EdwardRaff

Hah, not bad at all! Thats some nifty code actually! > Bagging - is there a weak classifier that in general can be assumed to be an "ok" starting point?...

The diabetes dataset is pretty hard and not a lot of data, your probably see bigger differences on larger datasets. MNIST could be a good benchmark Some of those, like...

> Errr... I'm not entirely sure. I got some that error with > "Network needs categorical attributes to work" > or "At least 2 categorical variables are needed for ODE"...

Hmm, so some things like LogisticRegression I didn't have implement BinaryScoreClassifier because they already produce calibrated probabilities, and the class was introduced so that methods that didn't have probabilities could...

HI Michael, do you have an example of this problem you could share by chance? Either way, it sounds like adding an infinity check to [this line](https://github.com/EdwardRaff/JSAT/blob/7d5286ba5e524c7ec612d1d2da045ed2a1146764/JSAT/src/jsat/regression/OrdinaryKriging.java#L129) fixed it for...

Please start with the [wiki page ](https://github.com/EdwardRaff/JSAT/wiki), which has some examples on using a classifier and loading data. You will simply need to change the imports/constructor from the example classifier...

Can you provide a link/reference to what you think is the correct way to do PAM?

I think what is in the paper is equivalent to the code I have for PAM. BUILD basically describes just the seed selection step, where the first seed is the...

My institution does not have access to Willey, do you have any other references on the details? I find the original PAM paper very hard to parse on what the...

Erich, I’m just not on the same page in understanding how the SWAP procedure works. If your paper has this in more explicit detail I’m happy to add it to...