sanghyun-son / EDSR-PyTorch

PyTorch version of the paper 'Enhanced Deep Residual Networks for Single Image Super-Resolution' (CVPRW 2017)
MIT License
2.44k stars 671 forks source link

Own model #227

Closed shakib-rahimi closed 4 years ago

shakib-rahimi commented 5 years ago

Hello, I would like to improve the model with new pictures of faces, how can I do that? Should my dataset have a specific shape?


I trained the model again with DIV2K. with this command:

python main.py - template GAN --scale 4 --save edsr_gan --reset --patch_size 96 --loss 5 VGG54 + 0.15 GAN --dir_data / home / rahimi / projects / dataset

How can I use the trained model now? I have used the following command but unfortunately it does not work: python main.py --model edsr_gan --data_test demo --scale 4 --test_only --save_results

Thank you very much Feedback geben Verlauf Gespeichert Community

momoda-mht commented 4 years ago

you had better add the parameter ' --pre_train', and you need to make sure that your model‘s location is right