The subdirectory emulationResults/trainedParameters
stores pretrained emulation parameters. The commands below show how to use these parameters to make predictions on the test data.
Predictions can be run as follows:
python -m main --mode="evaluate" --data_path="TwistingBeam" --dir_label="_preTrained" --trained_params_dir="emulationResults/trainedParameters/TwistingBeam/"
Predictions can be run as follows:
python -m main --mode="evaluate" --data_path="Liver" --dir_label="_preTrained" --trained_params_dir="emulationResults/trainedParameters/Liver/"
Predictions can be run as follows:
python -m main --mode="evaluate" --data_path="LeftVentricle" --dir_label="_preTrained" --trained_params_dir="emulationResults/trainedParameters/LeftVentricle/"
Note, as discussed in the README
, this is a slightly different version of the LeftVentricle data that was used in the paper.