Loïc Estève

MNT Handle NaNs in scipy dev rankdata

3

One of the failure in the scipy-dev build https://github.com/scikit-learn/scikit-learn/issues/23626 (turns out it fixes all the scipy-dev failures for some reason ...) This is a change in scipy 1.10.dev: https://github.com/scipy/scipy/pull/16140 ```py...

Support for full pip requirements.txt syntax in environment.yml?

11

This may be stretching conda-lock a bit, but would there be a chance to support full pip requirements.txt syntax in environment.yml like the following one? This is supported by conda...

on main: conda-lock install fails with "Cannot find platform in lockfile" if lock filename does not end with conda-lock.yml

3

On commit 67f8da0 (latest commit at the time of writing). This is not a real blocker since I can make sure conda lock files end with `conda-lock.yml` but it is...

subprocess stdout capture does not seem to work on Windows

2

ipykernel is capturing stdout from subprocesses since ipykernel 6.0.0 according to the [changelog](https://github.com/ipython/ipykernel/blob/main/CHANGELOG.md#600). On Windows it does not seem like the output of subprocesses is correctly captured. Here is a...

Consider adding links to Google colab?

12

Just a thought (IIRC I saw that in a semi-recent tweet by Jake van der Plas) if you a notebook on github e.g.: https://github.com/sphinx-gallery/sphinx-gallery.github.io/blob/master/notebooks/auto_examples/sin_func/plot_sin.ipynb You can prepend `https://colab.research.google.com/github` and get...

Yeah first bug report ;-)! I thought I tried on scikit-learn before the release but actually I did not try configuring binder fully ... Note that there is a work-around...

Run the generated notebooks as part of the tests

4

We generate notebooks but we never make sure they run. It would be nice to be able to run them and make sure that there is no exception. From a...

WIP: Add unified test suite infrastructure

2

Closes #8. The idea is that there is a lot of tests that we would want to run for all the `FooCluster` classes. At the moment there is some duplication...

Make submission script more human readable

5

```py from dask_jobqueue import SLURMCluster cluster = SLURMCluster(cores=1, memory='1GB') print(cluster.job_script()) ``` ``` #!/usr/bin/env bash #SBATCH -J dask-worker #SBATCH -n 1 #SBATCH --cpus-per-task=1 #SBATCH --mem=954M #SBATCH -t 00:30:00 /home/lesteve/miniconda3/bin/python -m distributed.cli.dask_worker...

Left-over from #307. From https://github.com/dask/dask-jobqueue/pull/307#issuecomment-531300128: > it would be great if we could find a way not to have to cd in a different folder (jobqueue scheduler-specific, mostly PBS and...