Computer Vision News - February 2023

34 Spectral Imaging Tool test _ target _ data: 'None' # path to testing target data ('None' if N/A) shuffle: 'True' # shuffle data [True/False] seed: 'None' # random seed for data shuffling [integer value or 'None'] train _ split: 'None' # percentage of data used for training set [0.0 - 1.0 or 'None'] val _ split: 'None' # percentage of data used for validation set [0.0 - 1.0 or 'None'] test _ split: 'None' # percentage of data used for testing set [0.0 - 1.0 or 'None'] State _ Dicts: net _ state _ dict: 'None' # network state dict (if pretrained _ network != 'None') optimizer _ state _ dict: 'None' # optimizer state dict (if pretrained _ network != 'None') scheduler _ state _ dict: 'None' # scheduler state dict (if pretrained _ network != 'None') To evaluate or apply a pretrained model, you can also run: spectrai_evaluate and spectrai_ apply , in which case you need to make sure that your Training_Options and State_Dicts parameters in the config file contain information about the pretrained network! We also tried the Matlab GUI (requirement: Matlab v.2020b). For this, make sure you have the .mlapp file in your folder. Then go to the Matlab command window and setup the correct Python version: >> pyversion('/usr/local/bin/python3.8') >> pyenv ans = PythonEnvironment with properties : Version: " 3.8 " Executable: "/usr/local/opt/python@ 3.8 /bin/python3. 8 " Library: "/usr/local/Cellar/python@ 3.8 / 3.8 . 16 /Frameworks/ Python.framework/Versions/ 3.8 /lib/libpython3. 8. dylib"