fit_cpp: C++ function to fit interval censored models with elastic net...
In anujkhare/iregnet: Regularized interval regression

This function is not meant to be called on it's own! Please use the iregnet function which includes data validation and pre-processing.

1
2
3

fit_cpp(X, y, family, lambda_path, debug, out_status, intercept, alpha,
  scale_init, estimate_scale, unreg_sol, flag_standardize_x, max_iter,
  threshold, num_lambda, eps_lambda)

`X`	Design matrix.
`y`	Output matrix in a 2 column format. `NA`s denote censoring.
`family`	String denoting the distribution to be fit.
`lambda_path`	Vector containing the path of hyper-parameter lambda values.
`debug`	Integer used for debugging during development. Unused otherwise.
`out_status`	Vector containing censoring status of each observation. If not provided, it will be calculated using the `y` matrix.
`intercept`	If `true`, intercept is to be fit. The first column of X must be `1`s if `intercept` is `true`.
`alpha`	Hyper parameter for the elastic-net penalty.
`scale_init`	The initial value of `scale` to be used for the fit. If not provided, a value is calculated depending on the distribution.
`estimate_scale`	If `true`, `scale` is estimated. Else, the `scale` remains fixed at `scale_init`.
`unreg_sol`	If `true`, the last model fit will be unregularized, i.e., the final `lambda` value will be `0`. Only used if `lambda_path` is not specified.
`flag_standardize_x`	If `true`, the design matrix `X` will column-wise standardized.
`max_iter`	Maximum number of iterations to allow for each model (`lambda` value) along the regularization path.
`threshold`	Convergence detected if absolute change in a coefficient is less than this value.
`num_lambda`	Number of lambda values to use in the regularization path. Only used if `lambda_path` is not specified.
`eps_lambda`	Ratio between the maximum and minimum values of `lambda`. Maximum value of `lambda` is calculated based on the distribution and the data. Only used if `lambda_path` is not specified.