tests/testthat/test-ConditionalGaussianSampler.R

In xplainfi: Feature Importance Methods for Global Explanations

# Tests for ConditionalGaussianSampler

test_that("ConditionalGaussianSampler initialization works", {
	task = tgen("friedman1")$generate(n = 100)
	sampler = ConditionalGaussianSampler$new(task)

	expect_s3_class(sampler, "ConditionalGaussianSampler")
	expect_s3_class(sampler, "ConditionalSampler")
	expect_equal(sampler$label, "Gaussian Conditional Sampler")
	expect_s3_class(sampler$param_set, "ParamSet")

	# Check stored statistics
	checkmate::expect_numeric(sampler$mu, len = task$n_features, any.missing = FALSE)
	checkmate::expect_matrix(sampler$sigma, nrows = task$n_features, ncols = task$n_features)
	expect_true(isSymmetric(sampler$sigma))
})

test_that("ConditionalGaussianSampler rejects non-numeric tasks", {
	task = tsk("penguins")
	expect_error(ConditionalGaussianSampler$new(task), "unsupported feature types")
})

test_that("ConditionalGaussianSampler sampling works", {
	task = tgen("friedman1")$generate(n = 100)
	sampler = ConditionalGaussianSampler$new(task)

	# Marginal sampling
	expect_marginal_sampling(sampler, feature = "important1", row_ids = 1:20)

	# Conditional sampling with single conditioning feature
	expect_conditional_sampling(
		sampler,
		feature = "important2",
		conditioning_set = "important1",
		row_ids = 1:20
	)

	# Conditional sampling with multiple features
	expect_conditional_sampling(
		sampler,
		feature = c("important2", "important3"),
		conditioning_set = "important1",
		row_ids = 1:20
	)
})

test_that("ConditionalGaussianSampler sample_newdata works", {
	task = tgen("friedman1")$generate(n = 100)
	sampler = ConditionalGaussianSampler$new(task)
	test_data = task$data(rows = 1:10)

	sampled = sampler$sample_newdata(
		feature = c("important2", "important3"),
		newdata = test_data,
		conditioning_set = "important1"
	)

	expect_sampler_output_structure(sampled, task, nrows = 10)
	expect_feature_type_consistency(sampled, task)
	expect_non_sampled_unchanged(sampled, test_data, "important1")
})

test_that("ConditionalGaussianSampler handles single observation", {
	task = tgen("friedman1")$generate(n = 100)
	sampler = ConditionalGaussianSampler$new(task)
	test_data = task$data(rows = 1)

	sampled = sampler$sample_newdata(
		feature = "important2",
		newdata = test_data,
		conditioning_set = "important1"
	)

	expect_sampler_output_structure(sampled, task, nrows = 1)
	expect_non_sampled_unchanged(sampled, test_data, "important1")
})

test_that("ConditionalGaussianSampler is reproducible with seed", {
	task = tgen("friedman1")$generate(n = 100)
	sampler = ConditionalGaussianSampler$new(task)
	test_data = task$data(rows = 1:10)

	sampled1 = withr::with_seed(123, {
		sampler$sample_newdata(
			feature = "important2",
			newdata = test_data,
			conditioning_set = "important1"
		)
	})

	sampled2 = withr::with_seed(123, {
		sampler$sample_newdata(
			feature = "important2",
			newdata = test_data,
			conditioning_set = "important1"
		)
	})

	expect_identical(sampled1$important2, sampled2$important2)
})

test_that("ConditionalGaussianSampler conditioning_set parameter behavior", {
	task = tgen("friedman1")$generate(n = 100)
	test_conditioning_set_behavior(ConditionalGaussianSampler, task)
})

test_that("ConditionalGaussianSampler preserves feature types", {
	test_sampler_feature_types(ConditionalGaussianSampler)
})