Skip to content
New issue

Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.

By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.

Already on GitHub? Sign in to your account

trt llm qwen #4

Open
wants to merge 6 commits into
base: main
Choose a base branch
from
Open
Changes from 1 commit
Commits
File filter

Filter by extension

Filter by extension

Conversations
Failed to load comments.
Loading
Jump to
Jump to file
Failed to load files.
Loading
Diff view
Diff view
Next Next commit
1
xiangxinhello committed Jul 29, 2024
commit 5f18faed5c50116070140192df07f0ed7dff1812
41 changes: 34 additions & 7 deletions demo_QAMaking_from_pdf.py
Original file line number Diff line number Diff line change
@@ -2,7 +2,7 @@
from gradio_pdf import PDF
import fitz
from chat import ChatModel

import time

def _launch_demo(model, template):
pdftxt = []
@@ -62,13 +62,17 @@ def makeQA(_chatbot, task_history):

message = {"role": "user", "content": chat_query}
respones = ""
start_time = time.time()
for new_text in model.stream_chat([message]):
respones += new_text
#print(respones)
_chatbot[-1] = (chat_query, respones)
#print(_chatbot)
#_chatbot += (p, new_text)
yield _chatbot
end_time = time.time()
execution_time = end_time - start_time
print(f"代码运行时间: {execution_time} 秒")
return _chatbot


@@ -125,7 +129,7 @@ def setbar():
QA_bad.click(logQA, [chatbot,QA_bad], [])
#run.click(makeQA, [chatbot, content], [chatbot], show_progress=True)

demo.launch(server_name="0.0.0.0" ,server_port=8504)
demo.launch(server_name="127.0.0.1" ,server_port=6006)


if __name__ == '__main__':
@@ -134,19 +138,42 @@ def setbar():

parser = argparse.ArgumentParser()

parser.add_argument('--model_name_or_path', type=str, default="/workspace/mnt/storage/zhaozhijian/pt-fs/Qwen1___5-7B-Chat/")

parser.add_argument('--model_name_or_path', type=str, default="/workspace/mnt/storage/xiangxin/llm_storge/qwen1.5-7b-lora/examples/qwen/tmp/Qwen1.5-7B")
# parser.add_argument('--model_name_or_path', type=str, default="/root/autodl-tmp/1-gpu/")
parser.add_argument('--template', type=str, default="qwen")
parser.add_argument('--prompt_path', type=str, default="template/QAtemplat.txt")
parser.add_argument('--trt_llm', type=str, default='1', help='An optional parameter with a default empty string.')

args = parser.parse_args()

with open(args.prompt_path, 'r') as f:
template = ''.join(f.readlines())

#print(vars(args).keys())

model = ChatModel(vars(args))

_launch_demo(model, template)
if args.trt_llm != '':
parser.add_argument('--max_output_len', type=int, default=2048)
parser.add_argument(
'--tokenizer_dir',
type=str,
default='/workspace/mnt/storage/xiangxin/llm_storge/qwen1.5-7b-lora/examples/qwen/tmp/Qwen1.5-7B'
)
parser.add_argument(
'--engine_dir',
type=str,
default='/workspace/mnt/storage/xiangxin/llm_storge/qwen1.5-7b-lora/examples/qwen/tmp/qwen/7B/trt_engines/fp16/1-gpu'
)
parser.add_argument(
'--input_text',
type=str,
nargs='+',
default=["你好,请问你叫什么?"]
)

args = parser.parse_args()

#print(vars(args).keys())

model = ChatModel(vars(args))

_launch_demo(model, template)