ychfan / scn Goto Github PK

我在使用”python trainer.py --dataset div2k --eval_datasets div2k set5 bsds100 urban100 --model scn --num_blocks 32 --num_residual_units 128 --scale 4 --job_dir ./scn_x4_retrain” 这条命令进行训练后，程序显示GPU显存占用1000多M，但是没有显示进行到多少步，我在train.py程序中加了epoch的显示，但是一直卡在第一个epoch，请问训练大概时间是多久，如果不是训练很慢是哪里出了问题？

您好，我对这份工作很感兴趣，但因为是初学者，代码不是很熟，想请问您训练好了以后怎么做测试，谢谢！

When I start to train the scn models, it has bug and show "trainer.py: error: the following arguments are required: --num_scales ", but I could't find this parameter in the train.py. So how can I solve this problem?

Hi,
I am reading your paper, I am confused that why not compare your model with RCAN?

Could you please provide the pre-training model of SCN? I would like to compare it with my training results.thank you