Error when running ADE20k pre-trained model

[jijeng]

Thanks for your wonderful work. I want to use your pre-trained model on my datasets.
using this command

python test.py --name ade20k --dataset_mode ade20k --dataroot ./imgs/ade20k --gpu_ids 0 --nThreads 0 --batchSize 6 --use_attention --maskmix --noise_for_mask --warp_mask_losstype direct --PONO --PONO_C --which_epoch 90

I thought it was not necessary to use ref mask when using a pre-trained model, so I removed all the .png images in imgs/ade20k. But there was an error

I wonder if I can use your pre-trained model if I don't have segmentation images ( masking images), thanks

[panzhang0212]

If you removed all the .png images in imgs/ade20k, how do you give the input mask(not ref mask) to network, which makes len(self.label_paths) = 0 , and raise out range error.

[jijeng]

if I removed the .png in ade20k/training or ade20k/validation, then there also some errors.

[panzhang0212]

I mean you couldn't delete all .png image. because some .png images are input mask. If you delete all .png images, the code could not find the input mask.