fastMRI Knees Multicoil

Training/Testing

Important

The fastMRI datasets are natively supported in atommic. Therefore, you do not need to create a custom dataset class. You just need to set the dataset_format argument in the configuration file to fastMRI. Also the FFT needs to be centered. For example:

model:
    fft_centered: true
    fft_normalization: ortho

train_ds:
    dataset_format: fastMRI
    fft_centered: true
    fft_normalization: ortho

validation_ds:
    dataset_format: fastMRI
    fft_centered: true
    fft_normalization: ortho

test_ds:
    dataset_format: fastMRI
    fft_centered: true
    fft_normalization: ortho

For training a model, you just need to set up the data and export paths to the configuration file of the model you want to train. In train_ds and validation_ds please set the data_path to the generated json files. In exp_manager please set the exp_dir to the path where you want to save the model checkpoints and tensorboard or wandb logs.

You can train a model with the following command:

atommic run -c /projects/REC/fastMRIKneesMulticoil/conf/train/{model}.yaml

For testing a model, you just need to set up the data and export paths to the configuration file model you want to test. In checkpoint (line 2) set the path the trained model checkpoint and in test_ds please set the data_path. In exp_manager please set the exp_dir to the path where the predictions and logs will be saved.

You can test a model with the following command:

atommic run -c /projects/REC/fastMRIKneesMulticoil/conf/test/{model}.yaml

Note: The default logger is tensorboard.