tests/testthat/test-plot_train_valid_error.R
In UBC-MDS/RMLViz: Machine learning results visualization helper in R
# Created on March 5, 2020
#
# @author: Fanli Zhou
#
# This script tests the plot_train_valid_error function
# in the RMLViz package.
#
# plot_train_valid_error function takes in a model name,
# train/validation data sets, a parameter name and a vector
# of parameter values to try and then plots train/validation
# errors vs. parameter values.
data(iris)
# use the iris data for unittest
set.seed(123)
split <- caTools::sample.split(iris$Species, SplitRatio = 0.75)
training_set <- subset(iris, split == TRUE)
valid_set <- subset(iris, split == FALSE)
X_train <- training_set[, -5]
y_train <- training_set[, 5]
X_valid <- valid_set[, -5]
y_valid <- valid_set[, 5]
# test that output is a ggplot2 plot
test_that("The return type of knn is not correct.", {
expect_true(ggplot2::is.ggplot(plot_train_valid_error('knn',
X_train, y_train,
X_valid, y_valid,
'k', seq(1, 50))))
})
test_that("The return type of decision tree is not correct.", {
expect_true(ggplot2::is.ggplot(plot_train_valid_error('decision tree',
X_train, y_train,
X_valid, y_valid,
'maxdepth', c(5, 10, 15, 20))))
})
test_that("The return type of svm is not correct.", {
expect_true(ggplot2::is.ggplot(plot_train_valid_error('svm',
X_train, y_train,
X_valid, y_valid,
'cost', c(0.00001, 0.0001, 0.001, 0.01, 0.1, 1, 10))))
})
test_that("The return type of svm is not correct.", {
expect_true(ggplot2::is.ggplot(plot_train_valid_error('svm',
X_train, y_train,
X_valid, y_valid,
'gamma', c(0.00001, 0.0001, 0.001, 0.01, 0.1, 1, 10))))
})
test_that("The return type of random forests is not correct.", {
expect_true(ggplot2::is.ggplot(plot_train_valid_error('random forests',
X_train, y_train,
X_valid, y_valid,
'ntree', c(5, 10, 15, 20))))
})
# test for error if input is of a wrong type
test_that("X_train should be a data frame.", {
expect_error(plot_train_valid_error('knn',
c(1, 2, 3), y_train,
X_valid, y_valid,
'k', seq(1, 50)),
"Sorry, X_train should be a data frame.")
})
test_that("y_train should be a vector or factor.", {
expect_error(plot_train_valid_error('knn',
X_train, matrix(c(1, 2, 3, 4), nrow=1, ncol=4),
X_valid, y_valid,
'k', seq(1, 50)),
"Sorry, y_train should be a vector or factor.")
})
test_that("X_valid should be a data frame.", {
expect_error(plot_train_valid_error('knn',
X_train, y_train,
c(1, 2, 3), y_valid,
'k', seq(1, 50)),
"Sorry, X_valid should be a data frame.")
})
test_that("y_valid should be a vector or factor.", {
expect_error(plot_train_valid_error('knn',
X_train, y_train,
X_valid, matrix(c(1, 2, 3, 4), nrow=1, ncol=4),
'k', seq(1, 50)),
"Sorry, y_valid should be a vector or factor.")
})
test_that("All elements in X_train should be numeric.", {
expect_error(plot_train_valid_error('knn',
tibble::tibble(a = c(1, 2, '3')), y_train,
X_valid, y_valid,
'k', seq(1, 50)),
"Sorry, all elements in X_train should be numeric.")
})
test_that("All elements in y_train should be numeric or factor.", {
expect_error(plot_train_valid_error('knn',
X_train, c(1, 2, '3'),
X_valid, y_valid,
'k', seq(1, 50)),
"Sorry, all elements in y_train should be numeric or factor.")
})
test_that("All elements in X_valid should be numeric.", {
expect_error(plot_train_valid_error('knn',
X_train, y_train,
tibble::tibble(a = c(1, 2, '3')), y_valid,
'k', seq(1, 50)),
"Sorry, all elements in X_valid should be numeric.")
})
test_that("All elements in y_valid should be numeric or factor.", {
expect_error(plot_train_valid_error('knn',
X_train, y_train,
X_valid, c(1, 2, '3'),
'k', seq(1, 50)),
"Sorry, all elements in y_valid should be numeric or factor.")
})
test_that("All elements in para_vec should be numeric.", {
expect_error(plot_train_valid_error('knn',
X_train, y_train,
X_valid, y_valid,
'k', c(1, 2, '3')),
"Sorry, all elements in para_vec should be numeric.")
})
test_that("All elements in para_vec should be non-negative.", {
expect_error(plot_train_valid_error('knn',
X_train, y_train,
X_valid, y_valid,
'k', c(1, 2, -3)),
"Sorry, all elements in para_vec should be non-negative.")
})
test_that("X_train and y_train should have the same number of rows.", {
expect_error(plot_train_valid_error('knn',
tibble::tibble(a = c(1, 2, 3), b = c(1, 2, 3)), c(1, 2, 3, 4),
tibble::tibble(a = c(1, 2, 3)), c(1, 2, 3),
'k', seq(1, 50)),
"Sorry, X_train and y_train should have the same number of rows.")
})
test_that("X_valid and y_valid should have the same number of rows.", {
expect_error(plot_train_valid_error('knn',
tibble::tibble(a = c(1, 2, 3)), c(1, 2, 3),
tibble::tibble(a = c(1, 2, 3), b = c(1, 2, 3)), c(1, 2, 3, 4),
'k', seq(1, 50)),
"Sorry, X_valid and y_valid should have the same number of rows.")
})
test_that("X_train and X_valid should have the same number of columns.", {
expect_error(plot_train_valid_error('knn',
tibble::tibble(a = c(1, 2, 3), b = c(1, 2, 3)), c(1, 2, 3),
tibble::tibble(a = c(1, 2, 3)), c(1, 2, 3),
'k', seq(1, 50)),
"Sorry, X_train and X_valid should have the same number of columns.")
})
# test for error if input is of a wrong value
test_that("Only the hyperparameter 'k' is allowed for a 'knn' model.", {
expect_error(plot_train_valid_error('knn',
X_train, y_train,
X_valid, y_valid,
'n', seq(1, 50)),
"Sorry, only the hyperparameter 'k' is allowed for a 'knn' model.")
})
test_that("Only the hyperparameter 'maxdepth' is allowed for a 'decision tree' model.", {
expect_error(plot_train_valid_error('decision tree',
X_train, y_train,
X_valid, y_valid,
'm', c(5, 10, 15, 20)),
"Sorry, only the hyperparameter 'maxdepth' is allowed for a 'decision tree' model.")
})
test_that("Only the hyperparameter, 'cost' and 'gamma', are allowed for a 'svm' model.", {
expect_error(plot_train_valid_error('svm',
X_train, y_train,
X_valid, y_valid,
'c', c(0.00001, 0.0001, 0.001, 0.01, 0.1, 1, 10)),
"Sorry, only the hyperparameters, 'cost' and 'gamma', are allowed for a 'svm' model.")
})
test_that("Only the hyperparameter 'ntree' is allowed for a 'random forests' model.", {
expect_error(plot_train_valid_error('random forests',
X_train, y_train,
X_valid, y_valid,
'cv', c(0.00001, 0.0001, 0.001, 0.01, 0.1, 1, 10)),
"Sorry, only the hyperparameter 'ntree' is allowed for a 'random forests' model.")
})
test_that("The model_name should be chosen from 'knn', 'decision tree', 'svc' and 'random forests'.", {
expect_error(plot_train_valid_error('r',
X_train, y_train,
X_valid, y_valid,
'cv', c(0.00001, 0.0001, 0.001, 0.01, 0.1, 1, 10)),
"Sorry, the model_name should be chosen from 'knn', 'decision tree', 'svc' and 'random forests'.")
})
UBC-MDS/RMLViz documentation built on April 1, 2020, 4:35 a.m.
R Package Documentation
Browse R Packages
We want your feedback!
Note that we can't provide technical support on individual packages. You should contact the package authors for that.
# Created on March 5, 2020
#
# @author: Fanli Zhou
#
# This script tests the plot_train_valid_error function
# in the RMLViz package.
#
# plot_train_valid_error function takes in a model name,
# train/validation data sets, a parameter name and a vector
# of parameter values to try and then plots train/validation
# errors vs. parameter values.
data(iris)
# use the iris data for unittest
set.seed(123)
split <- caTools::sample.split(iris$Species, SplitRatio = 0.75)
training_set <- subset(iris, split == TRUE)
valid_set <- subset(iris, split == FALSE)
X_train <- training_set[, -5]
y_train <- training_set[, 5]
X_valid <- valid_set[, -5]
y_valid <- valid_set[, 5]
# test that output is a ggplot2 plot
test_that("The return type of knn is not correct.", {
expect_true(ggplot2::is.ggplot(plot_train_valid_error('knn',
X_train, y_train,
X_valid, y_valid,
'k', seq(1, 50))))
})
test_that("The return type of decision tree is not correct.", {
expect_true(ggplot2::is.ggplot(plot_train_valid_error('decision tree',
X_train, y_train,
X_valid, y_valid,
'maxdepth', c(5, 10, 15, 20))))
})
test_that("The return type of svm is not correct.", {
expect_true(ggplot2::is.ggplot(plot_train_valid_error('svm',
X_train, y_train,
X_valid, y_valid,
'cost', c(0.00001, 0.0001, 0.001, 0.01, 0.1, 1, 10))))
})
test_that("The return type of svm is not correct.", {
expect_true(ggplot2::is.ggplot(plot_train_valid_error('svm',
X_train, y_train,
X_valid, y_valid,
'gamma', c(0.00001, 0.0001, 0.001, 0.01, 0.1, 1, 10))))
})
test_that("The return type of random forests is not correct.", {
expect_true(ggplot2::is.ggplot(plot_train_valid_error('random forests',
X_train, y_train,
X_valid, y_valid,
'ntree', c(5, 10, 15, 20))))
})
# test for error if input is of a wrong type
test_that("X_train should be a data frame.", {
expect_error(plot_train_valid_error('knn',
c(1, 2, 3), y_train,
X_valid, y_valid,
'k', seq(1, 50)),
"Sorry, X_train should be a data frame.")
})
test_that("y_train should be a vector or factor.", {
expect_error(plot_train_valid_error('knn',
X_train, matrix(c(1, 2, 3, 4), nrow=1, ncol=4),
X_valid, y_valid,
'k', seq(1, 50)),
"Sorry, y_train should be a vector or factor.")
})
test_that("X_valid should be a data frame.", {
expect_error(plot_train_valid_error('knn',
X_train, y_train,
c(1, 2, 3), y_valid,
'k', seq(1, 50)),
"Sorry, X_valid should be a data frame.")
})
test_that("y_valid should be a vector or factor.", {
expect_error(plot_train_valid_error('knn',
X_train, y_train,
X_valid, matrix(c(1, 2, 3, 4), nrow=1, ncol=4),
'k', seq(1, 50)),
"Sorry, y_valid should be a vector or factor.")
})
test_that("All elements in X_train should be numeric.", {
expect_error(plot_train_valid_error('knn',
tibble::tibble(a = c(1, 2, '3')), y_train,
X_valid, y_valid,
'k', seq(1, 50)),
"Sorry, all elements in X_train should be numeric.")
})
test_that("All elements in y_train should be numeric or factor.", {
expect_error(plot_train_valid_error('knn',
X_train, c(1, 2, '3'),
X_valid, y_valid,
'k', seq(1, 50)),
"Sorry, all elements in y_train should be numeric or factor.")
})
test_that("All elements in X_valid should be numeric.", {
expect_error(plot_train_valid_error('knn',
X_train, y_train,
tibble::tibble(a = c(1, 2, '3')), y_valid,
'k', seq(1, 50)),
"Sorry, all elements in X_valid should be numeric.")
})
test_that("All elements in y_valid should be numeric or factor.", {
expect_error(plot_train_valid_error('knn',
X_train, y_train,
X_valid, c(1, 2, '3'),
'k', seq(1, 50)),
"Sorry, all elements in y_valid should be numeric or factor.")
})
test_that("All elements in para_vec should be numeric.", {
expect_error(plot_train_valid_error('knn',
X_train, y_train,
X_valid, y_valid,
'k', c(1, 2, '3')),
"Sorry, all elements in para_vec should be numeric.")
})
test_that("All elements in para_vec should be non-negative.", {
expect_error(plot_train_valid_error('knn',
X_train, y_train,
X_valid, y_valid,
'k', c(1, 2, -3)),
"Sorry, all elements in para_vec should be non-negative.")
})
test_that("X_train and y_train should have the same number of rows.", {
expect_error(plot_train_valid_error('knn',
tibble::tibble(a = c(1, 2, 3), b = c(1, 2, 3)), c(1, 2, 3, 4),
tibble::tibble(a = c(1, 2, 3)), c(1, 2, 3),
'k', seq(1, 50)),
"Sorry, X_train and y_train should have the same number of rows.")
})
test_that("X_valid and y_valid should have the same number of rows.", {
expect_error(plot_train_valid_error('knn',
tibble::tibble(a = c(1, 2, 3)), c(1, 2, 3),
tibble::tibble(a = c(1, 2, 3), b = c(1, 2, 3)), c(1, 2, 3, 4),
'k', seq(1, 50)),
"Sorry, X_valid and y_valid should have the same number of rows.")
})
test_that("X_train and X_valid should have the same number of columns.", {
expect_error(plot_train_valid_error('knn',
tibble::tibble(a = c(1, 2, 3), b = c(1, 2, 3)), c(1, 2, 3),
tibble::tibble(a = c(1, 2, 3)), c(1, 2, 3),
'k', seq(1, 50)),
"Sorry, X_train and X_valid should have the same number of columns.")
})
# test for error if input is of a wrong value
test_that("Only the hyperparameter 'k' is allowed for a 'knn' model.", {
expect_error(plot_train_valid_error('knn',
X_train, y_train,
X_valid, y_valid,
'n', seq(1, 50)),
"Sorry, only the hyperparameter 'k' is allowed for a 'knn' model.")
})
test_that("Only the hyperparameter 'maxdepth' is allowed for a 'decision tree' model.", {
expect_error(plot_train_valid_error('decision tree',
X_train, y_train,
X_valid, y_valid,
'm', c(5, 10, 15, 20)),
"Sorry, only the hyperparameter 'maxdepth' is allowed for a 'decision tree' model.")
})
test_that("Only the hyperparameter, 'cost' and 'gamma', are allowed for a 'svm' model.", {
expect_error(plot_train_valid_error('svm',
X_train, y_train,
X_valid, y_valid,
'c', c(0.00001, 0.0001, 0.001, 0.01, 0.1, 1, 10)),
"Sorry, only the hyperparameters, 'cost' and 'gamma', are allowed for a 'svm' model.")
})
test_that("Only the hyperparameter 'ntree' is allowed for a 'random forests' model.", {
expect_error(plot_train_valid_error('random forests',
X_train, y_train,
X_valid, y_valid,
'cv', c(0.00001, 0.0001, 0.001, 0.01, 0.1, 1, 10)),
"Sorry, only the hyperparameter 'ntree' is allowed for a 'random forests' model.")
})
test_that("The model_name should be chosen from 'knn', 'decision tree', 'svc' and 'random forests'.", {
expect_error(plot_train_valid_error('r',
X_train, y_train,
X_valid, y_valid,
'cv', c(0.00001, 0.0001, 0.001, 0.01, 0.1, 1, 10)),
"Sorry, the model_name should be chosen from 'knn', 'decision tree', 'svc' and 'random forests'.")
})
R Package Documentation
Browse R Packages
We want your feedback!
Note that we can't provide technical support on individual packages. You should contact the package authors for that.
Embedding an R snippet on your website
Add the following code to your website.
For more information on customizing the embed code, read Embedding Snippets.