Skip to content

Conversation

@peterkentkiernan
Copy link

Multiplying by a constant without specifying adds an extra parameter, which breaks the code. Instead, we need to specify that the parameter in the constant kernel should not be changed. (I also fixed some minor oversights on checking if e or g was out of bounds.)

rmjarvis and others added 14 commits October 3, 2022 12:16
Multiplying by a constant normally adds an extra parameter from the ConstantKernel, which breaks the code; we need to call ConstantKernel directly and specify that this should be an immutable parameter.

Also removed the frivolous differences between rmjarvis/treegp (which this was originally cloned from) and pfleget/treegp (which this is being pushed to.
@PFLeget
Copy link
Owner

PFLeget commented Dec 6, 2023

@peterkentkiernan I am sorry to come back so late on that PR.

I tried to reproduce the error but I am not able to reproduce it... I am using the version 1.3.2 of sklearn.

Do you have an example of the problem?

else:
L = get_correlation_length_matrix(corr_length, g1, g2)
invLam = np.linalg.inv(L)
kernel_used = sigma**2 * self.kernel_class(invLam=invLam)
Copy link
Owner

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

Having an example of this failing can be nice. I am not able to reproduce this error.

Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment

Labels

None yet

Projects

None yet

Development

Successfully merging this pull request may close these issues.

3 participants