Enable Qwen-7B-Chat

[huiyan2021]

Type of Change

Add support for Qwen-7B-Chat

Description

1. Add support for Qwen-7B-Chat
2. Modify deploy_chatbot_on_xpu.ipynb again seems the modification I made before has not been merged

Expected Behavior & Potential Risk

Qwen-7B-Chat can be run correctly

How has this PR been tested?

This PR can be tested in build_chatbot_on_spr.ipynb and build_chatbot_on_xpu.ipynb after environment prepared

from intel_extension_for_transformers.neural_chat import build_chatbot from intel_extension_for_transformers.neural_chat.config import PipelineConfig config = PipelineConfig(model_name_or_path="Qwen/Qwen-7B-Chat") chatbot = build_chatbot(config) response = chatbot.predict("千问是什么？") print(response) 

Dependency Change?

see requirements_cpu.txt and requirements_xpu.txt

[hshen14]

@huiyan2021 please add UT to improve the coverage of your newly-added qwen model code.

[huiyan2021]

@huiyan2021 please add UT to improve the coverage of your newly-added qwen model code.

@lvliang-intel will help on this, thanks!