I do not know why exactly this is occurring, but when using the PcaAD method and the same K and c values, I am receiving different outputs for labeled anomalies. Has anyone else experienced this?
It almost appears as if there is randomness being incorporated. I believe the sklearn library PCA does include random_state as one of the parameters due to the randomness used in it's SVD. See link: https://gregorygundersen.com/blog/2019/01/17/randomized-svd/ .
Is there a random state parameter? I did not see one in the documentation.
Hello,
I do not know why exactly this is occurring, but when using the PcaAD method and the same K and c values, I am receiving different outputs for labeled anomalies. Has anyone else experienced this?
It almost appears as if there is randomness being incorporated. I believe the sklearn library PCA does include random_state as one of the parameters due to the randomness used in it's SVD. See link: https://gregorygundersen.com/blog/2019/01/17/randomized-svd/ .
Is there a random state parameter? I did not see one in the documentation.