Thanks to visit codestin.com
Credit goes to github.com

Skip to content

[fix] add sum over Poisson.log_prob in loss function #20

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
Merged
merged 1 commit into from
Feb 3, 2025
Merged

[fix] add sum over Poisson.log_prob in loss function #20

merged 1 commit into from
Feb 3, 2025

Conversation

@felixzinn
Copy link
Contributor

in the calculation of the poisson term of the likelihood function the sum has to be taken before the constraints are added with loss_val += evm.util.sum_over_leaves(constraints) because Poisson.log_prob returns a multi-dimensional array with the same shape as observation

so it has to be

@eqx.filter_jit
def loss(dynamic_model, static_model, hists, observation):
    model = eqx.combine(dynamic_model, static_model)
    expectations = model(hists)
    constraints = evm.loss.get_log_probs(model)
    loss_val = (
        evm.pdf.Poisson(evm.util.sum_over_leaves(expectations))
        .log_prob(observation)
        .sum()
    )
    # add constraint
    loss_val += evm.util.sum_over_leaves(constraints)
    return -jnp.sum(loss_val)

changed in the documentation and examples

@pfackeldey
Copy link
Owner

Thanks 👍

@pfackeldey pfackeldey merged commit e863f9a into pfackeldey:main Feb 3, 2025
3 checks passed
@felixzinn felixzinn deleted the fix/examples branch February 17, 2025 15:09
Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment

Reviewers

No reviews

Assignees

No one assigned

Labels

None yet

Projects

None yet

Milestone

No milestone

Development

Successfully merging this pull request may close these issues.

2 participants

@felixzinn @pfackeldey