[mlpack-git] [mlpack/mlpack] Adds a GammaDistribution object for estimating gamma parameters (#729)

[Ryan Curtin]

> +    /* Access to tolerance. */
> +    double Tol(void) const { return tol; };
> +    /* Change tolerance. */
> +    void Tol(double tolerance) { tol = tolerance; };
> +
> +    // Access to Gamma Distribution Parameters.
> +    /* Get alpha parameters of each dimension */
> +    arma::Col<double>& Alpha(void) { return alpha; };
> +    /* Get beta parameters of each dimension */
> +    arma::Col<double>& Beta(void) { return beta; };
> +
> +  private:
> +    arma::Col<double> alpha; // Array of fitted alphas.
> +    arma::Col<double> beta; // Array of fitted betas.
> +    arma::mat referenceSet; // Matrix of reference set.
> +    double tol; // Convergence tolerance.

I think both `referenceSet` and `tol` should be parameters to `Train()`, because they're only used for training.  I think it makes more sense if the class itself only holds the arrays of alphas and betas.

---
You are receiving this because you are subscribed to this thread.
Reply to this email directly or view it on GitHub:
https://github.com/mlpack/mlpack/pull/729/files/ee69680e03696c9abde8e722afb5babe6fab9a36#r71821627
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <https://mailman.cc.gatech.edu/pipermail/mlpack-git/attachments/20160721/465738fd/attachment.html>