Repository navigation
docs: document conda install and training co-install - #67
Merged
Merged
Conversation
bentaculum
force-pushed
the
bentaculum/conda-install-docs
branch
from
July 21, 2026 22:05
942e29a to
1369a1f
Compare
bentaculum
changed the title
Add a conda-forge install line next to the pip one, and note the training-dependency co-install next to pip install trackastra[train] (the conda-forge package is core-only, conda-forge/trackastra-feedstock#9).
bentaculum
force-pushed
the
bentaculum/conda-install-docs
branch
from
July 21, 2026 22:06
1369a1f to
96be677
Compare
bentaculum
marked this pull request as ready for review
July 21, 2026 22:06
This file contains hidden or bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
Add this suggestion to a batch that can be applied as a single commit.This suggestion is invalid because no changes were made to the code.Suggestions cannot be applied while the pull request is closed.Suggestions cannot be applied while viewing a subset of changes.Only one suggestion per line can be applied in a batch.Add this suggestion to a batch that can be applied as a single commit.Applying suggestions on deleted lines is not supported.You must change the existing code in this line in order to create a valid suggestion.Outdated suggestions cannot be applied.This suggestion has been applied or marked resolved.Suggestions cannot be applied from pending reviews.Suggestions cannot be applied on multi-line comments.Suggestions cannot be applied while the pull request is queued to merge.Suggestion cannot be applied right now. Please check back later.
What
Documents installing Trackastra via conda in the README.
conda install -c conda-forge trackastraline next to the existingpip install trackastra.pip install "trackastra[train]", notes that the conda-forge package is core-only (feedstock#9) and gives the training-dependency co-install one-liner.Why
conda-forge/trackastra-feedstock#9 switches the conda package to ship the core/inference dependencies only (the
trainextras —lightning,wandb,tensorboard,configargparse,matplotlib,kornia,gitpython— are dropped from the hardrundeps). The README previously documented onlypip. Conda has no equivalent of pip'strackastra[train]extra, so the training deps are documented as a co-install. Docs-only change.Verification
kornia≥0.7,matplotlib), andtrackastraitself is published there.setup.cfg [options.extras_require] train, notinstall_requires.