Jan

Results 60 issues of Jan

**Describe the bug** When using VIPosterior with "mps" as the GPU device, and setting `q="nsf"` with `num_dim=1`, sampling from `q` leads to NaNs. **To Reproduce** To reproduce the issue, follow...

bug
blocked
hackathon
gpu-handling

once #928 is fixed, we need to test making a new release using only the git tag as version reference (details here: https://carpentries-incubator.github.io/python_packaging/instructor/05-publishing.html#extra-consistent-versioning-with-setuptools-scm)

documentation
less-urgent

For previous releases we crafted the `CHANGELOG.md` by hand. For the upcoming release this is a lot of work. So I looked into automated tools and came across: - https://www.conventionalcommits.org/en/v1.0.0/...

question

At the moment, we require `python>=3.8`, see `pyproject.toml`. However, the security support for 3.8 will end in October. I suggest that we change it to `python>=3.9` with the upcoming release...

question

there is some code repetition in these tests https://github.com/sbi-dev/sbi/blob/bb35def3341a0896a66b7038d9c02931f6797113/tests/metrics_test.py#L133-L172 which can be reduced by using `@pytest.mark.parametrize` (see other tests for examples.).

good first issue

When building the most recent `docs` I noticed several problems: Upon running `jupyter nbconvert --to markdown ../examples/*.ipynb --output-dir docs/examples/ && jupyter nbconvert --to markdown ../tutorials/*.ipynb --output-dir docs/tutorial/` as described in...

bug

Initiated by @tomMoral's input in #1019 we are planning to give more flexibility to users for defining their custom density estimator, by adding another layer of abstraction -- an `Estimator`...

enhancement

In line with the planned license change in [sbi](https://github.com/sbi-dev/sbi/pull/997), I suggest to change the license of pyknos to Apache 2.0 as well. Please see https://github.com/sbi-dev/sbi/pull/997 for details. Please indicate your...

The MDN class has several inline comments and TODOs that are unclear. We should refactor it and improve documentation.

enhancement

Hi there, thanks for this package, it's really helpful! On a cluster with multiple GPUs, I have my model on device `cuda:1`. When calculating FID with a passed `gen` function,...