neuronets / trained-models

Trained TensorFlow models for 3D image processing
https://neuronets.dev/trained-models
22 stars 15 forks source link

cleanup singularity test command in new_model.yml and update_model.yml #95

Closed hvgazula closed 1 year ago

hvgazula commented 1 year ago

Note that the path bindings (--bind) in the singularity calls in https://github.com/neuronets/trained-models/blob/8b4d3422f3d1a4beb5737026ca00111b1093ff37/.github/workflows/new_model.yml#L519 and https://github.com/neuronets/trained-models/blob/8b4d3422f3d1a4beb5737026ca00111b1093ff37/.github/workflows/update_model.yml#L449 are redundant. The destination /output is never used in the test command. See below for why this works- image

For more info, please refer to the outputs of the successful workflow runs on a simplified version of new_model.yml.

hvgazula commented 1 year ago

The suggested cleaner version is

singularity exec --nv ./${{ needs.create_new_branch.outputs.MODELPATH }}/docker/${{ needs.build-docker.outputs.MODELNAME }}.sif ${{ needs.create_new_branch.outputs.TESTCOMMAND }}

@gaiborjosue While I am very sure that this suggestion won't fail, if you don't mind, could you please test it for me?

gaiborjosue commented 1 year ago

Hello @hvgazula, sure I will test it.