Problematic casting of integers in convert_observed_data
#7331
Comments
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
{{ message }}
Problematic casting of integers in
Description
When we use
pytensor.config.floatX=="float32", integer data is downcast to"int16"which has a pretty limited range of 32k. For count-based likelihoods this is way too narrow. I am not sure we should be doing anything with integers to begin with. Why are PyTensor casting rules (and customization flags) not sufficient?The text was updated successfully, but these errors were encountered: