Train UNet Model using MONAI with Multiple Runs and MLflow

Train a UNet model to predict segmentation masks using MONAI from multiple runs with MLflow tracking.

Arguments

--copick_config_path

Path to the Copick configuration JSON file. (default value: PARAMETER_VALUE)

--run_names

Comma-separated list of Copick run names to process. (default value: PARAMETER_VALUE)

--voxel_spacing

Voxel spacing to be used. (default value: PARAMETER_VALUE)

--tomo_type

Type of tomogram to process. (default value: PARAMETER_VALUE)

--seg_type

Type of segmentation labels to use. (default value: PARAMETER_VALUE)

--num_epochs

Number of training epochs. (default value: 100)

--batch_size

Batch size for training. (default value: 4)

--learning_rate

Learning rate for the optimizer. (default value: 0.0001)

--experiment_name

Name of the MLflow experiment. (default value: PARAMETER_VALUE)

--debug

Enable debugging output. (default value: 0)

Please follow this link for details on how to install and run this solution.