GetTrainData

GetTrainData[data,batch size,patch]

creates a training batch of size batch size with patch size patch. The input data can be out of memory in the form of a list of "*wxf" files that contain the data, segmentation and voxel size or a list of "*.nii" files in the form {{"data.nii", "segmentation.nii"}..}. The input data can be in memory in a list in the form {{data, segmentation, vox}..}

GetTrainData[data,batch size,patch,nClass]

If nClass is set to an value n > 0 the segmentations are decoded in n classes. If MaskedPretraining->True the target becomes the clean data itself and the input is masked with noise filled blocks, for self supervised pretraining.

Details
Details and Options Details and Options
QMRITools`
QMRITools`

GetTrainData

GetTrainData[data,batch size,patch]

creates a training batch of size batch size with patch size patch. The input data can be out of memory in the form of a list of "*wxf" files that contain the data, segmentation and voxel size or a list of "*.nii" files in the form {{"data.nii", "segmentation.nii"}..}. The input data can be in memory in a list in the form {{data, segmentation, vox}..}

GetTrainData[data,batch size,patch,nClass]

If nClass is set to an value n > 0 the segmentations are decoded in n classes. If MaskedPretraining->True the target becomes the clean data itself and the input is masked with noise filled blocks, for self supervised pretraining.

Details

  • The following options can be given:
  • PatchesPerSet1PatchesPerSet is an option for GetTrainData. Defines how many random patches per dataset are created within the batch.
    AugmentDataTrueAugmentData is an option for GetTrainData and TrainSegmentationNetwork. If set True the training data is augmented. It can also be set to "2D" or "3D" to control if augmentation is done through plane or only in-plane.
    PadDataFalsePadData is an option for GetTrainData and TrainSegmentationNetwork.. If set to an integers the that number of slices on the top and bottom of the data can be made 0. This is done to learn cut of datasets.
    MaskedPretrainingFalseMaskedPretraining is an option for GetTrainData and TrainSegmentationNetwork. If set to True self supervised masked-pretraining data/mode is used instead of segmentation data: the target becomes the clean data itself and the input is masked with noise filled blocks.