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.

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

Method Error when trying to estimate degree of freedom with TDist #673

Closed
LukasUlrych opened this issue Feb 4, 2019 · 3 comments
Closed

Method Error when trying to estimate degree of freedom with TDist #673

LukasUlrych opened this issue Feb 4, 2019 · 3 comments

Comments

@LukasUlrych
Copy link

Hi,
the following code throws an error (on Julia 1.0.3)

using Turing y = rand(TDist(4),100) @model est_TD(y) = begin nu ~ LogNormal(0,2) [y[i] ~ TDist(nu) for i = 1:length(y)] end out_HMC = sample(est_TD(y),HMC(500,1e-4,5))

It seems that it has something to do with differentiation. Is there something wrong with trying to estimate the degree of freedom from the data? Has anyone encountered this before and/or found some workaround?

I will be thankful for any help
@mohamed82008
Copy link
Member

Looks like another one of #661. Requires a fix like #664.

@mohamed82008
Copy link
Member

A bit of context: StatsFuns.jl uses ccall to find the logpdf of a number of distributions. ccall doesn't support AD, so it needs to be worked around.

@mohamed82008 mohamed82008 mentioned this issue Feb 4, 2019
Closed
@yebai
Copy link
Member

yebai commented Mar 1, 2019

Closed in favour of #674

@yebai yebai closed this as completed Mar 1, 2019
Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment
Assignees
No one assigned
Labels
None yet
Projects
None yet
Milestone
No milestone
Development

No branches or pull requests
3 participants
@yebai @mohamed82008 @LukasUlrych