Skip to content
This repository has been archived by the owner. It is now read-only.

Automatic Augmentation of Data in BalancedPathFilter #516

Closed
AlexDBlack opened this issue Feb 28, 2018 · 2 comments
Closed

Automatic Augmentation of Data in BalancedPathFilter #516

AlexDBlack opened this issue Feb 28, 2018 · 2 comments
Labels

Comments

@AlexDBlack
Copy link
Contributor

@AlexDBlack AlexDBlack commented Feb 28, 2018

Moved from DL4J issue: eclipse/deeplearning4j#4737

Currently the BalancedPathFilter Class removes training examples from classes where there are more examples to match classes where there are less examples to balance the training data.
Proposal is to have an option of also automatically increasing the numbers of training examples for the classes with less training examples to match the number of the classes with more training examples. This way we give us more option on how to treat imbalanced Training datasets

Note I'm in favor of having this functionality, but probably would make a new class instead of trying to modify BalancedPathFilter.

@saudet

This comment has been minimized.

Copy link
Contributor

@saudet saudet commented Mar 1, 2018

I think that's pretty much what is proposed in pull #439. We just need someone to polish it up a bit!
/cc @Ngosti2000

@raver119 raver119 added the ETL label Apr 29, 2018
@AlexDBlack

This comment has been minimized.

Copy link
Contributor Author

@AlexDBlack AlexDBlack commented May 16, 2018

@AlexDBlack AlexDBlack closed this May 16, 2018
Sign up for free to subscribe to this conversation on GitHub. Already have an account? Sign in.
Projects
None yet
Linked pull requests

Successfully merging a pull request may close this issue.

None yet
3 participants
You can’t perform that action at this time.