bofei5675

https://github.com/IDEA-Research/Grounded-Segment-Anything/issues/15 address this problem by looking at this one

@ChenSteventx This problem is cuased by the cuda version your torch compiled on is different to the one on your current device. For example, you can check your current python...

Get same error

I have the same question when setting `--per_device_train_batch_size` greater than 1.

> I have the same question when setting `--per_device_train_batch_size` greater than 1. Address by setting `use_reentrant=False` ```python training_args.gradient_checkpointing_kwargs = { "use_reentrant": False } trainer = CPMTrainer( model=model, tokenizer=tokenizer, args=training_args, **data_module,...

Address this issue by setting `scripts/zero2.json` or `scripts/zero3.json` ```json { "overlap_comm": false } ```

@pooruss There are some possible bugs or I might mis-understand the code in https://github.com/xlang-ai/OSWorld/blob/main/mm_agents/uitars_agent.py Try to fix on my side but not sure that's correct Fix 1: [Line 636](https://github.com/xlang-ai/OSWorld/blob/c179d0de122cf65315aa30ecb4aa2aa1eb52292f/mm_agents/uitars_agent.py#L636) Since...