Kaeli
Kaeli
I have the same question. But I solved it by reinstalling `torchdiffeq==0.0.1`.
Sorry, my notebook is the latest version. But I still met the problem. @ysaito8015
数据预处理时,针对LLaMA模型需要做一些修改。 - 将 `tencentpretrain/utils/constants.py` 文件第4行`special_tokens_map.json` 修改为 `llama_special_tokens_map.json` - 参考[llama-training](https://github.com/Tencent/TencentPretrain/blob/main/documents/llama.md)
> Hi,你认为目前现有PaddleOCR里的版本分析模型的部署方案有哪些点觉得需要提升吗 原部署方案依赖的框架较大,需要更多的运行内存。期望一个更精简的方案。
Thank you for your interest in this work! Sorry we've been so busy lately. The training is simple and you can easily reproduce it. We will provide training code and...
I have read the CLA Document and I hereby sign the CLA
recheck