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

GP regression with derivative information #462

Merged
merged 25 commits into from Jan 27, 2019
Merged

GP regression with derivative information #462

merged 25 commits into from Jan 27, 2019

Conversation

dme65
Copy link
Collaborator

@dme65 dme65 commented Jan 11, 2019

Adding a new kernel RBFKernelGrad and a mean ConstantMeanGrad that can be used together with MultiTask to train GPs from value and derivative information.

gpleiss
gpleiss requested changes Jan 23, 2019
View reviewed changes
Copy link
Member

@gpleiss gpleiss left a 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 changes to makes sure we don't have bugs down the road if this is used more generally.

gpytorch/kernels/scale_kernel.py Show resolved Hide resolved
gpytorch/means/constant_mean_grad.py Show resolved Hide resolved
gpytorch/kernels/rbf_kernel_grad.py Outdated Show resolved Hide resolved
dme65 and others added 25 commits January 27, 2019 18:22
@dme65 @gpleiss
Working on GPs with derivative observations.
4dc61b5
@jacobrgardner @gpleiss
WIP
ab65069
@dme65 @gpleiss
Working gpderiv in 1d
7d54fcf
@dme65 @gpleiss
Temporarily adding _create_input_grid + fixing diagonal
6b5188a
@dme65 @gpleiss
rbf_kernel_grad works in more than 1d + example
6b02be6
@dme65 @gpleiss
Reordering operations to get rid of cancellation
212e1f2
@dme65 @gpleiss
Adding comments to notebooks
cf6adac
@dme65 @gpleiss
Adding tests + cleaning up batch handling
556baff
@dme65 @gpleiss
Adding comments
6153d68
@dme65 @gpleiss
Docs + linting
d6bd723
@dme65 @gpleiss
Adding GPU support
a0ad6c0
@dme65 @gpleiss
Adding CUDA test
1ede625
@jacobrgardner @gpleiss
JIT for distance computation, kernels can override with custom JIT sc…
11b5c1b 
…ripts
@jacobrgardner @gpleiss
JIT script internals of Linear CG
a1e8bcc
@jacobrgardner @gpleiss
Matern kernel fix
b33a396
@jacobrgardner @gpleiss
Code reuse for JIT scripts
80864e1
@jacobrgardner @gpleiss
Kernels now pass scripts for handling postprocessing only
0ff6416
@jacobrgardner @gpleiss
Call postprocess in diag mode
8e7d3ed
@jacobrgardner @gpleiss
Remove double negative variable name
1f88504
@jacobrgardner @gpleiss
Implement Gauss-Hermite quadrature for use with likelihoods
6faba6a
@KeAWang @gpleiss
remove log_ kwargs deprecation
195fa8b 
addresses comments from #478
@KeAWang @gpleiss
fix GaussianLikelihood initialize noise bug
4108f3b 
closes #478
@KeAWang @gpleiss
update test files to not use log parameters
fc51fac
@dme65 @gpleiss
Adding size method to decorator kernels
7f469e4
@gpleiss
Fix squeeze in RBFKernelGrad, update examples
5edcbaa
@jacobrgardner jacobrgardner merged commit 5fa7ace into master Jan 27, 2019
@jacobrgardner jacobrgardner deleted the gpderiv branch January 27, 2019 23:30
Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment
Reviewers

@gpleiss gpleiss gpleiss approved these changes

Assignees
No one assigned
Labels
None yet
Projects
None yet
Milestone
No milestone
Development

Successfully merging this pull request may close these issues.

None yet
4 participants
@dme65 @gpleiss @jacobrgardner @KeAWang