Enhance _cwt.py by introducing a configurable hop size parameter #804
Conversation
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
Generating a scalogram from the full output of the Continuous Wavelet Transform (CWT) entails high computational cost while providing limited performance gains in acoustic recognition models based on deep learning. Therefore, this update proposes reducing the output size during the intermediate processing stage—rather than after CWT generation—to improve computational efficiency of CWT. This pull request reflects the research findings presented in the following paper. Phan, D. T., Huynh, T. A., Pham, V. T., Tran, C. M., Mai, V. T., & Tran, N. Q. (2025). Optimal Scalogram for Computational Complexity Reduction in Acoustic Recognition Using Deep Learning. arXiv preprint arXiv:2505.13017.
rgommers
requested changes
May 26, 2025
There was a problem hiding this comment.
Thanks @phandangthoai. A few quick review comments inline. This change also needs tests, and it would be good to demonstrate what the gain of this is with a small benchmark.
|
|
||
| def cwt(data, scales, wavelet, sampling_period=1., method='conv', axis=-1): | ||
| try: | ||
| import scipy |
There was a problem hiding this comment.
This seems unrelated to the goal of this PR, and reintroduces the optional dependency - why?
| return 2**ceil(np.log2(n)) | ||
|
|
||
|
|
||
| def cwt(data, scales, wavelet, hop_size=1, sampling_period=1., method='conv', axis=-1): |
There was a problem hiding this comment.
hop_size is new so should be documented. Including an example may be useful too. As well as the reference to the paper.
|
Hello @rgommers, |
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.
Generating a scalogram from the full output of the Continuous Wavelet Transform (CWT) entails high computational cost while providing limited performance gains in acoustic recognition models based on deep learning. Therefore, this update proposes reducing the output size during the intermediate processing stage—rather than after CWT generation—to improve computational efficiency of CWT. This pull request reflects the research findings presented in the following paper.
Phan, D. T., Huynh, T. A., Pham, V. T., Tran, C. M., Mai, V. T., & Tran, N. Q. (2025). Optimal Scalogram for Computational Complexity Reduction in Acoustic Recognition Using Deep Learning. arXiv preprint arXiv:2505.13017.