Adding heteroskedastic tests #1508
Merged
Conversation
This file contains bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
st--
reviewed
Jun 16, 2020
tests/gpflow/likelihoods/test_heteroskedastic_constant_variance.py
Outdated
Show resolved
Hide resolved
tests/gpflow/likelihoods/test_heteroskedastic_constant_variance.py
Outdated
Show resolved
Hide resolved
tests/gpflow/likelihoods/test_heteroskedastic_constant_variance.py
Outdated
Show resolved
Hide resolved
tests/gpflow/likelihoods/test_heteroskedastic_constant_variance.py
Outdated
Show resolved
Hide resolved
tests/gpflow/likelihoods/test_heteroskedastic_constant_variance.py
Outdated
Show resolved
Hide resolved
| ) | ||
|
|
||
|
|
||
| @pytest.mark.skip("Currently broken as it returns the sum over outputs when given multiple outputs") |
There was a problem hiding this comment.
Do you want to fix the code to make this test pass or merge it as is and I can change the code (and reactivate the test)?
There was a problem hiding this comment.
I think this would be easier for you to fix, than for me to fix. I would rather merge it as is, such that you can change the code and reactivate the test
Co-authored-by: st-- <st--@users.noreply.github.com>
|
The type-check issues in circleci are the same issues as seen in the source branch PR: #1455 |
gustavocmv
added a commit
that referenced
this pull request
Sep 15, 2020
* HeteroskedasticLikelihood base class draft * fixup * cleanup * cleanup heteroskedastic * multioutput likelihood WIP * Notebook exemplifying HeteroskedasticTFPDistribution usage (#1462) * fixes * typo fix; reshaping fix * notebook showing how to use HeteroskedasticTFPDistribution likelihood * converting to .pct.py format * removed .ipynb * better descriptions * black auto-formatting Co-authored-by: Gustavo Carvalho <gustavo.carvalho@delfosim.com> * note and bugfix * add comment * Adding heteroskedastic tests (#1508) These tests ensure that heteroskedastic likelihood with a constant variance, will give the same results as a Gaussian likelihood with the same variance. * testing * added QuadratureLikelihood to base, refactored ScalarLikelihood to use it * fix * using the first dimension to hold the quadrature summation * adapting ndiagquad wrapper * merged with gustavocmv/quadrature-change-shape * removed unecessary tf.init_scope * removed print and tf.print * removed print and tf.print * Type annotations Co-authored-by: Vincent Dutordoir <dutordoirv@gmail.com> * Work * Fix test * Remove multioutput from PR * Fix notebook * Add student t test * More tests * Copyright * Removed NDiagGHQuadratureLikelihood class in favor of non-abstract QuadratureLikelihood * _set_latent_and_observation_dimension_eagerly * n_gh ---> num_gauss_hermite_points * removed NDiagGHQuadratureLikelihood from test * black * bugfix * removing NDiagGHQuadratureLikelihood from test * fixed bad commenting * black * refactoring scalar likelihood * adding dtype casts to quadrature * black * small merging fixes * DONE: swap n_gh for num_gauss_hermite_points * black Co-authored-by: ST John <st@prowler.io> Co-authored-by: gustavocmv <47801305+gustavocmv@users.noreply.github.com> Co-authored-by: Gustavo Carvalho <gustavo.carvalho@delfosim.com> Co-authored-by: st-- <st--@users.noreply.github.com> Co-authored-by: joshuacoales-pio <47976939+joshuacoales-pio@users.noreply.github.com>
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
Add this suggestion to a batch that can be applied as a single commit.
This suggestion is invalid because no changes were made to the code.
Suggestions cannot be applied while the pull request is closed.
Suggestions cannot be applied while viewing a subset of changes.
Only one suggestion per line can be applied in a batch.
Add this suggestion to a batch that can be applied as a single commit.
Applying suggestions on deleted lines is not supported.
You must change the existing code in this line in order to create a valid suggestion.
Outdated suggestions cannot be applied.
This suggestion has been applied or marked resolved.
Suggestions cannot be applied from pending reviews.
Suggestions cannot be applied on multi-line comments.
Suggestions cannot be applied while the pull request is queued to merge.
Suggestion cannot be applied right now. Please check back later.
This adds some tests to @st-- 's heteroskedastic branch.
These tests ensure that heteroskedastic likelihood with a constant variance, will give the same results as a Gaussian likelihood with the same variance.
3 of these tests are failing and currently skipped.
conditional_mean()andconditional_variance()tests are skipped because they are not implemented in Heteroskedastic likelihood, which cause the tests to fail. I've asked in GPFlow slack as to whether anyone uses these directly.predict_log_density()test is skipped because it does not return the same result, as it returns the sum over outputs when given multiple outputs. I have also asked about this in GPFlow slack.