Update app.py

app.py

@@ -1,63 +1,67 @@
 import gradio as gr
-from huggingface_hub import InferenceClient
-
-"""
-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("HuggingFaceH4/zephyr-7b-beta")
-
-
-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,
-        temperature=temperature,
-        top_p=top_p,
-    ):
-        token = message.choices[0].delta.content
-
-        response += token
-        yield response
-
-"""
-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 friendly Chatbot.", 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)",
-        ),
+import torch
+from transformers import AutoModelForCausalLM, AutoTokenizer
+
+class MedS_Llama3:
+    def __init__(self, model_path: str):
+        # 加载模型到CPU
+        self.model = AutoModelForCausalLM.from_pretrained(
+            model_path,
+            device_map='cpu',  # 指定加载到CPU
+            torch_dtype=torch.float32  # 使用标准的float32精度
+        )
+        self.model.config.pad_token_id = self.model.config.eos_token_id = 128009
+        
+        self.tokenizer = AutoTokenizer.from_pretrained(
+            model_path,
+            model_max_length=2048,
+            padding_side="right"
+        )
+        self.tokenizer.pad_token = self.tokenizer.eos_token
+        self.model.eval()
+        print('Model and tokenizer loaded on CPU!')
+
+    def chat(self, query: str, instruction: str = "If you are a doctor, please perform clinical consulting with the patient.") -> str:
+        input_sentence = f"{instruction}\n\n{query}"
+        input_tokens = self.tokenizer(
+            input_sentence,
+            return_tensors="pt",
+            padding=True,
+            truncation=True
+        )
+
+        output = self.model.generate(
+            **input_tokens,
+            max_new_tokens=512,  # 降低生成的最大新tokens数目来节省内存
+            eos_token_id=128009
+        )
+
+        generated_text = self.tokenizer.decode(
+            output[0][input_tokens['input_ids'].shape[1]:], 
+            skip_special_tokens=True
+        )
+
+        return generated_text.strip()
+
+# 实例化模型
+model_path = "Henrychur/MMedS-Llama-3-8B"  # 确保这里是模型的正确路径
+chat_model = MedS_Llama3(model_path)
+
+# 定义 Gradio 接口中使用的响应函数
+def respond(message, system_message):
+    # 每次对话结束后清空历史，只使用当前输入和系统指令
+    response = chat_model.chat(query=message, instruction=system_message)
+    yield response
+
+# 设置 Gradio 聊天界面
+demo = gr.Interface(
+    fn=respond,
+    inputs=[
+        gr.Textbox(label="What is the treatment for diabetes?"),
+        gr.Textbox(value="If you are a doctor, please perform clinical consulting with the patient.", label="System message")
     ],
+    outputs="text"
 )
 
-
 if __name__ == "__main__":
-    demo.launch()
+    demo.launch()