LSSTDESC / td_env

Sets up td_env at NERSC and deploys docker images to dockerhub
BSD 3-Clause "New" or "Revised" License
2 stars 1 forks source link

Add packages for elasticc #53

Closed heather999 closed 2 years ago

heather999 commented 2 years ago

See Slack

lalsuite (https://anaconda.org/conda-forge/lalsuite; https://pypi.org/project/lalsuite/) ligo.skymap (https://anaconda.org/conda-forge/ligo.skymap; https://pypi.org/project/ligo.skymap/) ligo.em-bright (https://anaconda.org/conda-forge/ligo.em-bright; https://pypi.org/project/ligo.em-bright/) gwpy (https://anaconda.org/conda-forge/gwpy; https://pypi.org/project/gwpy/) gwemlightcurves (https://pypi.org/project/gwemlightcurves/)

heather999 commented 2 years ago

Test install revealed that ligo.em-bright has a hard requirement on scikit-learn >=0.24,<0.25 The LSST Sci Pipelines don't have a strict requirement on scikit-learn, and we were on a recent release scikit-learn=1.0.2, so we'll need to downgrade to include ligo.em-bright. At some point, we'd probably want to encourage the maintainers of ligo.em-bright to loosen that restriction and support the more recent scikit-learn releases - because it might become problematic in the future.

heather999 commented 2 years ago

Done!