Mock Annotation and PyTorch Training on Copick Data

A solution that creates mock annotations based on multilabel segmentation, trains a PyTorch segmentation model in steps, and generates predictions.

Arguments

--copick_config_path

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

--voxel_spacing

Voxel spacing used to scale pick locations. (default value: PARAMETER_VALUE)

--tomo_type

Tomogram type to use for each tomogram. (default value: PARAMETER_VALUE)

--embedding_name

Name of the embedding features to use. (default value: PARAMETER_VALUE)

--input_session_id

Session ID for the input segmentation. (default value: PARAMETER_VALUE)

--input_user_id

User ID for the input segmentation. (default value: PARAMETER_VALUE)

--input_label_name

Name of the input label segmentation. (default value: PARAMETER_VALUE)

--num_annotation_steps

Number of annotation steps to perform. (default value: PARAMETER_VALUE)

--run_name

Name of the run to process. (default value: PARAMETER_VALUE)

--output_dir

Directory to save trained models. (default value: PARAMETER_VALUE)

--random_seed

Random seed for reproducibility. (default value: 17171)

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