[GitHub] madlib pull request #243: MLP: Add minibatch gradient descent solver

[kaknikhil]

Github user kaknikhil commented on a diff in the pull request:

    https://github.com/apache/madlib/pull/243#discussion_r175872591
  
    --- Diff: src/modules/convex/task/mlp.hpp ---
    @@ -111,6 +117,57 @@ class MLP {
     template <class Model, class Tuple>
     double MLP<Model, Tuple>::lambda = 0;
     
    +template <class Model, class Tuple>
    +double
    +MLP<Model, Tuple>::getLossAndUpdateModel(
    +        model_type           &model,
    +        const Matrix         &x_batch,
    +        const Matrix         &y_true_batch,
    +        const double         &stepsize) {
    +
    +    uint16_t N = model.u.size(); // assuming nu. of layers >= 1
    +    size_t n = x_batch.rows();
    --- End diff --
    
    is there a reason we chose N and n as variable names ? Can we use more 
descriptive names ?
    



---