starlight-apk/Qwen
1
0
Fork 0
mirror of https://github.com/QwenLM/Qwen.git synced 2026-05-20 16:35:47 +08:00
Code Issues Packages Projects Releases Wiki Activity

fix no history problem

Browse Source
This commit is contained in:
wsl-wy
2023-08-07 21:01:42 +08:00
parent df25dee167
commit ad66116fe5
1 changed files with 35 additions and 21 deletions
Download Patch File Download Diff File Expand all files Collapse all files

56
web_demo.py
View File

@@ -9,12 +9,23 @@ from transformers import AutoModelForCausalLM, AutoTokenizer
from transformers.generation import GenerationConfig from transformers.generation import GenerationConfig
import sys import sys
tokenizer = AutoTokenizer.from_pretrained("Qwen/Qwen-7B-Chat", trust_remote_code=True) tokenizer = AutoTokenizer.from_pretrained(
model = AutoModelForCausalLM.from_pretrained("Qwen/Qwen-7B-Chat", device_map="auto", trust_remote_code=True).eval() "Qwen/Qwen-7B-Chat", trust_remote_code=True, resume_download=True
model.generation_config = GenerationConfig.from_pretrained("Qwen/Qwen-7B-Chat", trust_remote_code=True) )
model = AutoModelForCausalLM.from_pretrained(
"Qwen/Qwen-7B-Chat",
device_map="auto",
offload_folder="offload",
trust_remote_code=True,
resume_download=True,
).eval()
model.generation_config = GenerationConfig.from_pretrained(
"Qwen/Qwen-7B-Chat", trust_remote_code=True, resume_download=True
)
if len(sys.argv) > 1 and sys.argv[1] == "--exit": if len(sys.argv) > 1 and sys.argv[1] == "--exit":
exit(0) sys.exit(0)
def postprocess(self, y): def postprocess(self, y):
if y is None: if y is None:
@@ -58,21 +69,26 @@ def parse_text(text):
line = line.replace("(", "(") line = line.replace("(", "(")
line = line.replace(")", ")") line = line.replace(")", ")")
line = line.replace("$", "$") line = line.replace("$", "$")
lines[i] = "<br>"+line lines[i] = "<br>" + line
text = "".join(lines) text = "".join(lines)
return text return text
def predict(input, chatbot, history, past_key_values):
task_history = []
def predict(input, chatbot):
print('Q: ' + parse_text(input)) print('Q: ' + parse_text(input))
chatbot.append((parse_text(input), "")) chatbot.append((parse_text(input), ""))
fullResponse = ""; fullResponse = ""
for response in model.chat(tokenizer, input, history=history, stream=True): for response in model.chat(tokenizer, input, history=task_history, stream=True):
chatbot[-1] = (parse_text(input), parse_text(response)) chatbot[-1] = (parse_text(input), parse_text(response))
yield chatbot, history, past_key_values yield chatbot
fullResponse = parse_text(response); fullResponse = parse_text(response)
task_history.append((input, fullResponse))
print("A: " + parse_text(fullResponse)) print("A: " + parse_text(fullResponse))
@@ -81,7 +97,8 @@ def reset_user_input():
def reset_state(): def reset_state():
return [], [], None task_history = []
return []
with gr.Blocks() as demo: with gr.Blocks() as demo:
@@ -91,19 +108,16 @@ with gr.Blocks() as demo:
with gr.Row(): with gr.Row():
with gr.Column(scale=4): with gr.Column(scale=4):
with gr.Column(scale=12): with gr.Column(scale=12):
user_input = gr.Textbox(show_label=False, placeholder="Input...", lines=10).style( query = gr.Textbox(
container=False) show_label=False, placeholder="Input...", lines=10
).style(container=False)
with gr.Column(min_width=32, scale=1): with gr.Column(min_width=32, scale=1):
submitBtn = gr.Button("Submit", variant="primary") submitBtn = gr.Button("Submit", variant="primary")
with gr.Column(scale=1): with gr.Column(scale=1):
emptyBtn = gr.Button("Clear History") emptyBtn = gr.Button("Clear History")
history = gr.State([]) submitBtn.click(predict, [query, chatbot], [chatbot], show_progress=True)
past_key_values = gr.State(None) submitBtn.click(reset_user_input, [], [query])
emptyBtn.click(reset_state, outputs=[chatbot], show_progress=True)
submitBtn.click(predict, [user_input, chatbot, history, past_key_values],
[chatbot, history, past_key_values], show_progress=True)
submitBtn.click(reset_user_input, [], [user_input])
emptyBtn.click(reset_state, outputs=[chatbot, history, past_key_values], show_progress=True)
demo.queue().launch(share=False, inbrowser=True, server_port=80, server_name="0.0.0.0") demo.queue().launch(share=False, inbrowser=True, server_port=80, server_name="0.0.0.0")