Skip to content
New issue

Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.

Sign up for GitHub

By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.

Already on GitHub? Sign in to your account

Jump to bottom

Enable Many-to-Many training for rolling window classifiers #36

Open
lucasczz opened this issue Aug 8, 2022 · 0 comments
Open

Enable Many-to-Many training for rolling window classifiers #36

lucasczz opened this issue Aug 8, 2022 · 0 comments
Assignees
@lucasczz
Labels
enhancement New feature or request

Comments

@lucasczz
Copy link
Collaborator

lucasczz commented Aug 8, 2022 •
edited
Loading

For LSTM-based models it would probably make sense to use the loss of the predictions over the entire rolling window (see figure below) since they are computed anyways to obtain prediction for the last timestep.

image
Source: http://karpathy.github.io/2015/05/21/rnn-effectiveness/
@lucasczz lucasczz self-assigned this Aug 8, 2022
@lucasczz lucasczz added the enhancement New feature or request label Aug 8, 2022
@lucasczz lucasczz changed the title Enable Many-to-Many prediction for rolling window classifiers Enable Many-to-Many training for rolling window classifiers Aug 30, 2022
Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment
Assignees

@lucasczz lucasczz

Labels
enhancement New feature or request
Projects
None yet
Milestone
No milestone
Development

No branches or pull requests
1 participant
@lucasczz