[mlpack-git] [mlpack/mlpack] Added the discretize function. Design changes as discussed. (#703)

Marcus Edel notifications at github.com
Sun Aug 14 11:44:42 EDT 2016


> +  FtrDim = arma::vec(2);
> +
> +  const size_t shrink = this->params.Shrink();
> +  const size_t pSize = this->params.PSize();
> +  const size_t numCell = this->params.NumCell();
> +  const size_t numOrient = this->params.NumOrient();
> +  const size_t nColorCh = params.RGBD() == 0 ? 3 : 4;
> +  const size_t nCh = nColorCh + 2 * (1 + numOrient);
> +  FtrDim[0] = std::pow((pSize / shrink) , 2) * nCh;
> +  FtrDim[1] = std::pow(numCell , 2) * (std::pow (numCell, 2) - 1) / 2 * nCh;
> +}
> +
> +
> +template<typename MatType, typename CubeType>
> +void StructuredForests<MatType, CubeType>::
> +DistanceTransform1D(const arma::vec& f, const size_t n, const double inf,

Is this function from the "Distance Transforms of Sampled Functions" paper? If that's the case, we should cite the paper, as we did here: https://github.com/mlpack/mlpack/blob/master/src/mlpack/methods/ann/layer/dropout_layer.hpp. Also do you think, we could define inf inside the function and probably use DBL_MAX and -DBL_MAX instead of 999999.99?

-- 
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/703/files/72eb1ef22c7db1ea33af3de1cd043cdb277ec562#r74704039
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <https://mailman.cc.gatech.edu/pipermail/mlpack-git/attachments/20160814/34f5b929/attachment.html>


More information about the mlpack-git mailing list