Harsha
Harsha
I was trying to install nobrainer in different environments to check for consistency. All I did was create a new conda environment and installed nobrainer using `pip install nobrainer` 1....
- [Note](https://gablab.slack.com/archives/G01DLL64143/p1708004272198329) - [Code](https://github.com/lab-smile/GRACE) - [Paper](https://direct.mit.edu/imag/article/doi/10.1162/imag_a_00090/119208/Precise-and-rapid-whole-head-segmentation-from)
From @satra to a question of maintaining a mapping between models and storage siblings- "you can push to all get from first/closest"
It makes much more sense and is cleaner to use datalad checkout instead of git checkout as the siblings are not readily known with the latter.
Currently, the issue branch is retained in the remote repo and, (ideally) we want this branch to be deleted if the testing workflow fails. However, the key stumbling block is...
https://github.com/neuronets/trained-models/actions/runs/6762690478/job/18378961087#step:8:1 The input arguments don't match the order of arguments the function is expecting.
Partial workflow at https://github.com/hvgazula/trained-models/actions/runs/6806826215 barring pushing the branch and docker image on successful testing.
### Model's repository path PialNN/pialnn/1.0.0 ### Best model weights https://drive.google.com/uc?id=18KdBZLwd-8ld3SZoQCXAXwvrCtdeABU-&confirm=t ### Docker information https://github.com/gaiborjosue/DeepCSR-testing/tree/main/dockerP ### Model's python scripts https://github.com/gaiborjosue/DeepCSR-testing/blob/main/pial/predict.py https://github.com/gaiborjosue/DeepCSR-testing/blob/main/pial/utils.py https://github.com/hvgazula/pialnn-forked/tree/harsha/data https://github.com/hvgazula/pialnn-forked/tree/harsha/model https://github.com/gaiborjosue/DeepCSR-testing/blob/main/pial/config.py ### Model's card and spec https://github.com/gaiborjosue/trained-models-fork/blob/master/DeepCSR/deepcsr/1.0/spec.yaml https://github.com/gaiborjosue/trained-models-fork/blob/master/DeepCSR/deepcsr/1.0/model_card.yaml...
closes #102