You signed in with another tab or window. Reload to refresh your session.You signed out in another tab or window. Reload to refresh your session.You switched accounts on another tab or window. Reload to refresh your session.Dismiss alert
Note that the Dirichlet distribution is "degenerate" as it has one of the parameters that is zero. However generally the output from the sampling method is just a zero in the corresponding position, while with that specific seed it gives NaN.
The text was updated successfully, but these errors were encountered:
It seems that some seeds produce nans when sampling from a Dirichlet distribution. Any idea why? Example script below that was tested on Google Colab.
Note that the Dirichlet distribution is "degenerate" as it has one of the parameters that is zero. However generally the output from the sampling method is just a zero in the corresponding position, while with that specific seed it gives NaN.
The text was updated successfully, but these errors were encountered: