-
Notifications
You must be signed in to change notification settings - Fork 83
New issue
Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.
By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.
Already on GitHub? Sign in to your account
refactor: new module wrapping design and algorithm implementation design #24
Conversation
Co-authored-by: github-actions[bot] <41898282+github-actions[bot]@users.noreply.github.com>
Co-authored-by: github-actions[bot] <41898282+github-actions[bot]@users.noreply.github.com>
Co-authored-by: github-actions[bot] <41898282+github-actions[bot]@users.noreply.github.com>
Co-authored-by: github-actions[bot] <41898282+github-actions[bot]@users.noreply.github.com>
previous buckets and hooks need to be cleared before reinitbagua/bagua/torch_api/distributed_dev.py Lines 13 to 18 in 71a759f
This comment was generated by todo based on a
|
document thisbagua/bagua/torch_api/distributed_dev.py Lines 34 to 39 in 71a759f
This comment was generated by todo based on a
|
broadcast optimizer parametersbagua/bagua/torch_api/distributed_dev.py Lines 61 to 66 in 71a759f
This comment was generated by todo based on a
|
bagua/torch_api/dev/tensor.py
Outdated
|
||
import torch | ||
|
||
class BaguaTensor(torch.Tensor): |
There was a problem hiding this comment.
Choose a reason for hiding this comment
The reason will be displayed to describe this comment to others. Learn more.
[blackfmt] reported by reviewdog 🐶
class BaguaTensor(torch.Tensor): | |
class BaguaTensor(torch.Tensor): |
bagua/torch_api/dev/tensor.py
Outdated
def is_registered(self) -> bool: | ||
return not (self.bagua_backend is None) | ||
|
||
|
There was a problem hiding this comment.
Choose a reason for hiding this comment
The reason will be displayed to describe this comment to others. Learn more.
[blackfmt] reported by reviewdog 🐶
@ganshaoduo check if this should be usedbagua/bagua/torch_api/algorithms/onebit_adam.py Lines 192 to 197 in 1cfccfe
This comment was generated by todo based on a
|
remove parameter group logicbagua/bagua/torch_api/distributed.py Lines 276 to 281 in 1cfccfe
This comment was generated by todo based on a
|
@shjwudp check whether these are still neededbagua/bagua/torch_api/utils.py Lines 261 to 264 in 1cfccfe
This comment was generated by todo based on a
|
@ganshaoduo check if this should be usedbagua/bagua/torch_api/algorithms/onebit_adam.py Lines 192 to 197 in 96cb6fe
This comment was generated by todo based on a
|
remove parameter group logicbagua/bagua/torch_api/distributed.py Lines 276 to 281 in 96cb6fe
This comment was generated by todo based on a
|
@shjwudp check whether these are still neededbagua/bagua/torch_api/utils.py Lines 261 to 264 in 96cb6fe
This comment was generated by todo based on a
|
Goal: To support easier implementation of new algorithms like OneBitAdam. Decouple bucketing policy from wrapper.