tdarec (version 0.2.0)

step_vpd_persistence_block: Persistence Block Vectorization of Persistent Homology

Description

The function step_vpd_persistence_block() creates a specification of a recipe step that will convert a list-column of 3-column matrices of persistence data to a list-column of 1-row matrices of vectorizations.

Usage

step_vpd_persistence_block(
  recipe,
  ...,
  role = "predictor",
  trained = FALSE,
  hom_degree = 0L,
  xseq = NULL,
  xmin = NULL,
  xmax = NULL,
  xlen = NULL,
  xby = NULL,
  yseq = NULL,
  ymin = NULL,
  ymax = NULL,
  ylen = NULL,
  yby = NULL,
  block_size = 0.3,
  columns = NULL,
  keep_original_cols = TRUE,
  skip = FALSE,
  id = rand_id("vpd_persistence_block")
)

Value

An updated version of recipe with the new step added to the sequence of any existing operations.

Arguments

recipe

A recipe object. The step will be added to the sequence of operations for this recipe.

...

One or more selector functions to choose variables for this step. See selections() for more details.

role

For model terms created by this step, what analysis role should they be assigned? By default, the new columns created by this step from the original variables will be used as predictors in a model.

trained

A logical to indicate if the quantities for preprocessing have been estimated.

hom_degree

The homological degree of the features to be transformed.

xseq

A discretization grid, as an increasing numeric vector. xseq overrides the other x* parameters with a warning.

xmin, xmax, xlen, xby

Limits and resolution of a discretization grid; specify only one of xlen and xby.

yseq

Combined with xseq to form a 2-dimensional discretization grid.

ymin, ymax, ylen, yby

Limits and resolution of a discretization grid; specify only one of ylen and yby.

block_size

The scaling factor of the squares used to obtain persistence blocks. The side length of the square centered at a feature \((b,p)\) is obtained by multiplying \(2p\) by this factor.

columns

A character string of the selected variable names. This field is a placeholder and will be populated once prep() is used.

keep_original_cols

A logical to keep the original variables in the output. Defaults to FALSE.

skip

A logical. Should the step be skipped when the recipe is baked by bake()? While all operations are baked when prep() is run, some operations may not be able to be conducted on new data (e.g. processing the outcome variable(s)). Care should be taken when using skip = TRUE as it may affect the computations for subsequent operations.

id

A character string that is unique to this step to identify it.

Engine

The persistence block vectorization deploys TDAvec::computePersistenceBlock(). See there for definitions and references.

Tuning Parameters

This step has 2 tuning parameters:

• hom_degree: Homological degree (type: integer, default: 0L)

• block_size: Square side length scaling factor (type: double, default: 0.3)

Details

Persistent homology is usually encoded as birth--death pairs (barcodes or diagrams), but the space of persistence data sets does not satisfy convenient statistical properties. Such applications as hypothesis testing and machine learning benefit from transformations of persistence data, often to Hilbert spaces (vector spaces with inner products and induced metrics).

Examples

library(recipes)

# inspect vectorized features
volc_dat <- data.frame(image = I(list(volcano / 10)))
recipe(~ image, data = volc_dat) %>% 
  step_pd_raster(image, method = "link_join") %>% 
  step_vpd_persistence_block(image, hom_degree = 1, block_size = 1) %>% 
  print() -> volc_rec
print(volc_rec)
volc_rec %>% 
  prep(training = volc_dat) %>% 
  bake(new_data = volc_dat)

# dimension-reduce using vectorized features
data(permeability_qsar, package = "modeldata")
permeability_qsar %>% 
  transform(perm_cut = cut(permeability, breaks = seq(0, 60, 10))) %>% 
  subset(select = -permeability) %>% 
  tidyr::nest(chem_fp = -perm_cut) %>% 
  print() -> perm_dat
recipe(perm_cut ~ chem_fp, data = perm_dat) %>% 
  step_pd_point_cloud(chem_fp, max_hom_degree = 2) %>% 
  step_vpd_persistence_block(chem_fp, hom_degree = 1, block_size = 1) %>% 
  step_pca(starts_with("chem_fp_"), num_comp = 2) %>%
  print() -> perm_rec
perm_est <- prep(perm_rec, training = perm_dat)
perm_res <- bake(perm_est, new_data = perm_dat)
# inspect results
tidy(perm_rec)
tidy(perm_rec, number = 2)
tidy(perm_est, number = 2)
# visualize results
with(perm_res, {
  plot(PC1, PC2, type = "n", asp = 1)
  text(PC1, PC2, labels = perm_cut)
})