Search results

How can I use the trained model for prediction? is it right to use prediction_model.predict(new_x) ?

Status: Open.

#17 In yaringal/multi-task-learning-example;

· Qianqian-Yang opened on Apr 17, 2022

Why return torch.mean(loss)?

Status: Open.

#16 In yaringal/multi-task-learning-example;

· edshkim98 opened on Apr 6, 2021

Log var can become negative and explode

Status: Open.

#14 In yaringal/multi-task-learning-example;

· snie2012 opened on Feb 5, 2021

Question about the loss

Status: Open.

#13 In yaringal/multi-task-learning-example;

· caishiqing opened on Dec 23, 2020

This is a lucky demo When I change the data generation process, the prediction of the variance is wrong

Status: Open.

#12 In yaringal/multi-task-learning-example;

· conan-ux opened on Nov 26, 2020

uncertainty for self-supervised learning

Status: Open.

#11 In yaringal/multi-task-learning-example;

· griffintin opened on Apr 14, 2020

some questions about formulation 10 in paper

Status: Open.

#10 In yaringal/multi-task-learning-example;

· neil-yc opened on Apr 9, 2020

Sounds like a lucky result comes from a wrong formula deduction

Status: Open.

#9 In yaringal/multi-task-learning-example;

· edfall opened on Feb 26, 2020

Question on relative weights

Status: Open.

#8 In yaringal/multi-task-learning-example;

· kmbmjn opened on Nov 28, 2019

Any way to incorporate these methods in other tasks easily?

Status: Open.

#7 In yaringal/multi-task-learning-example;

· Jacobew opened on Sep 22, 2019

The loss might be nagative value

Status: Open.

#6 In yaringal/multi-task-learning-example;

· begeekmyfriend opened on Apr 23, 2019

MergeLoss with regular item is \log_{sigma} in paper but \log_{sigma}^2 in code

Status: Open.

#4 In yaringal/multi-task-learning-example;

· songzeballboy opened on Feb 25, 2019