Open MammadTavakoli opened 7 years ago
hey, I have another question: evaluate.py [-h] [--model MODEL] [--top {patchwise,weighted}] [--gpu GPU] INPUT [REF] So, what is trained model ???? I can't get it ? :(
hey, I have another question: evaluate.py [-h] [--model MODEL] [--top {patchwise,weighted}] [--gpu GPU] INPUT [REF] So, what is trained model ???? I can't get it ? :(
For Full-Reference Image Quality Assessment you have to specify a reference image such as "python evaluate.py --model models/fr_tid_weighted.model --top weighted Name_of_InputImage.JPG Name_of_ReferenceImage.JPG". If you don't have a reference image, you have to load a no reference model, f.e. such that "python evaluate.py --model models/nr_tid_weighted.model --top weighted Name_of_InputImage.JPG".
How you compute PLCC and SROCC ?