Mariusz Musiał

The following code from DataCamp's Intermediate Portfolio Analysis in R course fails in optimize.portfolio function of PortfolioAnalytics_1.1.0: ``` asset_names

PLS regression with mixOmics engine - predictions not extracted from predict object

5

Hi, `vi_permute` fails on PLS perdict object. ``` r library(plsmod) # parsnip helper for pls #> Loading required package: parsnip library(mixOmics) # for pls regression #> Loading required package: MASS...

That would be helpful when dealing with large number of predictors, or smaller numbers with some being unimportant and just taking up the space. This would also homogenize a bit...

Running `predict_parts` with type set to oscillations of any kind results in a set of warnings of an argument being neither numeric or logical, with all oscillations in the result...

Hi, trying to load and transform png images using load_image function based on one of torchvision's vignettes ``` load_image % transform_grayscale(num_output_channels = 1) %>% transform_resize(c(105, 105)) x sessionInfo() R version...

Error: C stack usage xxx is too close to the limit, with certain input to pin_write metadata parameter

2

The following code causes Error: C stack usage xxx is too close to the limit: ``` mbm %>% pin_write(wfset_fits1, name = wfset_s1_name, description = glue::glue("{prod_name} meanlog prediction workflowsets {transf_msg} {this_rev}"),...

## The problem The usual method of finalizing tunable parameters in a workflow set is not working with `num_comp` for `discrim_flexible` workflows ## Reproducible example ``` r library(tidymodels) library(tidyverse) library(discrim)...