GraphGPT

GraphGPT copied to clipboard

Error in saving model after stage1 tuning

1

After stage1 tuning, use safe_save_model_for_hf_trainer to save model params, but with “AttributeError: 'Trainer' object has no attribute 'deepspeed'” error when fixing this, meet same probs: "AttributeError: 'Trainer' object has no...

applekeyword

Hi, thanks for the great work. I've noted that the option 'use_graph_start_end' is set to false in the self-supervised tuning stage, but set to true in the task-specific tuning stage....

zfkarl

训练选择显卡设备

2

请问如何调整训练的显卡，比如说想用第二张卡训练这种，或者选择多卡训练

wwwww-k

Thank you for such an excellent work, it inspires me a lot. I would like to ask you if you can open source the raw TAG data of cora, pubmed,...

jujulili888

I found that in stage 1 and stage 2, the training data lacked benchmarks. Can you help explain this? Thank you! graph_matching.json in stage 1: Counter({'Industrial': 84003, 'arxiv': 74075, 'cora':...

zhuochunli

We ran into the same problem as issue#17, but we still got an error even though we had to comment out "replace_llama_attn_with_flash_attn()" ``` 0%| | 0/274797 [00:00

octopusStar218

关于stage_1输出

3

在使用轻量级第一阶段微调后，只得到last.ckpt这一个文件正常吗

zzzzzzyc

是我下载的clip_gt_arxiv_pub不正确么，还是有其它错误。

CodeCollectorliu