In Python XGBoost one can provide weights for each row of the data, see http://xgboost.readthedocs.io/en/latest/python/python_api.html#xgboost.XGBClassifier.fit. I tried to look for a way to specify such weights in SharpLearning, but could not find it. Is this possible?

This issue documents the way to use this package for Nested Cross-Validation. If you have any question, welcome to comment below.

Flat cross-validation vs. nested cross-validation

To clarify the meaning of these two terms in this specific issue, let me first describe them.

Flat cross-validation

Let us use 5-Fold as an example. In a 5-Fold flat cross-validation, you split the dataset

This came up when I was trying to use the weights.psis function to actually make an importance sampler. The weights that are returned are not the log-PSIS weights, but rather the log-PSIS weights minus the largest log ratio. @avehtari documented the solution here. https://discourse.mc-stan.org/t/trouble-using-loo-psis-for-importance-sampling/9209/2

This is not super urgent, but it would

Hi,

I was wondering if there's a mistake in the FAQ section of the documentation ("Should I apply ICA first or autoreject first?").
According to the MNE docs the reject parameter in ica.fit() only applies to an instance of Raw, not Epochs.
However, here ica.fit() is used with reject on epochs several times, e.g.:

>>> reject = get_rejection_threshold(epochs)
>>> ica.fit(epo

not just compare different feature sets with a fixed builtin model, but users can also input a model of their own choice. it does not limit exploration of new models or pipelines - they can use implementation of best practices while evaluating the such new models on features of their choice