CVector | A C vector library similar to the C STL vector

You're trying to free memory you didn't malloc. And you don't free the one piece of memory you did malloc. In your case you probably meant:

What they call y_score is just the predicted probabilities outputted by your ML algorithm.

In multinomial nb and in a decision tree (I suppose that's what you mean by LearningTree?), you can do this with the method .predict_proba:

result = bar+foo; could also be written as result = bar.operator+(foo);.

So in the case of bar + foo we are calling the member function operator+ on bar and giving it the parameter foo.

We could write it as follows and it's the same principle.

Compiling with the flag -ffast-math results in fast performance.

Eigen uses a signed type to store sizes and for indexing. This type is Eigen::Index which by default is a typedef for std::ptr_diff. Just replace your size_t by Eigen::Index and while your are doing that, you can also replace your constructor implementations by these:

First in your Rotation for example in RotateZ you should save the x in some temporary because if you modify it & then try to use it for the y it's obviously gonna cause you an error, ie you should do something like this

This is a non standard g++ extension that is allowing g++ to accept the code. g++ allows you to declare a VLA (Variable Length Array) which is not part of the C++ standard. To make this code standard compliant you can move dim1 and dim2 into the template parameters and pass them as compile time constants like:

Yes, you need to free also vector->data, the rule is: one call to free per each call to malloc

if you are under C99, you can use flexible array members:

Almost all of statsmodels and all the inference is designed for the case when the number of observations is much larger than the number of features.

Logit.fit_regularized uses an interior point algorithm with scipy optimizers which needs to keep all features in memory. Inference for the parameters requires the covariance of the parameter estimate which has shape n_features by n_features. The use case for which it was designed is when the number of features is relatively small compared to the number of observations, and the Hessian can be used in-memory.

GLM.fit_regularized estimates elastic net penalized parameters and uses coordinate descend. This can possibly handle a large number of features, but it does not have any inferential results available.

Inference after Lasso and similar penalization that select variables has only been available in recent research. See for example selective inference in Python https://github.com/selective-inference/Python-software for which also a R package is available.