Evaluate a trained model

After training a model or obtaining a trained model, we can readily get our model to be used for forward pass. For this matter, we provide a script to get the results for a trained model. We provide an example brain segmentation model for LPBA40 dataset in the following link: https://drive.google.com/file/d/1QYj-5T2mxSKEXX_V3lzZaE5v4hA8TCtt/view?usp=sharing

Our split can be found here (absolute paths needs to be fixed): https://drive.google.com/drive/folders/1b2oygVEFQ0xQb9-rBwUjO2H154D4YfcR?usp=sharing

Arguments:
    -m : the path to model file,
    -o  : the directory you want to have the uploads at.
    -i: absolute path to image
    -l: absolute path to label
    -ts : setting file for the evaluation
    -txt:

This script outputs the 3D segmentation to {-o}/seg/res/records/3D, under the name of {image_name}_test_iter_0_output.nii.gz.

An example arguments could be as following:

python eval_seg.py -m MODEL_PATH -o outputs -i s22.nii.gz -l s22.nii.gz -ts settings_for_lpba/seg_test/

OR
python eval_seg.py -m MODEL_PATH -o outputs -txt outputs/25case/test/file_path_list.txt -ts settings_for_lpba/seg_test/