-
Notifications
You must be signed in to change notification settings - Fork 49
Support for ablation studies for AddBiomechanics ECCV paper #212
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.
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
base: master
Are you sure you want to change the base?
Conversation
nickbianco
left a comment
There was a problem hiding this comment.
Choose a reason for hiding this comment
The reason will be displayed to describe this comment to others. Learn more.
Looks good! Just a few minor comments to clarify some potential changes to the dynamic fitter algorithm.
| int driftCorrectionBlurRadius = 250, | ||
| int driftCorrectionBlurInterval = 250); | ||
| int driftCorrectionBlurInterval = 250, | ||
| s_t regularizeUnobservedTimesteps = 0.01); |
There was a problem hiding this comment.
Choose a reason for hiding this comment
The reason will be displayed to describe this comment to others. Learn more.
The default value for regularizeUnobservedTimesteps was changed from 0.1 to 0.01 here -- will that have any meaningful effect on the dynamics fitter algorithm?
|
|
||
| // Now that we've formulated our problem, we can attempt to solve: | ||
|
|
||
| // Eigen::VectorXs tentativeResult = |
There was a problem hiding this comment.
Choose a reason for hiding this comment
The reason will be displayed to describe this comment to others. Learn more.
Remove all this commented out code?
| if (maxIterations > 1000) | ||
| { | ||
| maxIterations = 1000; | ||
| } |
There was a problem hiding this comment.
Choose a reason for hiding this comment
The reason will be displayed to describe this comment to others. Learn more.
Capping at a 1000 iterations shouldn't change much right? This usually converges in a handful of iterations if everything is set up right.
This PR adds support for a few methods we needed in order to run some ablation tests requested by reviewers for ECCV. A few small GUI optimization fixes also got wrapped up in this PR.