tests/testthat/test-python.R

In shaelebrown/TDAML: Machine Learning and Inference for Topological Data Analysis

# all python tests are skipped to avoid build errors, even though they succeed locally
# to run the tests the reticulate package must be installed, correctly hooked up to
# python, and the ripser module must be downloaded.

test_that("ripser can be imported and verified.",{
  
  skip_if(T)
  ripser <- import_ripser()
  expect_invisible(check_ripser(ripser))
  expect_error(check_ripser(2),"ripser object")
  expect_error(check_ripser(NULL),"ripser object")
  np <- reticulate::import("numpy")
  expect_error(check_ripser(np),"ripser object")
  
})

test_that("PyH can detect bad input parameters.",{
  
  skip_if(T)
  ripser <- import_ripser()
  expect_error(PyH(X = data.frame(),maxdim = 1,thresh = 1,distance_mat = F,ripser = ripser),"two rows")
  expect_error(PyH(X = NULL,maxdim = 1,thresh = 1,distance_mat = F,ripser = ripser),"dataframe")
  expect_error(PyH(X = data.frame(x = 1:2,y = c("1","2")),maxdim = 1,thresh = 1,distance_mat = F,ripser = ripser),"numeric")
  expect_error(PyH(X = data.frame(x = c(1,NA,2)),maxdim = 1,thresh = 1,distance_mat = F,ripser = ripser),"missing")
  expect_error(PyH(X = data.frame(x = 1:3),maxdim = 1,thresh = 1,distance_mat = T,ripser = ripser),"square")
  expect_error(PyH(X = data.frame(x = 1:3,y = 1:3,z = 1:3),maxdim = 1,thresh = 1,distance_mat = T,ripser = ripser),"matrix")
  expect_error(PyH(X = data.frame(x = 1:3,y = 1:3,z = 1:3),maxdim = NA,thresh = 1,distance_mat = T,ripser = ripser),"maxdim")
  expect_error(PyH(X = data.frame(x = 1:3,y = 1:3,z = 1:3),maxdim = -1,thresh = 1,distance_mat = T,ripser = ripser),"maxdim")
  expect_error(PyH(X = data.frame(x = 1:3,y = 1:3,z = 1:3),maxdim = 1,thresh = 1,distance_mat = NULL,ripser = ripser),"NULL")
  expect_error(PyH(X = data.frame(x = 1:3,y = 1:3,z = 1:3),maxdim = 1,thresh = 1,distance_mat = NA,ripser = ripser),"NA")
  expect_error(PyH(X = data.frame(x = 1:3,y = 1:3,z = 1:3),maxdim = 1,thresh = 1,distance_mat = c(T,F),ripser = ripser),"logical")
  expect_error(PyH(X = data.frame(x = 1:3,y = 1:3,z = 1:3),maxdim = 1,thresh = 1,distance_mat = T,ripser = ripser,ignore_infinite_cluster = NULL),"NULL")
  expect_error(PyH(X = data.frame(x = 1:3,y = 1:3,z = 1:3),maxdim = 1,thresh = 1,distance_mat = T,ripser = ripser,ignore_infinite_cluster = c(T,F)),"single")
  expect_error(PyH(X = data.frame(x = 1:3,y = 1:3,z = 1:3),maxdim = 1,thresh = 1,distance_mat = T,ripser = ripser,ignore_infinite_cluster = NA),"NA")
  expect_error(PyH(X = data.frame(x = 1:3,y = 1:3,z = 1:3),maxdim = 1,thresh = 1,distance_mat = T,ripser = ripser,calculate_representatives = NULL),"NULL")
  expect_error(PyH(X = data.frame(x = 1:3,y = 1:3,z = 1:3),maxdim = 1,thresh = 1,distance_mat = T,ripser = ripser,calculate_representatives = c(T,F)),"single")
  expect_error(PyH(X = data.frame(x = 1:3,y = 1:3,z = 1:3),maxdim = 1,thresh = 1,distance_mat = T,ripser = ripser,calculate_representatives = NA),"NA")
  
})

test_that("PyH is computing correctly.",{
  
  skip_if(T)
  skip_if_not_installed("TDAstats")
  D1 <- data.frame(x = stats::rnorm(20),y = stats::rnorm(20))
  D2 <- data.frame(x = stats::rnorm(20),y = stats::rnorm(20))
  D3 <- data.frame(x = stats::rnorm(20),y = stats::rnorm(20))
  
  phom_TDA_1 <- diagram_to_df(TDAstats::calculate_homology(D1,threshold = 5))
  phom_TDA_2 <- diagram_to_df(TDAstats::calculate_homology(D2,threshold = 5))
  phom_TDA_3 <- diagram_to_df(TDAstats::calculate_homology(D3,threshold = 5))
  
  ripser <- import_ripser()
  
  phom_py_1 <- PyH(D1,thresh = 5,ripser = ripser)
  phom_py_2 <- PyH(D2,thresh = 5,ripser = ripser)
  phom_py_3 <- PyH(D3,thresh = 5,ripser = ripser)
  
  expect_equal(phom_TDA_1,phom_py_1,tolerance = 0.00001)
  expect_equal(phom_TDA_2,phom_py_2,tolerance = 0.00001)
  expect_equal(phom_TDA_3,phom_py_3,tolerance = 0.000001)
  
  phom_with_extra_cluster <- PyH(D1,thresh = 5,ripser = ripser,ignore_infinite_cluster = F)
  
  expect_length(which(phom_with_extra_cluster$dimension == 0),20)
  
  phom_with_reps <- PyH(D1,thresh = 5,ripser = ripser,calculate_representatives = T)
  expect_type(phom_with_reps,"list")
  
  circ <- TDAstats::circle2d[sample(1:100,10),]
  phom_with_empty_dim <- PyH(circ,thresh = 2,ripser = ripser,maxdim = 2)
  expect_s3_class(phom_with_empty_dim,"data.frame")
  
})

test_that("bootstrap function can detect PyH errors correctly.",{
  
  skip_if(T)
  skip_if_not_installed("TDAstats")
  ripser = import_ripser()
  D <- TDAstats::circle2d[sample(1:100,10),]
  expect_error(bootstrap_persistence_thresholds(X = D,FUN_diag = "PyH",maxdim = 1,thresh = 2,calculate_representatives = T,return_diag = T,ripser = ripser,num_workers = 2,num_samples = 3,return_subsetted = T,ignore_infinite_cluster = NULL),"NULL")
  expect_error(bootstrap_persistence_thresholds(X = D,FUN_diag = "PyH",maxdim = 1,thresh = 2,calculate_representatives = T,return_diag = T,ripser = ripser,num_workers = 2,num_samples = 3,return_subsetted = T,ignore_infinite_cluster = 2),"logical")
  expect_error(bootstrap_persistence_thresholds(X = D,FUN_boot = "PyH",maxdim = 1,thresh = 2,calculate_representatives = T,return_diag = T,ripser = ripser,num_workers = 2,num_samples = 3,return_subsetted = T,ignore_infinite_cluster = NA),"NA")
  
})

test_that("PyH functionality works in bootstrap function.",{
  
  skip_if(T)
  skip_if_not_installed("TDAstats")
  ripser = import_ripser()
  D <- TDAstats::circle2d[sample(1:100,10),]
  
  # PyH with multiple thresholds
  bs <- bootstrap_persistence_thresholds(X = D,FUN_diag = "PyH",FUN_boot = "PyH",maxdim = 1,thresh = 2,calculate_representatives = T,return_diag = T,ripser = ripser,num_workers = 2,num_samples = 3,return_subsetted = T,ignore_infinite_cluster = F)
  expect_length(bs$representatives[[2]],length(which(bs$diag$dimension == 1)))
  expect_length(bs$thresholds,2)
  expect_gt(bs$thresholds[[1]],0)
  expect_gt(bs$thresholds[[2]],0)
  expect_lte(length(bs$subsetted_representatives),nrow(bs$subsetted_diag) + 1)
  if(length(which(bs$subsetted_diag$dimension == 0)) > 0)
  {
    expect_true(min(bs$subsetted_diag[which(bs$subsetted_diag$dimension == 0),]$death - bs$subsetted_diag[which(bs$subsetted_diag$dimension == 0),]$birth) >= bs$thresholds[[1]])
  }
  
  expect_true(min(bs$subsetted_diag[which(bs$subsetted_diag$dimension == 1),]$death - bs$subsetted_diag[which(bs$subsetted_diag$dimension == 1),]$birth) > bs$thresholds[[2]])
  
  # check on circle
  bs <- bootstrap_persistence_thresholds(X = D,FUN_diag = "PyH",maxdim = 1,thresh = 2,return_diag = T,ripser = ripser,num_workers = 2,num_samples = 3)
  expect_lte(length(bs$subsetted_diag$dimension),1)
  bs <- bootstrap_persistence_thresholds(X = D,FUN_diag = "PyH",maxdim = 1,thresh = 2,return_diag = T,ripser = ripser,ignore_infinite_cluster = F,num_workers = 2,num_samples = 3)
  expect_lte(length(bs$subsetted_diag$dimension),2)

})