How to use pycharm single-step debugging . Tips: run remote in server

[huizhilei]

I edit the configurations of finetune.py in pycharm as below:

--model_name_or_path facebook/galactica-1.3b --dataset_path /root/LMFlow/data/alpaca/train --output_dir /root/LMFlow/output_models/finetune_with_lora --overwrite_output_dir --num_train_epochs 0.01 --learning_rate 1e-4 --block_size 512 --per_device_train_batch_size 1 --use_lora 1 --lora_r 8 --save_aggregated_lora 0 --deepspeed /root/LMFlow/configs/ds_config_zero2.json --bf16 --run_name finetune_with_lora --validation_split_percentage 0 --logging_steps 20 --do_train --ddp_timeout 72000 --save_steps 5000 --dataloader_num_workers 4

When i run finetune.py, it works;; BUT when i debug the finetune.py ,it show below error: ImportError: cannot import name 'AutoPipeline' from partially initialized module 'lmflow.pipeline.auto_pipeline' (most likely due to a circular import) (/root/LMFlow/src/lmflow/pipeline/auto_pipeline.py)

[HuihuiChyan]

把evaluate.py这个文件改成evaluation.py，然后在对应的脚本里也改一下就行。 现在出现这个BUG是因为evaluate.py这个文件和包evaluate混淆了。

[huizhilei]

把evaluate.py这个文件改成evaluation.py，然后在对应的脚本里也改一下就行。 现在出现这个BUG是因为evaluate.py这个文件和包evaluate混淆了。

好嘞，谢谢哥！ 哥，再请教个问题，跑测试脚本的时候，在pycharm里面参数怎么配置py文件啊？ 这个CUDA_VISIBLE_DEVICES=0参数在pycharm中怎么配置呀？或者需要配置吗

[HuihuiChyan]

建议改用vscode

[shizhediao]

This issue has been marked as stale because it has not had recent activity. If you think this still needs to be addressed please feel free to reopen this issue. Thanks