Без опису

Yaroslav Halchenko b32a971f29 Place desired inputs listing into external file 10 місяців тому
.datalad d6959d2965 [DATALAD] new dataset 1 рік тому
.github 979ba07add upload build artifact 10 місяців тому
.reproman c8e0b80f21 [ReproMan] save results for 20230706-100912-74dd 11 місяців тому
code 7c205993b8 Add figure generation deps 10 місяців тому
inputs 6c23921c0f restore inputs/mousebraintemplates to latest 10 місяців тому
outputs 7c83e8a9c2 Fix podman run \ 1 рік тому
paper 9d499af050 Add list of things-to-do-differently 10 місяців тому
reproman d03d61b91e reproman progress 11 місяців тому
.gitattributes 1500ae9b41 commit fonts directly to git to simplify life 10 місяців тому
.gitignore 84dd63be0e ignore other article side artifacts 10 місяців тому
.gitmodules ce6da4d909 .git at the end of url upsets git annex 11 місяців тому
Makefile 1af3cf09be Add explanatory "all" rule and "article" rule 10 місяців тому
README.md 5e496b7113 Add gin repo for mouse-brain-templates 10 місяців тому
inputs.txt b32a971f29 Place desired inputs listing into external file 10 місяців тому
yoh-notes 05d6e837c1 WIP reproman 11 місяців тому

README.md

Opfvta

This is a yoda-style[TODO cite] repository that contains all of the code, data, configuration, and prose to generate an article replicating the [TODO original cite].

How to clone this repository

This repository (and all yoda-style repos) contain. submodules/subdatasets.

datalad install -r https://github.com/con/opfvta-replication-2023.git cd opfvta-replication-2023

Some of the subdatasets/submodules are git-annex enabled, which means that we can use datalad to retrieve the files (which are just references, no "bits").

TMP: Add smaug remote (need access):

SshConfig

Host smaug smaug.dartmouth.edu
  Hostname smaug.dartmouth.edu
  AddKeysToAgent yes
  port 11110
  user <username>

git remote add smaug smaug:/mnt/btrfs/datasets/incoming/con/opfvta-replication-2023.git datalad get . -s smaug

git remote add gin https://gin.g-node.org/TheChymera/mouse-brain-templates datalad get . -s gin

Run datalad get . for each git-annex enabled repository below:

  • /opfvta-replication-2023/code/opfvta
  • /opfvta-replication-2023/inputs/opfvta_bidsdata

How to build and use the images

Analysis image

This container executes all parts of the original work:

- preprocessing
- analysis
- data visualization rendering
- latex rendering into original article with updated data

Configure the variables at the top of the root level makefile (provided values are what we used). Then the OCI image can be built with:

make oci-build

The OCI container can be run with the following command. We require the scratch directory to be explicitly specified, as it will end up containing large amounts of data (in excess of 200 GB), and any default might accidentally clutter a difficult to locate directory.

OPFVTA_SCRATCH_DIR="/your/cache/dir" make oci-run

The container image can be pushed to a container registry:

make push

You may need to build a singularity/apptainer image with the following command:

make apptainer-build

After the apptainer image is built, it will need to be committed and the new bits pushed to a git annex enabled repository.

LaTeX builder image

This container renders the LaTeX for the replication paper. The image is blang/latex with some extra dependencies.

make build-latex

The paper can be rendered into a pdf with:

cd paper/source/
make article