Matthew R. Becker
Matthew R. Becker
This is a tracking issue to enable windows arm64. We think the path forward is to cross-compile and do out of band testing on native hardware. Let's put todo items...
We're having a lot of trouble with a token rotation right now (https://github.com/conda-forge/status/issues/137) due to travis API rate limits. Some of this was a poorly coded bot on my part...
This issue is for tracking our move to unique tokens per provider. - [ ] conda-smithy supports new per-provider tokens (https://github.com/conda-forge/conda-smithy/pull/1716, https://github.com/conda-forge/conda-smithy/pull/1717) - [x] conda-forge-ci-setup provides `provider` key in http...
We have deprecated drone, circle ci, and appveyor for various reasons. I propose that for conda-forge, we drop them for good, cleaning out all of the webhooks and the GitHub...
This PR is a start on some apodization options for Piff. Right now I have fixed things to avoid edge cases for adaptive algorithms.
This code doesn't follow sklearn APIs and conventions closely enough to use in things like pipelines etc.
The changes in the CI scripts in https://github.com/LSSTDESC/CCL/pull/912 likely need to be ported to release branches.
Right now we have a lot of C layer warnings which are - not descriptive - not correct - override other warnings so that the error is not clear at...
Right now we still have some strcpy stuff happening. We need to make sure to use the internal API.
There are potential problems with using `__del__` to manage C memory. We may want to try using a weak reference instead. http://code.activestate.com/recipes/577242-calling-c-level-finalizers-without-__del__/