''' install Pytorch 2.1.2 and torchvision 0.16.2 ''' conda install pytorch torchvision torchaudio pytorch-cuda=12.1 -c pytorch -c nvidia ''' Relevant dependencies have been described in requirement.txt, just install them as required. ''' conda install --yes --file requirements.txt '''
We use UBFC, NBHR, NICU-NPS．NICU-NPS will be made public after the paper is published. NBHR and UBFC are public databases.
We provide a few data samples for you to test demo.
You need to configure the data_root and save_path parameters in the clip_train.py file. After executing the training, you can get the adaptive lighting enhancement model parameter file in the save_path path. This model file will be used in end_to_end.py