Home - News ( United States | United Kingdom | Italy | Germany ) - Football scores

Showing content from https://github.com/tidymodels/parsnip/issues/195 below:

glmnet penalty parameter · Issue #195 · tidymodels/parsnip · GitHub

Right now, if a modeling function using glmnet gets a specific penalty value, the model cannot make predictions on any other values. If no value is given, the model can predict on anything but there is no default value to be used with predict().

I propose doing what caret does:

• if a penalty value is given, save it but do not pass that value to the lambda argument to glmnet.
• attach the supplied penalty value to the the glmnet fit object.
• when using predict(), use the value attached to the glmnet object.

It is suboptimal to modify the underlying object but that would enable use to have the best of both worlds; predict() works as expected (and without error) and multi_predict() can also be used.

jaredlander and mdancho84

RetroSearch is an open source project built by @garambo | Open a GitHub Issue

Search and Browse the WWW like it's 1997 | Search results from DuckDuckGo

HTML: 3.2 | Encoding: UTF-8 | Version: 0.7.4