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

Pyro backend VB #66

Open
pdh21 opened this issue Jul 14, 2020 · 1 comment
Open

Pyro backend VB #66

pdh21 opened this issue Jul 14, 2020 · 1 comment
Assignees

Comments

@pdh21
Copy link
Member

pdh21 commented Jul 14, 2020

Expanding to more complicated models is going to require moving to Variational inference. Pyro looks like the best backend to use to be able to implement this

@pdh21 pdh21 self-assigned this Jul 14, 2020
@pdh21
Copy link
Member Author

pdh21 commented Jul 14, 2020

Standard SPIRE model test

Time comparison

No subsampling

fit_pyro=SPIRE.all_bands([prior250,prior350,prior500],n_steps=500,lr=0.001)

CPU times: user 2min 9s, sys: 20.6 s, total: 2min 29s
Wall time: 1min 6s

with subsampling

fit_pyro=SPIRE.all_bands([prior250,prior350,prior500],n_steps=500,lr=0.001)

CPU times: user 1min 2s, sys: 7.24 s, total: 1min 9s
Wall time: 37.7 s

@pdh21 pdh21 mentioned this issue Jul 14, 2020
Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment
Projects
None yet
Development

No branches or pull requests

1 participant