Train one model
COPY_biglasso_part(X, y.train, ind.train, ind.col, covar.train, family,
lambda, center, scale, resid, alpha, eps, max.iter, dfmax, warn, ind.val,
covar.val, y.val, n.abort, nlam.min, b0, base.train, base.val)A named list with following variables:
A vector of intercepts, corresponding to each lambda.
The vector of coefficients that minimized the loss on the validation set.
A vector of length nlambda containing the number of
iterations until convergence at each value of lambda.
The sequence of regularization parameter values in the path.
Either "gaussian" or "binomial" depending on the
function used.
Input parameter.
A vector containing either the residual sum of squares
(for linear models) or negative log-likelihood (for logistic models)
of the fitted model at each value of lambda.
A vector containing the loss for the corresponding validation set.
The number of observations used in the model fitting. It's equal
to length(row.idx).
The number of dimensions (including covariables, but not the intercept).
The sample mean vector of the variables, i.e., column mean
of the sub-matrix of X used for model fitting.
The sample standard deviation of the variables, i.e.,
column standard deviation of the sub-matrix of X used for model
fitting.
The response vector used in the model fitting. Depending on
row.idx, it could be a subset of the raw input of the response vector
y.
The indices of features that have 'scale' value greater
than 1e-6. Features with 'scale' less than 1e-6 are removed from
model fitting.