Thank you for your great work!
I just wanted to ask if preparing the dataset - validation, training, test sets are necessary if one just want to test the pretrained model. So are the validation, training , test datasets just for training? Or do we also need them also for training
-- answering my own question - if you use the code in the dockers, no need to prepare csv file. Just follow the instruction and run it as explained in show_instruction.py
Thank you for your great work! I just wanted to ask if preparing the dataset - validation, training, test sets are necessary if one just want to test the pretrained model. So are the validation, training , test datasets just for training? Or do we also need them also for training
Thank you and have a nice day!