p = gr.Slider(label="Top-p", minimum=0.1, maximum=1.0, value=0.9, step=0.05)
 
-"""
-For information on how to customize the ChatInterface, peruse the gradio docs: https://www.gradio.app/docs/chatinterface
-"""
-demo = gr.ChatInterface(
-    respond,
-    additional_inputs=[
-        gr.Textbox(value="You are a code review assistant. Provide helpful feedback on the given code.", label="System message"),
-        gr.Slider(minimum=1, maximum=2048, value=512, step=1, label="Max new tokens"),
-        gr.Slider(minimum=0.1, maximum=4.0, value=0.7, step=0.1, label="Temperature"),
-        gr.Slider(
-            minimum=0.1,
-            maximum=1.0,
-            value=0.95,
-            step=0.05,
-            label="Top-p (nucleus sampling)",
-        ),
-    ],
-)
+    # 버튼
+    generate_button = gr.Button("리뷰 생성")
 
+    # 이벤트 연결
+    generate_button.click(
+        fn=generate_code_review,
+        inputs=[code_input, system_message, max_tokens, temperature, top_p],
+        outputs=review_output,
+    )
 
+# 실행
 if __name__ == "__main__":
     demo.launch()