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.

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

Add Gaussian Process Model #18

Open
muammar opened this issue Aug 1, 2019 · 1 comment
Open

Add Gaussian Process Model #18

muammar opened this issue Aug 1, 2019 · 1 comment
Assignees
Labels
enhancement New feature or request

Comments

@muammar
Copy link
Owner

muammar commented Aug 1, 2019

Another one to the zoo.

@muammar muammar self-assigned this Aug 3, 2019
@muammar muammar added the enhancement New feature or request label Aug 3, 2019
muammar added a commit that referenced this issue Aug 3, 2019
- ml4chem/data/parser.py: improved FakeCalculator class and its
  docstrings.
- ml4chem/models/kernelridge.py:
    * New `class_name` parameter is saved to operate with universal
      loading of models.
    * get_kernel_matrix() function can handle more input cases.
    * Improvement in docstrings.
- ml4chem/potentials.py: universal loading of models.
- setup.py: now the `ml4chem` command line tool is installed with pip.
- ml4chem/models/gaussian_process.py: the new gaussian process
  regression model for ml4chem. It uses the KernelRidge class in the
  kernelridge module as a base class to then compute the extra step
  referring to the inference variance. This is related to #18.
- Updated examples/.
@muammar
Copy link
Owner Author

muammar commented Aug 3, 2019

I think this is implemented now but the following aspects have to be improved:

  1. Compute training data K just once at inference time. See https://github.com/muammar/ml4chem/blob/master/ml4chem/models/gaussian_process.py#L171 and https://github.com/muammar/ml4chem/blob/master/ml4chem/potentials.py#L331.
  2. Add some tools to plot the kernels.

Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment
Labels
enhancement New feature or request
Projects
None yet
Development

No branches or pull requests

1 participant