Corrected docs to run JPQD in DDP mode #315
Merged
Conversation
This file contains 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
|
@vuiseng9 @yujiepan-work @AlexKoff88 please take a look |
|
LGTM |
|
Thanks @ljaljushkin ! It would be good to update the example tests too (https://github.com/huggingface/optimum-intel/blob/main/tests/openvino/test_training_examples.py ), so that the test tests what we show in the docs (probably good to test both methods for at least one example). We can also do that in a separate PR. |
|
The documentation is not available anymore as the PR was closed or merged. |
AlexKoff88
reviewed
May 15, 2023
AlexKoff88
approved these changes
May 15, 2023
ljaljushkin
commented
May 19, 2023
| @@ -92,4 +92,4 @@ python run_audio_classification.py \ | |||
| --seed 0 | |||
| ``` | |||
|
|
|||
| This script should take about 3 hours on a single V100 GPU and produce a quantized Wav2Vec2-base model with ~80% structured sparsity in its linear layers. The model accuracy should converge to about 97.5%. | |||
| This script should take about 3 hours on a single V100 GPU and produce a quantized Wav2Vec2-base model with ~80% structured sparsity in its linear layers. The model accuracy should converge to about 97.5%. For launching the script on multiple GPUs specify `--nproc-per-node=<number of GPU>`. Note, that different batch size and other hyperparameters might be required to achieve the same results as on a single GPU. | |||
There was a problem hiding this comment.
@vuiseng9 @yujiepan-work @AlexKoff88 mentioned necessity for hyperparameters tuning in case of multiple GPU
Thanks @helena-intel! It makes sense. I've corrected tests here: #319 |
echarlaix
approved these changes
Jun 7, 2023
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
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.
What does this PR do?
Corrects command line for launching JPQD. Currently, it launches training in DataParallel (DP) mode, which is slower than Distributed Data Parallel (DDP) and may have some issue with movement pruning openvinotoolkit/nncf#1582
Before submitting