12915494174

Results 6 issues of 12915494174

按照README.md文件直接打开界面,提问报了错误如下图: ![image](https://github.com/CrazyBoyM/llama3-Chinese-chat/assets/68487349/4dddd600-a796-4c7c-b0cc-1ea12a978490) 日志: ![image](https://github.com/CrazyBoyM/llama3-Chinese-chat/assets/68487349/ccc09477-a3bf-4be4-b5df-7e72b71f6a57) `root@autodl-container-30634997bd-fb05be27:~/llama3-Chinese-chat# streamlit run deploy/web_streamlit_for_instruct.py /root/autodl-tmp/llama-3-8b-Instruct-chinese_v2 --theme.base="dark" Collecting usage statistics. To deactivate, set browser.gatherUsageStats to false. You can now view your Streamlit app in your browser. Network...

LLaMA3-8B-Instruct+lora微调,单张A800(80GB显存)能否8192长度的token微调,我的任务场景较为特殊,需要使用较长的文本来微调。我使用了该代码库提供的代码,在微调过程中遇到了显存溢出的问题?

Hello, I encountered this error when trying to convert wasm into normal assembly using this tool. Is there a solution to this problem now? Error: Buffer still has 851309 bytes...

使用了GPU运行,但是CPU使用率依旧很高?根据性能报告,GPU确实生效。请问这个现象的原因是为什么? 我的代码如下: ```from funasr import AutoModel import time import json # paraformer-zh is a multi-functional asr model # use vad, punc, spk or not as you need model = AutoModel(...

是否支持时间戳识别出录音中每句话的情感?我是用funasr+emotion2vec_plus_large模型进行了测试,发现只能输出整段录音的情感,添加vad后依然是只有整段的。

我想将该项目作为一个依赖引入到springboot项目中,而现在我并没有找到该项目自定义词库动态更新的API。我想问下该版本是否有该设计?或者后续是否有此项设计的安排?