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

investigate side-effects of pandas upgrade #77

Open
k-dominik opened this issue Oct 25, 2019 · 0 comments
Open

investigate side-effects of pandas upgrade #77

k-dominik opened this issue Oct 25, 2019 · 0 comments

Comments

@k-dominik
Copy link
Contributor

We found an issue (ilastik/ilastik#2147) related to a change of behavior of pandas.concat from 0.24.x to 0.25.

Previously different types of indices would be handled more gracefully, now if something like uint32 and int64 is encountered as an index, this will be cased to object. h5py will then in turn not serialize this (without help) and right now I am not sure which parts of the code rely on the index being an integer type - need to investigate this with more time.

see also https://github.com/ilastik/ilastikrag/blob/8812be124f589bf68272852fe91a69e86d0371be/ilastikrag/rag.py#L524, where the index for the features is constructed as uint32, and https://github.com/ilastik/ilastik/blob/4b76c50442ad561eaa579745524808a912f129f7/ilastik/applets/edgeTraining/opEdgeTraining.py#L283 where the dataset to be concatenated is constructed with the default index type.

Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment
Labels
None yet
Projects
None yet
Development

No branches or pull requests

1 participant