more prep for CRAN submission

DESCRIPTION

@@ -11,7 +11,7 @@ Description: Implements the VEESA pipeline for explainable machine learning
   inputs to elastic shape analysis principal components that account for
   vertial and/or horizontal variability. Computes feature importance to 
   identify important principal components and visualizes variability captured
-  by functional principal components. See Goode et al (2025) arXiv
+  by functional principal components. See Goode et al (2025) <arXiv:2501.07602>
   for technical details about the methodology.
 License: MIT + file LICENSE
 Encoding: UTF-8

R/compute_pfi.R

@@ -38,7 +38,7 @@
 #'   select(data, group, id) |>
 #'   distinct() |>
 #'   group_by(data, group) |>
-#'   slice(1:5) |>
+#'   slice(1:4) |>
 #'   ungroup()
 #'
 #' # Create a smaller version of shifted data
@@ -95,7 +95,7 @@
 #'     x = model_data |> select(-group),
 #'     y = model_data$group,
 #'     f = rf,
-#'     K = 5,
+#'     K = 1,
 #'     metric = "accuracy"
 #'  )
 

R/prep_training_data.R

@@ -56,7 +56,7 @@
 #'   select(data, group, id) |>
 #'   distinct() |>
 #'   group_by(data, group) |>
-#'   slice(1:5) |>
+#'   slice(1:3) |>
 #'   ungroup()
 #'
 #' # Create a smaller version of shifted data

README.Rmd

@@ -14,20 +14,18 @@ always_allow_html: yes
 knitr::opts_chunk$set(echo = TRUE, message = FALSE, dpi = 300)
 ```
 
-`veesa` is an R package for implementing the VEESA pipeline for an explainable approach to training machine learning models with functional data inputs. (Paper appearing on arXiv soon!) The approach makes use of the [elastic shape analysis (ESA) framework](https://doi.org/10.1016/j.csda.2012.12.001) for functional data.
-
-We hope to have `veesa` available on CRAN soon! For now, install `veesa` from GitHub using the commands below.
+`veesa` is an R package for implementing the VEESA pipeline for an explainable approach to training machine learning models with functional data inputs. See a preprint manuscript describing the approach on [arXiv](https://arxiv.org/abs/2501.07602). We hope to have `veesa` available on CRAN soon! For now, install `veesa` from GitHub using the commands below.
 
 ```{r}
 #| eval: false
 remotes::install_github("sandialabs/veesa")
 ```
 
-Keep reading for an example training a random forest model using `veesa` with simulated functional data.
+Keep reading for an example using `veesa` to implement the VEESA pipeline.
 
 ## Demonstration
 
-### Set Up and Data Generation
+#### Set Up and Data Generation
 
 ```{r}
 #| message: false
@@ -109,7 +107,7 @@ Create a vector of times:
 times = sim_data$t %>% unique()
 ```
 
-### Alignment and fPCA
+#### Alignment and fPCA
 
 Prepare train data
 
@@ -196,7 +194,7 @@ bind_rows(train_plot_df_jfpca, test_plot_df_jfpca) %>%
   theme_bw()
 ```
 
-### Models
+#### Models
 
 Create response variable: 
 
@@ -226,7 +224,7 @@ set.seed(20211130)
 rf_jfpca = randomForest(x1 ~ ., data = rf_jfpca_df)
 ```
 
-### PFI
+#### PFI
 
 Compute PFI:
 
@@ -305,7 +303,6 @@ Principal directions of top PC for each jfPCA method:
 #| out-width: "60%"
 #| fig-align: "center"
 #| echo: false
-
 plot_pc_directions(
   fpcs = top_pc_jfpca,
   fdasrvf = train_transformed_jfpca$fpca_res,

README.md

@@ -10,24 +10,21 @@ coverage](https://codecov.io/gh/sandialabs/veesa/graph/badge.svg)](https://app.c
 
 `veesa` is an R package for implementing the VEESA pipeline for an
 explainable approach to training machine learning models with functional
-data inputs. (Paper appearing on arXiv soon!) The approach makes use of
-the [elastic shape analysis (ESA)
-framework](https://doi.org/10.1016/j.csda.2012.12.001) for functional
-data.
-
-We hope to have `veesa` available on CRAN soon! For now, install `veesa`
-from GitHub using the commands below.
+data inputs. See a preprint manuscript describing the approach on
+[arXiv](https://arxiv.org/abs/2501.07602). We hope to have `veesa`
+available on CRAN soon! For now, install `veesa` from GitHub using the
+commands below.
 
 ``` r
 remotes::install_github("sandialabs/veesa")
 ```
 
-Keep reading for an example training a random forest model using `veesa`
-with simulated functional data.
+Keep reading for an example using `veesa` to implement the VEESA
+pipeline.
 
 ## Demonstration
 
-### Set Up and Data Generation
+#### Set Up and Data Generation
 
 ``` r
 # Load R packages
@@ -95,7 +92,7 @@ Create a vector of times:
 times = sim_data$t %>% unique()
 ```
 
-### Alignment and fPCA
+#### Alignment and fPCA
 
 Prepare train data
 
@@ -129,7 +126,7 @@ Compare jfPCA coefficients from train and test data:
 
 <img src="README_files/figure-gfm/unnamed-chunk-11-1.png" width="75%" style="display: block; margin: auto;" />
 
-### Models
+#### Models
 
 Create response variable:
 
@@ -159,7 +156,7 @@ set.seed(20211130)
 rf_jfpca = randomForest(x1 ~ ., data = rf_jfpca_df)
 ```
 
-### PFI
+#### PFI
 
 Compute PFI: