Some question for rerun the code

[ttttsjjjj]

Hello ,i am new in research.I have already set all the settings through READ ME file.But when i run the code by "sh main.sh" some problems occurs:DataParallel' object has no attribute 'data_root' .Here are my settings in main.sh: python main.py \ --model 'LASTED' \ --train_file 'annotation/Train_Photos_num199244.txt' \ --num_class 2\ --val_ratio 0.005 \ --test_file 'annotation/Test_MidjourneyV5_num2000.txt' \ --isTrain 1 \ --lr 0.0001 \ --resume '' \ --data_size 448 \ --batch_size 6\ --gpu '1,2' \ 2>&1 | tee weights/log.log If I miss something please make me know .Thank you !

[HighwayWu]

Thanks for your asking. By default, all the images should be placed in the 'data/' folder, which is also the default 'data_root'. Also, please provide a more detailed error screenshot for analyzing.

[ttttsjjjj]

Thanks for your reply!Here are my training images in the 'data/'folder,(i just use the dataset'RealPhoto_LSUN'and 'SyntheticPhoto_ProGAN' for training)

Also i have prepared the training txt file named "Train_Photos_num199244" under the folder"annotation"through preprocess.py.

Then i set the main.sh file like this:

Finally,i run this code by "sh main.sh"in the terminal.But some problems occurs: If i miss something please tell me.Thank you again for taking the time to reply !

[HighwayWu]

Sorry for the inconvenience. The reason is that the "model" was incorrectly read as "args" in line. The error has been corrected, please refer to the new "main.py".

[ttttsjjjj]

Thank you very much for your reply, the previous issue is solved! But now a new question arises: Looking forward for your reply. Thank you again!

[HighwayWu]

Sorry again for the inconvenience. The "images" should have only 4 dims, i.e., batch-size, channels, H, and W. The "rep" shoule be removed. Therefore, the line should be "b, C, H, W = images.shape", and there is no need to do ".reshape" in the next line. Relevant codes have been updated, and thanks for pointing out this issue.

[ttttsjjjj]

Thanks again for your reply, the previous issue is solved! But another small problem arises: (ps:You are really very helpful and your reply really helped me a lot, thank you again!)

[HighwayWu]

"val_score" was forgotten to be defined. Please correct it to a specific metric, such as "val_auc" or "val_ap".

[ttttsjjjj]

All problems solved! Thanks a lot.