for the class

app.py

@@ -0,0 +1,44 @@
+import os
+from datetime import datetime
+import uuid
+import gradio as gr
+from transformers import AutoTokenizer, AutoModelForCausalLM
+import torch
+
+# Load model and tokenizer
+model_name = "meta-llama/Llama-2-7b-chat-hf"
+tokenizer = AutoTokenizer.from_pretrained(model_name)
+model = AutoModelForCausalLM.from_pretrained(model_name, torch_dtype=torch.float16, device_map="auto")
+
+def chat_with_model(messages):
+    input_ids = tokenizer.encode(str(messages), return_tensors="pt").to(model.device)
+    output = model.generate(input_ids, max_length=1000, num_return_sequences=1, temperature=0.7)
+    response = tokenizer.decode(output[0], skip_special_tokens=True)
+    return response
+
+def chat_with_model_gradio(message, history, session_id):
+    messages = [
+        {"role": "system", "content": f"너의 이름은 ChatMBTI. 사람들의 MBTI유형에 알맞은 상담을 진행할 수 있어. 상대방의 MBTI 유형을 먼저 물어보고, 그 유형에 알맞게 상담을 진행해줘. 참고로 현재 시각은 {datetime.now().strftime('%Y-%m-%d %H:%M:%S')}이야."},
+    ]
+    messages.extend([{"role": "user" if i % 2 == 0 else "assistant", "content": m} for i, m in enumerate(history)])
+    messages.append({"role": "user", "content": message})
+
+    response = chat_with_model(messages)
+    history.append((message, response))
+
+    return "", history
+
+def main():
+    session_id = str(uuid.uuid4())
+    with gr.Blocks() as demo:
+        chatbot = gr.Chatbot(label="ChatMBTI")
+        msg = gr.Textbox(label="메시지를 입력하세요")
+        clear = gr.Button("대화 초기화")
+
+        msg.submit(chat_with_model_gradio, [msg, chatbot, gr.State(session_id)], [msg, chatbot])
+        clear.click(lambda: None, None, chatbot, queue=False)
+
+    demo.launch()
+
+if __name__ == "__main__":
+    main()

requirements.txt

@@ -0,0 +1,2 @@
+transformers
+torch