[Gradient Compression] Explicitly specify the dtype of the error tensor#50985
Closed
wayi1 wants to merge 5 commits intogh/SciPioneer/48/basefrom
Closed
[Gradient Compression] Explicitly specify the dtype of the error tensor#50985wayi1 wants to merge 5 commits intogh/SciPioneer/48/basefrom
wayi1 wants to merge 5 commits intogh/SciPioneer/48/basefrom
Conversation
Explicitly specify the dtype of error tensor when it is initialized by zeros. Previously if the dtype of input tensor is FP16, the error tensor is still created in FP32, although later it will be assigned by another FP16 tensor (`input_tensor_cp` - `input_tensor`). This change will make the dtype of error tensor look more clear. Original PR issue: Investigate Applying PowerSGD to Communication Hook for Gradient Compression #47202 Differential Revision: [D26034988](https://our.internmc.facebook.com/intern/diff/D26034988/) [ghstack-poisoned]
Contributor
💊 CI failures summary and remediationsAs of commit 4f79850 (more details on the Dr. CI page):
🕵️ 3 new failures recognized by patternsThe following CI failures do not appear to be due to upstream breakages:
|
This was referenced Jan 23, 2021
Closed
… error tensor" Explicitly specify the dtype of error tensor when it is initialized by zeros. Previously if the dtype of input tensor is FP16, the error tensor is still created in FP32, although later it will be assigned by another FP16 tensor (`input_tensor_cp` - `input_tensor`). This change will make the dtype of error tensor look more clear. Original PR issue: Investigate Applying PowerSGD to Communication Hook for Gradient Compression #47202 Differential Revision: [D26034988](https://our.internmc.facebook.com/intern/diff/D26034988/) [ghstack-poisoned]
wayi1
pushed a commit
that referenced
this pull request
Jan 23, 2021
Pull Request resolved: #50985 Explicitly specify the dtype of error tensor when it is initialized by zeros. Previously if the dtype of input tensor is FP16, the error tensor is still created in FP32, although later it will be assigned by another FP16 tensor (`input_tensor_cp` - `input_tensor`). This change will make the dtype of error tensor look more clear. Original PR issue: Investigate Applying PowerSGD to Communication Hook for Gradient Compression #47202 ghstack-source-id: 120259409 Differential Revision: [D26034988](https://our.internmc.facebook.com/intern/diff/D26034988/)
… error tensor" Explicitly specify the dtype of error tensor when it is initialized by zeros. Previously if the dtype of input tensor is FP16, the error tensor is still created in FP32, although later it will be assigned by another FP16 tensor (`input_tensor_cp` - `input_tensor`). This change will make the dtype of error tensor look more clear. Additionally, also explicitly specify the dtype if rank-1 tensor buffer is empty. Original PR issue: Investigate Applying PowerSGD to Communication Hook for Gradient Compression #47202 Differential Revision: [D26034988](https://our.internmc.facebook.com/intern/diff/D26034988/) [ghstack-poisoned]
wayi1
pushed a commit
that referenced
this pull request
Jan 25, 2021
Pull Request resolved: #50985 Explicitly specify the dtype of error tensor when it is initialized by zeros. Previously if the dtype of input tensor is FP16, the error tensor is still created in FP32, although later it will be assigned by another FP16 tensor (`input_tensor_cp` - `input_tensor`). This change will make the dtype of error tensor look more clear. Additionally, also explicitly specify the dtype if rank-1 tensor buffer is empty. Original PR issue: Investigate Applying PowerSGD to Communication Hook for Gradient Compression #47202 ghstack-source-id: 120328964 Differential Revision: [D26034988](https://our.internmc.facebook.com/intern/diff/D26034988/)
rohan-varma
reviewed
Jan 26, 2021
… error tensor" Explicitly specify the dtype of error tensor when it is initialized by zeros. Previously if the dtype of input tensor is FP16, the error tensor is still created in FP32, although later it will be assigned by another FP16 tensor (`input_tensor_cp` - `input_tensor`). This change will make the dtype of error tensor look more clear. Additionally, also explicitly specify the dtype if rank-1 tensor buffer is empty. Original PR issue: Investigate Applying PowerSGD to Communication Hook for Gradient Compression #47202 Differential Revision: [D26034988](https://our.internmc.facebook.com/intern/diff/D26034988/) [ghstack-poisoned]
wayi1
pushed a commit
that referenced
this pull request
Jan 26, 2021
Pull Request resolved: #50985 Explicitly specify the dtype of error tensor when it is initialized by zeros. Previously if the dtype of input tensor is FP16, the error tensor is still created in FP32, although later it will be assigned by another FP16 tensor (`input_tensor_cp` - `input_tensor`). This change will make the dtype of error tensor look more clear. Additionally, also explicitly specify the dtype if rank-1 tensor buffer is empty. Original PR issue: Investigate Applying PowerSGD to Communication Hook for Gradient Compression #47202 ghstack-source-id: 120377786 Differential Revision: [D26034988](https://our.internmc.facebook.com/intern/diff/D26034988/)
… error tensor" Explicitly specify the dtype of error tensor when it is initialized by zeros. Previously if the dtype of input tensor is FP16, the error tensor is still created in FP32, although later it will be assigned by another FP16 tensor (`input_tensor_cp` - `input_tensor`). This change will make the dtype of error tensor look more clear. Additionally, also explicitly specify the dtype if rank-1 tensor buffer is empty. Original PR issue: Investigate Applying PowerSGD to Communication Hook for Gradient Compression #47202 Differential Revision: [D26034988](https://our.internmc.facebook.com/intern/diff/D26034988/) [ghstack-poisoned]
Contributor
|
This pull request has been merged in 9d731e8. |
laurentdupin
pushed a commit
to laurentdupin/pytorch
that referenced
this pull request
Apr 24, 2026
…or (pytorch#50985) Summary: Pull Request resolved: pytorch#50985 Explicitly specify the dtype of error tensor when it is initialized by zeros. Previously if the dtype of input tensor is FP16, the error tensor is still created in FP32, although later it will be assigned by another FP16 tensor (`input_tensor_cp` - `input_tensor`). This change will make the dtype of error tensor look more clear. Additionally, also explicitly specify the dtype if rank-1 tensor buffer is empty. Original PR issue: Investigate Applying PowerSGD to Communication Hook for Gradient Compression pytorch#47202 ghstack-source-id: 120377786 Test Plan: buck test mode/dev-nosan caffe2/test/distributed:c10d -- test_powerSGD_ddp_comm_hook_nccl buck test mode/dev-nosan caffe2/test/distributed:distributed_nccl_fork -- test_DistributedDataParallel_powerSGD_ddp_comm_hook Reviewed By: rohan-varma Differential Revision: D26034988 fbshipit-source-id: e0d323d0b77c6a2478cdbe8b31a1946ffd1a07da
This file contains hidden or bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
3 participants
Add this suggestion to a batch that can be applied as a single commit.This suggestion is invalid because no changes were made to the code.Suggestions cannot be applied while the pull request is closed.Suggestions cannot be applied while viewing a subset of changes.Only one suggestion per line can be applied in a batch.Add this suggestion to a batch that can be applied as a single commit.Applying suggestions on deleted lines is not supported.You must change the existing code in this line in order to create a valid suggestion.Outdated suggestions cannot be applied.This suggestion has been applied or marked resolved.Suggestions cannot be applied from pending reviews.Suggestions cannot be applied on multi-line comments.Suggestions cannot be applied while the pull request is queued to merge.Suggestion cannot be applied right now. Please check back later.
Stack from ghstack:
Explicitly specify the dtype of error tensor when it is initialized by zeros.
Previously if the dtype of input tensor is FP16, the error tensor is still created in FP32, although later it will be assigned by another FP16 tensor (
input_tensor_cp-input_tensor).This change will make the dtype of error tensor look more clear.
Additionally, also explicitly specify the dtype if rank-1 tensor buffer is empty.
Original PR issue: Investigate Applying PowerSGD to Communication Hook for Gradient Compression #47202
Differential Revision: D26034988