samet-akcay / skip-ganomaly

Source code for Skip-GANomaly paper
MIT License
192 stars 64 forks source link

bad result of mnist training #9

Open songsu1008 opened 4 years ago

songsu1008 commented 4 years ago

hello, I trained the mnist file by following your code(change nothing), (type like this in terminal : ython train.py --dataset mnist --isize 32 --nc 1 --niter 1 --abnormal_class 0 --model skipganomaly ) but the result is not good like below figure.(pink dots are normal_labels, and skyblue dots are abnormal_labels.) how can I increase difference of the anomaly scores between normal and abnormal?

Figure_1