Feiyang Lee

返回 embedding 和 huggingface 的返回结果不完全一致

4

比如 bert-base-chinese，作者是否有做过这方面的评估测试呀～

search功能基本不能用。。

2

中文貌似不能搜索诶

python 3.7 tensorflow 1.13.2 

ERNIE 1.0 中文 Base 模型(max_len=512) 考虑转化吗

1

https://github.com/PaddlePaddle/ERNIE/blob/develop/README.zh.md#%E9%A2%84%E8%AE%AD%E7%BB%83%E6%A8%A1%E5%9E%8B%E4%B8%8B%E8%BD%BD

model and code

1

Thanks for this job, I don’t know if the pre-train model & code will be released in the future~

Question about llama_flash_attn_monkey_patch

1

https://github.com/PKU-YuanGroup/Chat-UniVi/blob/main/ChatUniVi/train/llama_flash_attn_monkey_patch.py is different with https://github.com/haotian-liu/LLaVA/blob/main/llava/train/llama_flash_attn_monkey_patch.py for example: https://github.com/PKU-YuanGroup/Chat-UniVi/blob/main/ChatUniVi/train/llama_flash_attn_monkey_patch.py  and https://github.com/haotian-liu/LLaVA/blob/main/llava/train/llama_flash_attn_monkey_patch.py  it seems chat-univi change some code in **llama_flash_attn_monkey_patch**, can you help explain the reason for modifying the code?...