Adam Goodge

Results 4 comments of Adam Goodge

You mean reduce the weight of l2_loss? that would encourage the adversarial loss to decrease?

> Hi, Leng, I found that I used the pre-trained model to achieve 91.7% Where is the pre-trained model available?

Hi @suziyou12 We find nearest neighbors from only the training set because they are all normal points, instead of the test set where some points are anomalies.