Xiao-Feng

The picture format of original miniimage dataset just like 0153282900005.jpg，i try to get the format of n01532829_5.jpg according to the guidance.

Ok,thank you,that means we must use the whole ImageNet dataset？

I have solve this question.

sorry,the modifyed is "west_image_path = os.path.join(config['path']['Test_Result_SR'], file_name+'.jpg')"

i try to use the model test the image given,but how we write a ".yml" file to inference new images?

i use the trained model_final.pth or model_step59999.pth to inference,but get the wrong result,do you know what wrong with it? 

i meet the same problem as you,have you solved it?@yyyyojiro

If i need to build the registry by myself?

可以通过以下指令可视化： python tools/detection/test.py $your-config$ $your-checkpoint$ --show-dir $your-dir$可视化测试集中的结果。