ChatGLM-6B/web_demo.py

57 lines
2.2 KiB
Python

from transformers import AutoModel, AutoTokenizer
import gradio as gr
tokenizer = AutoTokenizer.from_pretrained("THUDM/chatglm-6b", trust_remote_code=True)
model = AutoModel.from_pretrained("THUDM/chatglm-6b", trust_remote_code=True).half().cuda()
model = model.eval()
MAX_TURNS = 20
MAX_BOXES = MAX_TURNS * 2
def btn_is_clickable(txt):
if txt is not None and txt.strip() != '':
return gr.update(interactive=True)
else:
return gr.update(interactive=False)
def user_message(user_message, history):
return "", history + [[user_message, None]]
def predict(input, max_length, top_p, temperature, history):
if len(history) > MAX_TURNS:
history = history[-20:]
for response, const_history in model.stream_chat(tokenizer, input, history, max_length=max_length, top_p=top_p,
temperature=temperature):
history[-1][1] = response
yield history
with gr.Blocks() as demo:
with gr.Row():
with gr.Column(scale=1):
gr.Markdown("https://github.com/THUDM/ChatGLM-6B")
max_length = gr.Slider(0, 4096, value=2048, step=1.0, label="Maximum length", interactive=True)
top_p = gr.Slider(0, 1, value=0.7, step=0.01, label="Top P", interactive=True)
temperature = gr.Slider(0, 1, value=0.95, step=0.01, label="Temperature", interactive=True)
with gr.Column(scale=4):
chatbot = gr.Chatbot([], elem_id="chatbot").style(height=750)
with gr.Row():
with gr.Column(scale=4):
txt = gr.Textbox(
show_label=False,
placeholder="有问题就会有答案"
).style(container=False)
with gr.Column(scale=1, min_width=0):
btn = gr.Button("发送", interactive=False)
# 控制按钮是否可以点击
txt.change(btn_is_clickable, txt, btn)
# 发送消息
btn.click(user_message, [txt, chatbot], [txt, chatbot], queue=False).then(
predict, [txt, max_length, top_p, temperature, chatbot], chatbot
)
demo.queue().launch(share=False, inbrowser=True)