Improving performance of data loading and collecting #223
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
added 2 commits
February 10, 2022 16:13
1. convert list first to np.ndarray before to torch.tensor 2. reorder ndarray faster
facebook-github-bot
added
the
CLA Signed
|
By the way, the time measurement in current code (i.e., from line 1534 - line 1596) is not accurate because it doesn't contain the time cost of the data collection before each iteration (i.e., line 1517), which actually costs almost as much time as the training process for Kaggle dataset. Line 1517 in 9c2fda7 After the optimization of this PR, the data collection process only costs about 2 seconds for Kaggle dataset. |
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
2 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Outdated suggestions cannot be applied.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.
This pull request fixes the performance issue #219