The hardware and bandwidth for this mirror is donated by dogado GmbH, the Webhosting and Full Service-Cloud Provider. Check out our Wordpress Tutorial.
If you wish to report a bug, or if you are interested in having us mirror your free-software or open-source project, please feel free to contact us at mirror[@]dogado.de.
hnp_umbrella() has a new interface. It now takes
separate X and Y arguments together with
importance_order instead of a single data frame
S and a class_col. It supports an arbitrary
number of ordered classes (T >= 2) rather than only
ternary classification.hnp_summary() now takes classifier,
X, Y and importance_order instead
of data and class_col, and accepts classifiers
that return class labels, probability matrices or score matrices, as
well as fitted model objects.hnp_map_classes() now accepts a variable number of
class labels via ... (in decreasing priority order) instead
of the fixed class_1, class_2,
class_3 arguments.probability_to_score_1(),
probability_to_score_2(), hnp_umbrella_flex()
and hnp_box_plot().T >= 2).hnp_umbrella() via the
pretrained_model and input_is_score
arguments.grid_search, grid_set, max_grid,
max_combinations) to minimize the weighted
misclassification objective, with a recursive multi-class threshold
search.hnp_boxplot() to visualize and summarize
under-classification and overall error from confusion matrices,
supporting single- and two-method comparisons.gen_data(), gen_normal_data(),
generate_ball_data(), etc.) and a neural-network scoring
helper (train_nn_and_get_scores()).hnp_delta_search() now uses
stats::pbinom() instead of an explicit
choose()-based summation, avoiding numerical overflow and
underflow for large samples.hnp_upper_bound() is now vectorized over the score
functions and adds validation for non-finite, NA and
length-mismatched score outputs.These binaries (installable software) and packages are in development.
They may not be fully stable and should be used with caution. We make no claims about them.
Health stats visible at Monitor.