danieltan07 / dagmm

My attempt at reproducing the paper Deep Autoencoding Gaussian Mixture Model for Unsupervised Anomaly Detection

danieltan07/dagmm Issues

Model without training works well?
Updated a year ago
nan values during training
Updated a year ago2
Select random data while train and test?Is that suitable?
Updated a year ago1
why only attack data during training
Updated a year ago2
Can anyone run this codes?
Updated a year ago2
Reconstruction Loss issue
Updated a year ago1
Problem with Shuffling
Updated a year ago1
The training data in the code is the same as the test data?
Updated a year ago1
Good results even with random initialization
Updated 2 years ago9
backward not implemented error
Updated 3 years ago9
cholesky_cpu: U(2,2) is zero, singular U.
Updated 3 years ago3
negative probability
Updated 4 years ago1
change the AE model every time when the dataset changes?
Updated 4 years ago
requesting an open-source license (eg. MIT)
Updated 4 years ago
Large standard deviation when reproducing experiment results
Updated 4 years ago
KDDCUP-99 dataset discrepancy
Closed 5 years ago3
use csv file instead of npz
Updated 5 years ago
Why the loss value does not decrease
Closed 5 years ago
Why the predict samples of high energy as anomalies?
Updated 5 years ago1
Train loss is not converged.
Closed 5 years ago
run the code without GPU devices?
Updated 5 years ago3
data_loader KDD99Loader inherit problem
Updated 5 years ago
Unable to reproduce reported performance
Updated 6 years ago6
About max_val
Updated 6 years ago2
Hello!
Closed 6 years ago1