[MLPerf] UNet3D dataloader (#4343)

* add support for train/val datasets for kits19

* split dataset into train and val sets

* add tests for kits19 dataloader

* add MLPerf dataset tests to CI

* update unet3d model_eval script

* fix linting

* add nibabel

* fix how mock dataset gets created

* update ref implementation with permalink and no edits

* clean up test and update rand_flip implementation

* cleanups
This commit is contained in:
Francis Lata
2024-04-28 22:34:18 -04:00
committed by GitHub
parent 82d0ed3cf3
commit bb849a57d1
6 changed files with 352 additions and 11 deletions

View File

@@ -232,6 +232,9 @@ jobs:
- if: ${{ matrix.task == 'onnx' }}
name: Test MLPerf metrics
run: GPU=1 python -m pytest -n=auto test/external/external_test_metrics.py --durations=20
- if: ${{ matrix.task == 'onnx' }}
name: Test MLPerf datasets
run: GPU=1 python -m pytest -n=auto test/external/external_test_datasets.py --durations=20
- if: ${{ matrix.task == 'onnx' }}
name: Test THREEFRY
run: PYTHONPATH=. THREEFRY=1 GPU=1 python3 -m pytest test/test_randomness.py test/test_jit.py