To execute the view aggrgation, provide the necessary paths and configuration settings through command line arguments:
--out_path
: path of the place create the view aggregation models.--npz_path
: path of the place with inference files to use.
Example to use:
python3 utils/create_2_D.py --out_path /localmount/volume-hd/users/uline/segmentation_results/miccai_2017/transfer_lr/test_set \
--npz_path /localmount/volume-hd/users/uline/segmentation_results/miccai_2017/transfer_lr/out
For the output file this structure needs to exists
Outfolder
├── 69c2dddd-cd4a-4501-887a-f7d6899d91e3
│ └── lesion.nii.gz
├── 790f3bed-e979-47d3-8c52-6ae84eb9a0ab
│ └── lesion.nii.gz
└──18978ce2-7faa-46e6-9ab6-36e931712365
└── lesion.nii.gz
Also the npz folder file needs this structure
out_dir
├── UNET_axial
├── UNET_sagital
├── UNET_coronal
├── CNN_axial
├── CNN_sagital
└── CNN_coronal
To execute the view aggrgation, provide the necessary paths and configuration settings through command line arguments:
--eval_path
: path of the place create a evaluation path
Example to use:
python3 evaluatefrom_dir.py --eval_path /localmount/volume-hd/users/uline/segmentation_results/miccai_2017/transfer_lr/test_set