implement mini-epochs in training #60
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
… go out of bounds
fixes crash when tf tries to call with input tensor of shape [None, None, None, None]
|
Testing with fine-tuning a model. Will roll out with r7. |
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Outdated suggestions cannot be applied.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.
This PR implements mini-epochs (multiple rounds of validation per full traversal of the training data) in Train.py.
I have found that running with
--mini_epochs 5slightly improves the accuracy of the resulting model, since models at intermediate stages of an epoch may be better than those at the end of an epoch. If no--mini_epochsparameter is provided, training should be the same as currently.To facilitate this change, training batch generation has been reimplemented as a
keras.utils.Sequenceclass, since this has anon_epoch_endmethod that can be used to count miniepochs and only shuffle when a full epoch has been completed. This code is based on theSequenceBatcherimplemented in Medaka.