deepspeedai / deepspeedai/DeepSpeed
[BUG][during training]AttributeError: 'DeepSpeedCPUAdam' object has no attribute 'ds_opt_adam'
Open
@GuanhuaWang is already working on this.
Since May 18, 2023.
bug
training
- Dominant language
- Python
- Stars
- 43.1k
- Forks
- 5k
- Avg merge
- 4d 15h
- Merged PRs (30d)
- 112
Description
Describe the bug
I ran my script to train successfully, but during the training process, the following error occurred:
self.ds_opt_adam.destroy_adam(self.opt_id)
AttributeError: 'DeepSpeedCPUAdam' object has no attribute 'ds_opt_adam'
The screenshot of the error message is as follows:
And I can run the following code without any errors:
python -c 'import deepspeed; deepspeed.ops.adam.cpu_adam.CPUAdamBuilder().load()'
I tried the training script again, and every time this error occurs in step 14251. So a bad batch data may trigger this error? Why did this happen?
Contributor guide
First steps
- Read the whole issue, then the project's contributing guide.
- Comment on the issue to say you are picking it up — it saves two people doing the same work.
- Fork the repository and make your change on a branch.
- Open a pull request that references the issue number.
Assessment
This issue has not been assessed yet.