Qwen-VL-Chat visual part #2
Description
起始日期 | Start Date
No response
实现PR | Implementation PR
No response
相关Issues | Reference Issues
No response
摘要 | Summary
Hi,
Did you make any modification to visual part in the file modeling_qwen.py. As I understand that you modified the encode method in visual.py file to receive image instead of its path. Could you help me locate where the visual embedding is forward to the LLM model during training?
Thanks
基本示例 | Basic Example
(No)
缺陷 | Drawbacks
(No)
未解决问题 | Unresolved questions
No response