-
-
Notifications
You must be signed in to change notification settings - Fork 258
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
Pathfinder model specification fix #674
Conversation
Check out this pull request on See visual diffs & provide feedback on Jupyter Notebooks. Powered by ReviewNB |
Sorry for the merge mess. Squash-merge should fix, yes? |
theta = pm.Normal("theta", mu=0, sigma=1, shape=J) | ||
theta_1 = mu + tau * theta | ||
z = pm.Normal("z", mu=0, sigma=1, shape=J) | ||
theta = mu + tau * z |
There was a problem hiding this comment.
Choose a reason for hiding this comment
The reason will be displayed to describe this comment to others. Learn more.
Ugh 🤦
There was a problem hiding this comment.
Choose a reason for hiding this comment
The reason will be displayed to describe this comment to others. Learn more.
ChatGPT?
The model still does a terrible job of fitting a very simple model. Hopefully the GSoC project will bear some fruit. |
Yup that was why I didn't bother to push the fix. It seem to be pretty terrible, but maybe it's just a hard model still. |
ADVI does a fine job, and I've seen Stan's implementation fit the same
model so I'm guessing it's a problem with the Black Jax version
specifically? Could be something as simple as bad defaults, who knows?
…On Thu, Jul 18, 2024, 11:00 AM Ricardo Vieira ***@***.***> wrote:
The model still does a terrible job of fitting a very simple model.
Hopefully the GSoC project will bear some fruit.
Yup that was why I didn't bother to push the fix. It seem to be pretty
terrible, but maybe it's just a hard model still.
—
Reply to this email directly, view it on GitHub
<#674 (comment)>,
or unsubscribe
<https://github.com/notifications/unsubscribe-auth/AAAT4RCGF3DQ2HODASYXDJLZM7RBRAVCNFSM6AAAAABLBVZQFSVHI2DSMVQWIX3LMV43OSLTON2WKQ3PNVWWK3TUHMZDEMZWHEZTAOJTHA>
.
You are receiving this because you authored the thread.Message ID:
***@***.***>
|
Re-mentioning the problem to @junpenglao @zaxtax ^ We should at least open an issue in blackjax? Pretty lame to not have it working :( |
Deals with the easy problem in #572, the mis-specification of the model. Also added an example
num_samples
argument that may be useful for new users.📚 Documentation preview 📚: https://pymc-examples--674.org.readthedocs.build/en/674/