Title:	Extra Recipes Steps for Dealing with Unbalanced Data
Version:	1.0.3
Description:	A dataset with an uneven number of cases in each class is said to be unbalanced. Many models produce a subpar performance on unbalanced datasets. A dataset can be balanced by increasing the number of minority cases using SMOTE 2011 <doi:10.48550/arXiv.1106.1813>, BorderlineSMOTE 2005 <doi:10.1007/11538059_91> and ADASYN 2008 https://ieeexplore.ieee.org/document/4633969. Or by decreasing the number of majority cases using NearMiss 2003 https://www.site.uottawa.ca/~nat/Workshop2003/jzhang.pdf or Tomek link removal 1976 https://ieeexplore.ieee.org/document/4309452.
License:	MIT + file LICENSE
URL:	https://github.com/tidymodels/themis, https://themis.tidymodels.org
BugReports:	https://github.com/tidymodels/themis/issues
Depends:	R (≥ 3.6), recipes (≥ 1.1.0)
Imports:	cli, gower, lifecycle (≥ 1.0.3), dplyr, generics (≥ 0.1.0), purrr, RANN, rlang (≥ 1.1.0), ROSE, tibble, withr, glue, hardhat, vctrs
Suggests:	covr, dials (≥ 1.2.0), ggplot2, modeldata, testthat (≥ 3.0.0)
Config/Needs/website:	tidyverse/tidytemplate
Config/testthat/edition:	3
Encoding:	UTF-8
LazyData:	true
RoxygenNote:	7.3.2
NeedsCompilation:	no
Packaged:	2025-01-22 23:40:53 UTC; emilhvitfeldt
Author:	Emil Hvitfeldt [aut, cre], Posit Software, PBC [cph, fnd]
Maintainer:	Emil Hvitfeldt <emil.hvitfeldt@posit.co>
Repository:	CRAN
Date/Publication:	2025-01-23 00:10:02 UTC

themis: Extra Recipes Steps for Dealing with Unbalanced Data

Description

A dataset with an uneven number of cases in each class is said to be unbalanced. Many models produce a subpar performance on unbalanced datasets. A dataset can be balanced by increasing the number of minority cases using SMOTE 2011 arXiv:1106.1813, BorderlineSMOTE 2005 doi:10.1007/11538059_91 and ADASYN 2008 https://ieeexplore.ieee.org/document/4633969. Or by decreasing the number of majority cases using NearMiss 2003 https://www.site.uottawa.ca/~nat/Workshop2003/jzhang.pdf or Tomek link removal 1976 https://ieeexplore.ieee.org/document/4309452.

Author(s)

Maintainer: Emil Hvitfeldt emil.hvitfeldt@posit.co (ORCID)

Other contributors:

• Posit Software, PBC [copyright holder, funder]

See Also

Useful links:

• https://github.com/tidymodels/themis

• https://themis.tidymodels.org

• Report bugs at https://github.com/tidymodels/themis/issues

Adaptive Synthetic Algorithm

Description

Generates synthetic positive instances using ADASYN algorithm.

Usage

adasyn(df, var, k = 5, over_ratio = 1)

Arguments

Details

All columns used in this function must be numeric with no missing data.

Value

A data.frame or tibble, depending on type of df.

References

Chawla, N. V., Bowyer, K. W., Hall, L. O., and Kegelmeyer, W. P. (2002). Smote: Synthetic minority over-sampling technique. Journal of Artificial Intelligence Research, 16:321-357.

See Also

step_adasyn() for step function of this method

Other Direct Implementations: bsmote(), nearmiss(), smote(), smotenc(), tomek()

Examples

circle_numeric <- circle_example[, c("x", "y", "class")]

res <- adasyn(circle_numeric, var = "class")

res <- adasyn(circle_numeric, var = "class", k = 10)

res <- adasyn(circle_numeric, var = "class", over_ratio = 0.8)

borderline-SMOTE Algorithm

Description

BSMOTE generates generate new examples of the minority class using nearest neighbors of these cases in the border region between classes.

Usage

bsmote(df, var, k = 5, over_ratio = 1, all_neighbors = FALSE)

Arguments

Details

This methods works the same way as smote(), expect that instead of generating points around every point of of the minority class each point is first being classified into the boxes "danger" and "not". For each point the k nearest neighbors is calculated. If all the neighbors comes from a different class it is labeled noise and put in to the "not" box. If more then half of the neighbors comes from a different class it is labeled "danger.

If all_neighbors = FALSE then points will be generated between nearest neighbors in its own class. If all_neighbors = TRUE then points will be generated between any nearest neighbors. See examples for visualization.

The parameter neighbors controls the way the new examples are created. For each currently existing minority class example X new examples will be created (this is controlled by the parameter over_ratio as mentioned above). These examples will be generated by using the information from the neighbors nearest neighbor of each example of the minority class. The parameter neighbors controls how many of these neighbor are used.

All columns used in this step must be numeric with no missing data.

Value

A data.frame or tibble, depending on type of df.

References

Hui Han, Wen-Yuan Wang, and Bing-Huan Mao. Borderline-smote: a new over-sampling method in imbalanced data sets learning. In International Conference on Intelligent Computing, pages 878–887. Springer, 2005.

See Also

step_bsmote() for step function of this method

Other Direct Implementations: adasyn(), nearmiss(), smote(), smotenc(), tomek()

Examples

circle_numeric <- circle_example[, c("x", "y", "class")]

res <- bsmote(circle_numeric, var = "class")

res <- bsmote(circle_numeric, var = "class", k = 10)

res <- bsmote(circle_numeric, var = "class", over_ratio = 0.8)

res <- bsmote(circle_numeric, var = "class", all_neighbors = TRUE)

Synthetic Dataset With a Circle

Description

A random dataset with two classes one of which is inside a circle. Used for examples to show how the different methods handles borders.

Usage

circle_example

Format

A data frame with 200 rows and 4 variables:

x

Numeric.

y

Numeric.

class

Factor, values "Circle" and "Rest".

id

character, ID variable.

Remove Points Near Other Classes

Description

Generates synthetic positive instances using nearmiss algorithm.

Usage

nearmiss(df, var, k = 5, under_ratio = 1)

Arguments

Details

All columns used in this function must be numeric with no missing data.

Value

A data.frame or tibble, depending on type of df.

References

Inderjeet Mani and I Zhang. knn approach to unbalanced data distributions: a case study involving information extraction. In Proceedings of workshop on learning from imbalanced datasets, 2003.

See Also

step_nearmiss() for step function of this method

Other Direct Implementations: adasyn(), bsmote(), smote(), smotenc(), tomek()

Examples

circle_numeric <- circle_example[, c("x", "y", "class")]

res <- nearmiss(circle_numeric, var = "class")

res <- nearmiss(circle_numeric, var = "class", k = 10)

res <- nearmiss(circle_numeric, var = "class", under_ratio = 1.5)

Objects exported from other packages

Description

These objects are imported from other packages. Follow the links below to see their documentation.

generics

required_pkgs, tidy, tunable

S3 methods for tracking which additional packages are needed for steps.

Description

S3 methods for tracking which additional packages are needed for steps.

Usage

## S3 method for class 'step_adasyn'
required_pkgs(x, ...)

## S3 method for class 'step_bsmote'
required_pkgs(x, ...)

## S3 method for class 'step_downsample'
required_pkgs(x, ...)

## S3 method for class 'step_nearmiss'
required_pkgs(x, ...)

## S3 method for class 'step_rose'
required_pkgs(x, ...)

## S3 method for class 'step_smote'
required_pkgs(x, ...)

## S3 method for class 'step_smotenc'
required_pkgs(x, ...)

## S3 method for class 'step_tomek'
required_pkgs(x, ...)

## S3 method for class 'step_upsample'
required_pkgs(x, ...)

Arguments

Value

A character vector

SMOTE Algorithm

Description

SMOTE generates new examples of the minority class using nearest neighbors of these cases.

Usage

smote(df, var, k = 5, over_ratio = 1)

Arguments

Details

The parameter neighbors controls the way the new examples are created. For each currently existing minority class example X new examples will be created (this is controlled by the parameter over_ratio as mentioned above). These examples will be generated by using the information from the neighbors nearest neighbor of each example of the minority class. The parameter neighbors controls how many of these neighbor are used. All columns used in this function must be numeric with no missing data.

Value

A data.frame or tibble, depending on type of df.

References

Chawla, N. V., Bowyer, K. W., Hall, L. O., and Kegelmeyer, W. P. (2002). Smote: Synthetic minority over-sampling technique. Journal of Artificial Intelligence Research, 16:321-357.

See Also

step_smote() for step function of this method

Other Direct Implementations: adasyn(), bsmote(), nearmiss(), smotenc(), tomek()

Examples

circle_numeric <- circle_example[, c("x", "y", "class")]

res <- smote(circle_numeric, var = "class")

res <- smote(circle_numeric, var = "class", k = 10)

res <- smote(circle_numeric, var = "class", over_ratio = 0.8)

SMOTENC Algorithm

Description

SMOTENC generates new examples of the minority class using nearest neighbors of these cases, and can handle categorical variables

Usage

smotenc(df, var, k = 5, over_ratio = 1)

Arguments

Details

The parameter neighbors controls the way the new examples are created. For each currently existing minority class example X new examples will be created (this is controlled by the parameter over_ratio as mentioned above). These examples will be generated by using the information from the neighbors nearest neighbor of each example of the minority class. The parameter neighbors controls how many of these neighbor are used. Columns can be numeric and categorical with no missing data.

Value

A data.frame or tibble, depending on type of df.

References

Chawla, N. V., Bowyer, K. W., Hall, L. O., and Kegelmeyer, W. P. (2002). Smote: Synthetic minority over-sampling technique. Journal of Artificial Intelligence Research, 16:321-357.

See Also

step_smotenc() for step function of this method

Other Direct Implementations: adasyn(), bsmote(), nearmiss(), smote(), tomek()

Examples

circle_numeric <- circle_example[, c("x", "y", "class")]

res <- smotenc(circle_numeric, var = "class")

res <- smotenc(circle_numeric, var = "class", k = 10)

res <- smotenc(circle_numeric, var = "class", over_ratio = 0.8)

Apply Adaptive Synthetic Algorithm

Description

step_adasyn() creates a specification of a recipe step that generates synthetic positive instances using ADASYN algorithm.

Usage

step_adasyn(
  recipe,
  ...,
  role = NA,
  trained = FALSE,
  column = NULL,
  over_ratio = 1,
  neighbors = 5,
  skip = TRUE,
  seed = sample.int(10^5, 1),
  id = rand_id("adasyn")
)

Arguments

Details

All columns in the data are sampled and returned by recipes::juice() and recipes::bake().

All columns used in this step must be numeric with no missing data.

When used in modeling, users should strongly consider using the option skip = TRUE so that the extra sampling is not conducted outside of the training set.

Value

An updated version of recipe with the new step added to the sequence of existing steps (if any). For the tidy method, a tibble with columns terms which is the variable used to sample.

Tidying

When you tidy() this step, a tibble is retruned with columns terms and id:

terms

character, the selectors or variables selected

id

character, id of this step

Tuning Parameters

This step has 2 tuning parameters:

• over_ratio: Over-Sampling Ratio (type: double, default: 1)

• neighbors: # Nearest Neighbors (type: integer, default: 5)

Case weights

The underlying operation does not allow for case weights.

References

He, H., Bai, Y., Garcia, E. and Li, S. 2008. ADASYN: Adaptive synthetic sampling approach for imbalanced learning. Proceedings of IJCNN 2008. (IEEE World Congress on Computational Intelligence). IEEE International Joint Conference. pp.1322-1328.

See Also

adasyn() for direct implementation

Other Steps for over-sampling: step_bsmote(), step_rose(), step_smote(), step_smotenc(), step_upsample()

Examples

library(recipes)
library(modeldata)
data(hpc_data)

hpc_data0 <- hpc_data %>%
  select(-protocol, -day)

orig <- count(hpc_data0, class, name = "orig")
orig

up_rec <- recipe(class ~ ., data = hpc_data0) %>%
  # Bring the minority levels up to about 1000 each
  # 1000/2211 is approx 0.4523
  step_adasyn(class, over_ratio = 0.4523) %>%
  prep()

training <- up_rec %>%
  bake(new_data = NULL) %>%
  count(class, name = "training")
training

# Since `skip` defaults to TRUE, baking the step has no effect
baked <- up_rec %>%
  bake(new_data = hpc_data0) %>%
  count(class, name = "baked")
baked

# Note that if the original data contained more rows than the
# target n (= ratio * majority_n), the data are left alone:
orig %>%
  left_join(training, by = "class") %>%
  left_join(baked, by = "class")

library(ggplot2)

ggplot(circle_example, aes(x, y, color = class)) +
  geom_point() +
  labs(title = "Without ADASYN")

recipe(class ~ x + y, data = circle_example) %>%
  step_adasyn(class) %>%
  prep() %>%
  bake(new_data = NULL) %>%
  ggplot(aes(x, y, color = class)) +
  geom_point() +
  labs(title = "With ADASYN")

Apply borderline-SMOTE Algorithm

Description

step_bsmote() creates a specification of a recipe step that generate new examples of the minority class using nearest neighbors of these cases in the border region between classes.

Usage

step_bsmote(
  recipe,
  ...,
  role = NA,
  trained = FALSE,
  column = NULL,
  over_ratio = 1,
  neighbors = 5,
  all_neighbors = FALSE,
  skip = TRUE,
  seed = sample.int(10^5, 1),
  id = rand_id("bsmote")
)

Arguments

Details

This methods works the same way as step_smote(), expect that instead of generating points around every point of of the minority class each point is first being classified into the boxes "danger" and "not". For each point the k nearest neighbors is calculated. If all the neighbors comes from a different class it is labeled noise and put in to the "not" box. If more then half of the neighbors comes from a different class it is labeled "danger.

If all_neighbors = FALSE then points will be generated between nearest neighbors in its own class. If all_neighbors = TRUE then points will be generated between any nearest neighbors. See examples for visualization.

The parameter neighbors controls the way the new examples are created. For each currently existing minority class example X new examples will be created (this is controlled by the parameter over_ratio as mentioned above). These examples will be generated by using the information from the neighbors nearest neighbor of each example of the minority class. The parameter neighbors controls how many of these neighbor are used.

All columns in the data are sampled and returned by recipes::juice() and recipes::bake().

All columns used in this step must be numeric with no missing data.

When used in modeling, users should strongly consider using the option skip = TRUE so that the extra sampling is not conducted outside of the training set.

Value

An updated version of recipe with the new step added to the sequence of existing steps (if any). For the tidy method, a tibble with columns terms which is the variable used to sample.

Tidying

When you tidy() this step, a tibble is retruned with columns terms and id:

terms

character, the selectors or variables selected

id

character, id of this step

Tuning Parameters

This step has 3 tuning parameters:

• over_ratio: Over-Sampling Ratio (type: double, default: 1)

• neighbors: # Nearest Neighbors (type: integer, default: 5)

• all_neighbors: All Neighbors (type: logical, default: FALSE)

Case weights

The underlying operation does not allow for case weights.

References

Hui Han, Wen-Yuan Wang, and Bing-Huan Mao. Borderline-smote: a new over-sampling method in imbalanced data sets learning. In International Conference on Intelligent Computing, pages 878–887. Springer, 2005.

See Also

bsmote() for direct implementation

Other Steps for over-sampling: step_adasyn(), step_rose(), step_smote(), step_smotenc(), step_upsample()

Examples

library(recipes)
library(modeldata)
data(hpc_data)

hpc_data0 <- hpc_data %>%
  select(-protocol, -day)

orig <- count(hpc_data0, class, name = "orig")
orig

up_rec <- recipe(class ~ ., data = hpc_data0) %>%
  # Bring the minority levels up to about 1000 each
  # 1000/2211 is approx 0.4523
  step_bsmote(class, over_ratio = 0.4523) %>%
  prep()

training <- up_rec %>%
  bake(new_data = NULL) %>%
  count(class, name = "training")
training

# Since `skip` defaults to TRUE, baking the step has no effect
baked <- up_rec %>%
  bake(new_data = hpc_data0) %>%
  count(class, name = "baked")
baked

# Note that if the original data contained more rows than the
# target n (= ratio * majority_n), the data are left alone:
orig %>%
  left_join(training, by = "class") %>%
  left_join(baked, by = "class")

library(ggplot2)

ggplot(circle_example, aes(x, y, color = class)) +
  geom_point() +
  labs(title = "Without SMOTE")

recipe(class ~ x + y, data = circle_example) %>%
  step_bsmote(class, all_neighbors = FALSE) %>%
  prep() %>%
  bake(new_data = NULL) %>%
  ggplot(aes(x, y, color = class)) +
  geom_point() +
  labs(title = "With borderline-SMOTE, all_neighbors = FALSE")

recipe(class ~ x + y, data = circle_example) %>%
  step_bsmote(class, all_neighbors = TRUE) %>%
  prep() %>%
  bake(new_data = NULL) %>%
  ggplot(aes(x, y, color = class)) +
  geom_point() +
  labs(title = "With borderline-SMOTE, all_neighbors = TRUE")

Down-Sample a Data Set Based on a Factor Variable

Description

step_downsample() creates a specification of a recipe step that will remove rows of a data set to make the occurrence of levels in a specific factor level equal.

Usage

step_downsample(
  recipe,
  ...,
  under_ratio = 1,
  ratio = deprecated(),
  role = NA,
  trained = FALSE,
  column = NULL,
  target = NA,
  skip = TRUE,
  seed = sample.int(10^5, 1),
  id = rand_id("downsample")
)

Arguments

Details

Down-sampling is intended to be performed on the training set alone. For this reason, the default is skip = TRUE.

If there are missing values in the factor variable that is used to define the sampling, missing data are selected at random in the same way that the other factor levels are sampled. Missing values are not used to determine the amount of data in the minority level

For any data with factor levels occurring with the same frequency as the minority level, all data will be retained.

All columns in the data are sampled and returned by recipes::juice() and recipes::bake().

Keep in mind that the location of down-sampling in the step may have effects. For example, if centering and scaling, it is not clear whether those operations should be conducted before or after rows are removed.

Value

An updated version of recipe with the new step added to the sequence of existing steps (if any). For the tidy method, a tibble with columns terms which is the variable used to sample.

Tidying

When you tidy() this step, a tibble is retruned with columns terms and id:

terms

character, the selectors or variables selected

id

character, id of this step

Tuning Parameters

This step has 1 tuning parameters:

• under_ratio: Under-Sampling Ratio (type: double, default: 1)

Case weights

This step performs an unsupervised operation that can utilize case weights. To use them, see the documentation in recipes::case_weights and the examples on tidymodels.org.

See Also

Other Steps for under-sampling: step_nearmiss(), step_tomek()

Examples

library(recipes)
library(modeldata)
data(hpc_data)

hpc_data0 <- hpc_data %>%
  select(-protocol, -day)

orig <- count(hpc_data0, class, name = "orig")
orig

up_rec <- recipe(class ~ ., data = hpc_data0) %>%
  # Bring the majority levels down to about 1000 each
  # 1000/259 is approx 3.862
  step_downsample(class, under_ratio = 3.862) %>%
  prep()

training <- up_rec %>%
  bake(new_data = NULL) %>%
  count(class, name = "training")
training

# Since `skip` defaults to TRUE, baking the step has no effect
baked <- up_rec %>%
  bake(new_data = hpc_data0) %>%
  count(class, name = "baked")
baked

# Note that if the original data contained more rows than the
# target n (= ratio * majority_n), the data are left alone:
orig %>%
  left_join(training, by = "class") %>%
  left_join(baked, by = "class")

library(ggplot2)

ggplot(circle_example, aes(x, y, color = class)) +
  geom_point() +
  labs(title = "Without downsample")

recipe(class ~ x + y, data = circle_example) %>%
  step_downsample(class) %>%
  prep() %>%
  bake(new_data = NULL) %>%
  ggplot(aes(x, y, color = class)) +
  geom_point() +
  labs(title = "With downsample")

Remove Points Near Other Classes

Description

step_nearmiss() creates a specification of a recipe step that removes majority class instances by undersampling points in the majority class based on their distance to other points in the same class.

Usage

step_nearmiss(
  recipe,
  ...,
  role = NA,
  trained = FALSE,
  column = NULL,
  under_ratio = 1,
  neighbors = 5,
  skip = TRUE,
  seed = sample.int(10^5, 1),
  id = rand_id("nearmiss")
)

Arguments

Details

This method retains the points from the majority class which have the smallest mean distance to the k nearest points in the minority class.

All columns in the data are sampled and returned by recipes::juice() and recipes::bake().

All columns used in this step must be numeric with no missing data.

When used in modeling, users should strongly consider using the option skip = TRUE so that the extra sampling is not conducted outside of the training set.

Value

An updated version of recipe with the new step added to the sequence of existing steps (if any). For the tidy method, a tibble with columns terms which is the variable used to sample.

Tidying

When you tidy() this step, a tibble is retruned with columns terms and id:

terms

character, the selectors or variables selected

id

character, id of this step

Tuning Parameters

This step has 2 tuning parameters:

• under_ratio: Under-Sampling Ratio (type: double, default: 1)

• neighbors: # Nearest Neighbors (type: integer, default: 5)

Case weights

The underlying operation does not allow for case weights.

References

Inderjeet Mani and I Zhang. knn approach to unbalanced data distributions: a case study involving information extraction. In Proceedings of workshop on learning from imbalanced datasets, 2003.

See Also

nearmiss() for direct implementation

Other Steps for under-sampling: step_downsample(), step_tomek()

Examples

library(recipes)
library(modeldata)
data(hpc_data)

hpc_data0 <- hpc_data %>%
  select(-protocol, -day)

orig <- count(hpc_data0, class, name = "orig")
orig

up_rec <- recipe(class ~ ., data = hpc_data0) %>%
  # Bring the majority levels down to about 1000 each
  # 1000/259 is approx 3.862
  step_nearmiss(class, under_ratio = 3.862) %>%
  prep()

training <- up_rec %>%
  bake(new_data = NULL) %>%
  count(class, name = "training")
training

# Since `skip` defaults to TRUE, baking the step has no effect
baked <- up_rec %>%
  bake(new_data = hpc_data0) %>%
  count(class, name = "baked")
baked

# Note that if the original data contained more rows than the
# target n (= ratio * majority_n), the data are left alone:
orig %>%
  left_join(training, by = "class") %>%
  left_join(baked, by = "class")

library(ggplot2)

ggplot(circle_example, aes(x, y, color = class)) +
  geom_point() +
  labs(title = "Without NEARMISS") +
  xlim(c(1, 15)) +
  ylim(c(1, 15))

recipe(class ~ x + y, data = circle_example) %>%
  step_nearmiss(class) %>%
  prep() %>%
  bake(new_data = NULL) %>%
  ggplot(aes(x, y, color = class)) +
  geom_point() +
  labs(title = "With NEARMISS") +
  xlim(c(1, 15)) +
  ylim(c(1, 15))

Apply ROSE Algorithm

Description

step_rose() creates a specification of a recipe step that generates sample of synthetic data by enlarging the features space of minority and majority class example. Using ROSE::ROSE().

Usage

step_rose(
  recipe,
  ...,
  role = NA,
  trained = FALSE,
  column = NULL,
  over_ratio = 1,
  minority_prop = 0.5,
  minority_smoothness = 1,
  majority_smoothness = 1,
  skip = TRUE,
  seed = sample.int(10^5, 1),
  id = rand_id("rose")
)

Arguments

Details

The factor variable used to balance around must only have 2 levels.

The ROSE algorithm works by selecting an observation belonging to class k and generates new examples in its neighborhood is determined by some matrix H_k. Smaller values of these arguments have the effect of shrinking the entries of the corresponding smoothing matrix H_k, Shrinking would be a cautious choice if there is a concern that excessively large neighborhoods could lead to blur the boundaries between the regions of the feature space associated with each class.

All columns in the data are sampled and returned by recipes::juice() and recipes::bake().

When used in modeling, users should strongly consider using the option skip = TRUE so that the extra sampling is not conducted outside of the training set.

Value

An updated version of recipe with the new step added to the sequence of existing steps (if any). For the tidy method, a tibble with columns terms which is the variable used to sample.

Tidying

When you tidy() this step, a tibble is retruned with columns terms and id:

terms

character, the selectors or variables selected

id

character, id of this step

Tuning Parameters

This step has 1 tuning parameters:

• over_ratio: Over-Sampling Ratio (type: double, default: 1)

Case weights

The underlying operation does not allow for case weights.

References

Lunardon, N., Menardi, G., and Torelli, N. (2014). ROSE: a Package for Binary Imbalanced Learning. R Jorunal, 6:82–92.

Menardi, G. and Torelli, N. (2014). Training and assessing classification rules with imbalanced data. Data Mining and Knowledge Discovery, 28:92–122.

See Also

Other Steps for over-sampling: step_adasyn(), step_bsmote(), step_smote(), step_smotenc(), step_upsample()

Examples

library(recipes)
library(modeldata)
data(hpc_data)

hpc_data0 <- hpc_data %>%
  mutate(class = factor(class == "VF", labels = c("not VF", "VF"))) %>%
  select(-protocol, -day)

orig <- count(hpc_data0, class, name = "orig")
orig

up_rec <- recipe(class ~ ., data = hpc_data0) %>%
  step_rose(class) %>%
  prep()

training <- up_rec %>%
  bake(new_data = NULL) %>%
  count(class, name = "training")
training

# Since `skip` defaults to TRUE, baking the step has no effect
baked <- up_rec %>%
  bake(new_data = hpc_data0) %>%
  count(class, name = "baked")
baked

orig %>%
  left_join(training, by = "class") %>%
  left_join(baked, by = "class")

library(ggplot2)

ggplot(circle_example, aes(x, y, color = class)) +
  geom_point() +
  labs(title = "Without ROSE")

recipe(class ~ x + y, data = circle_example) %>%
  step_rose(class) %>%
  prep() %>%
  bake(new_data = NULL) %>%
  ggplot(aes(x, y, color = class)) +
  geom_point() +
  labs(title = "With ROSE")

Apply SMOTE Algorithm

Description

step_smote() creates a specification of a recipe step that generate new examples of the minority class using nearest neighbors of these cases.

Usage

step_smote(
  recipe,
  ...,
  role = NA,
  trained = FALSE,
  column = NULL,
  over_ratio = 1,
  neighbors = 5,
  skip = TRUE,
  seed = sample.int(10^5, 1),
  id = rand_id("smote")
)

Arguments

Details

The parameter neighbors controls the way the new examples are created. For each currently existing minority class example X new examples will be created (this is controlled by the parameter over_ratio as mentioned above). These examples will be generated by using the information from the neighbors nearest neighbor of each example of the minority class. The parameter neighbors controls how many of these neighbor are used.

All columns in the data are sampled and returned by recipes::juice() and recipes::bake().

All columns used in this step must be numeric with no missing data.

When used in modeling, users should strongly consider using the option skip = TRUE so that the extra sampling is not conducted outside of the training set.

Value

An updated version of recipe with the new step added to the sequence of existing steps (if any). For the tidy method, a tibble with columns terms which is the variable used to sample.

Tidying

When you tidy() this step, a tibble is retruned with columns terms and id:

terms

character, the selectors or variables selected

id

character, id of this step

Tuning Parameters

This step has 2 tuning parameters:

• over_ratio: Over-Sampling Ratio (type: double, default: 1)

• neighbors: # Nearest Neighbors (type: integer, default: 5)

Case weights

The underlying operation does not allow for case weights.

References

Chawla, N. V., Bowyer, K. W., Hall, L. O., and Kegelmeyer, W. P. (2002). Smote: Synthetic minority over-sampling technique. Journal of Artificial Intelligence Research, 16:321-357.

See Also

smote() for direct implementation

Other Steps for over-sampling: step_adasyn(), step_bsmote(), step_rose(), step_smotenc(), step_upsample()

Examples

library(recipes)
library(modeldata)
data(hpc_data)

hpc_data0 <- hpc_data %>%
  select(-protocol, -day)

orig <- count(hpc_data0, class, name = "orig")
orig

up_rec <- recipe(class ~ ., data = hpc_data0) %>%
  # Bring the minority levels up to about 1000 each
  # 1000/2211 is approx 0.4523
  step_smote(class, over_ratio = 0.4523) %>%
  prep()

training <- up_rec %>%
  bake(new_data = NULL) %>%
  count(class, name = "training")
training

# Since `skip` defaults to TRUE, baking the step has no effect
baked <- up_rec %>%
  bake(new_data = hpc_data0) %>%
  count(class, name = "baked")
baked

# Note that if the original data contained more rows than the
# target n (= ratio * majority_n), the data are left alone:
orig %>%
  left_join(training, by = "class") %>%
  left_join(baked, by = "class")

library(ggplot2)

ggplot(circle_example, aes(x, y, color = class)) +
  geom_point() +
  labs(title = "Without SMOTE")

recipe(class ~ x + y, data = circle_example) %>%
  step_smote(class) %>%
  prep() %>%
  bake(new_data = NULL) %>%
  ggplot(aes(x, y, color = class)) +
  geom_point() +
  labs(title = "With SMOTE")

Apply SMOTENC algorithm

Description

step_smotenc() creates a specification of a recipe step that generate new examples of the minority class using nearest neighbors of these cases. Gower's distance is used to handle mixed data types. For categorical variables, the most common category along neighbors is chosen.

Usage

step_smotenc(
  recipe,
  ...,
  role = NA,
  trained = FALSE,
  column = NULL,
  over_ratio = 1,
  neighbors = 5,
  skip = TRUE,
  seed = sample.int(10^5, 1),
  id = rand_id("smotenc")
)

Arguments

Details

The parameter neighbors controls the way the new examples are created. For each currently existing minority class example X new examples will be created (this is controlled by the parameter over_ratio as mentioned above). These examples will be generated by using the information from the neighbors nearest neighbor of each example of the minority class. The parameter neighbors controls how many of these neighbor are used.

All columns in the data are sampled and returned by recipes::juice() and recipes::bake().

Columns can be numeric and categorical with no missing data.

When used in modeling, users should strongly consider using the option skip = TRUE so that the extra sampling is not conducted outside of the training set.

Value

An updated version of recipe with the new step added to the sequence of existing steps (if any). For the tidy method, a tibble with columns terms which is the variable used to sample.

Tidying

When you tidy() this step, a tibble is retruned with columns terms and id:

terms

character, the selectors or variables selected

id

character, id of this step

Tuning Parameters

This step has 2 tuning parameters:

• over_ratio: Over-Sampling Ratio (type: double, default: 1)

• neighbors: # Nearest Neighbors (type: integer, default: 5)

Case weights

The underlying operation does not allow for case weights.

References

Chawla, N. V., Bowyer, K. W., Hall, L. O., and Kegelmeyer, W. P. (2002). Smote: Synthetic minority over-sampling technique. Journal of Artificial Intelligence Research, 16:321-357.

See Also

smotenc() for direct implementation

Other Steps for over-sampling: step_adasyn(), step_bsmote(), step_rose(), step_smote(), step_upsample()

Examples

library(recipes)
library(modeldata)
data(hpc_data)

orig <- count(hpc_data, class, name = "orig")
orig

up_rec <- recipe(class ~ ., data = hpc_data) %>%
  step_impute_knn(all_predictors()) %>%
  # Bring the minority levels up to about 1000 each
  # 1000/2211 is approx 0.4523
  step_smotenc(class, over_ratio = 0.4523) %>%
  prep()

training <- up_rec %>%
  bake(new_data = NULL) %>%
  count(class, name = "training")
training

# Since `skip` defaults to TRUE, baking the step has no effect
baked <- up_rec %>%
  bake(new_data = hpc_data) %>%
  count(class, name = "baked")
baked

# Note that if the original data contained more rows than the
# target n (= ratio * majority_n), the data are left alone:
orig %>%
  left_join(training, by = "class") %>%
  left_join(baked, by = "class")

Remove Tomek’s Links

Description

step_tomek() creates a specification of a recipe step that removes majority class instances of tomek links.

Usage

step_tomek(
  recipe,
  ...,
  role = NA,
  trained = FALSE,
  column = NULL,
  skip = TRUE,
  seed = sample.int(10^5, 1),
  id = rand_id("tomek")
)

Arguments

Details

The factor variable used to balance around must only have 2 levels. All other variables must be numerics with no missing data.

A tomek link is defined as a pair of points from different classes and are each others nearest neighbors.

All columns in the data are sampled and returned by recipes::juice() and recipes::bake().

When used in modeling, users should strongly consider using the option skip = TRUE so that the extra sampling is not conducted outside of the training set.

Value

An updated version of recipe with the new step added to the sequence of existing steps (if any). For the tidy method, a tibble with columns terms which is the variable used to sample.

Tidying

When you tidy() this step, a tibble is retruned with columns terms and id:

terms

character, the selectors or variables selected

id

character, id of this step

Case weights

The underlying operation does not allow for case weights.

References

Tomek. Two modifications of cnn. IEEE Trans. Syst. Man Cybern., 6:769-772, 1976.

See Also

tomek() for direct implementation

Other Steps for under-sampling: step_downsample(), step_nearmiss()

Examples

library(recipes)
library(modeldata)
data(hpc_data)

hpc_data0 <- hpc_data %>%
  select(-protocol, -day)

orig <- count(hpc_data0, class, name = "orig")
orig

up_rec <- recipe(class ~ ., data = hpc_data0) %>%
  step_tomek(class) %>%
  prep()

training <- up_rec %>%
  bake(new_data = NULL) %>%
  count(class, name = "training")
training

# Since `skip` defaults to TRUE, baking the step has no effect
baked <- up_rec %>%
  bake(new_data = hpc_data0) %>%
  count(class, name = "baked")
baked

orig %>%
  left_join(training, by = "class") %>%
  left_join(baked, by = "class")

library(ggplot2)

ggplot(circle_example, aes(x, y, color = class)) +
  geom_point() +
  labs(title = "Without Tomek") +
  xlim(c(1, 15)) +
  ylim(c(1, 15))

recipe(class ~ x + y, data = circle_example) %>%
  step_tomek(class) %>%
  prep() %>%
  bake(new_data = NULL) %>%
  ggplot(aes(x, y, color = class)) +
  geom_point() +
  labs(title = "With Tomek") +
  xlim(c(1, 15)) +
  ylim(c(1, 15))

Up-Sample a Data Set Based on a Factor Variable

Description

step_upsample() creates a specification of a recipe step that will replicate rows of a data set to make the occurrence of levels in a specific factor level equal.

Usage

step_upsample(
  recipe,
  ...,
  over_ratio = 1,
  ratio = deprecated(),
  role = NA,
  trained = FALSE,
  column = NULL,
  target = NA,
  skip = TRUE,
  seed = sample.int(10^5, 1),
  id = rand_id("upsample")
)

Arguments

Details

Up-sampling is intended to be performed on the training set alone. For this reason, the default is skip = TRUE.

If there are missing values in the factor variable that is used to define the sampling, missing data are selected at random in the same way that the other factor levels are sampled. Missing values are not used to determine the amount of data in the majority level (see example below).

For any data with factor levels occurring with the same frequency as the majority level, all data will be retained.

All columns in the data are sampled and returned by recipes::juice() and recipes::bake().

Value

An updated version of recipe with the new step added to the sequence of existing steps (if any). For the tidy method, a tibble with columns terms which is the variable used to sample.

Tidying

When you tidy() this step, a tibble is retruned with columns terms and id:

terms

character, the selectors or variables selected

id

character, id of this step

Tuning Parameters

This step has 1 tuning parameters:

• over_ratio: Over-Sampling Ratio (type: double, default: 1)

Case weights

This step performs an unsupervised operation that can utilize case weights. To use them, see the documentation in recipes::case_weights and the examples on tidymodels.org.

See Also

Other Steps for over-sampling: step_adasyn(), step_bsmote(), step_rose(), step_smote(), step_smotenc()

Examples

library(recipes)
library(modeldata)
data(hpc_data)

hpc_data0 <- hpc_data %>%
  select(-protocol, -day)

orig <- count(hpc_data0, class, name = "orig")
orig

up_rec <- recipe(class ~ ., data = hpc_data0) %>%
  # Bring the minority levels up to about 1000 each
  # 1000/2211 is approx 0.4523
  step_upsample(class, over_ratio = 0.4523) %>%
  prep()

training <- up_rec %>%
  bake(new_data = NULL) %>%
  count(class, name = "training")
training

# Since `skip` defaults to TRUE, baking the step has no effect
baked <- up_rec %>%
  bake(new_data = hpc_data0) %>%
  count(class, name = "baked")
baked

# Note that if the original data contained more rows than the
# target n (= ratio * majority_n), the data are left alone:
orig %>%
  left_join(training, by = "class") %>%
  left_join(baked, by = "class")

library(ggplot2)

ggplot(circle_example, aes(x, y, color = class)) +
  geom_point() +
  labs(title = "Without upsample")

recipe(class ~ x + y, data = circle_example) %>%
  step_upsample(class) %>%
  prep() %>%
  bake(new_data = NULL) %>%
  ggplot(aes(x, y, color = class)) +
  geom_jitter(width = 0.1, height = 0.1) +
  labs(title = "With upsample (with jittering)")

Remove Tomek's links

Description

Removed observations that are part of tomek links.

Usage

tomek(df, var)

Arguments

Details

All columns used in this function must be numeric with no missing data.

Value

A data.frame or tibble, depending on type of df.

References

Tomek. Two modifications of cnn. IEEE Trans. Syst. Man Cybern., 6:769-772, 1976.

See Also

step_tomek() for step function of this method

Other Direct Implementations: adasyn(), bsmote(), nearmiss(), smote(), smotenc()

Examples

circle_numeric <- circle_example[, c("x", "y", "class")]

res <- tomek(circle_numeric, var = "class")

tunable methods for themis

Description

These functions define what parameters can be tuned for specific steps. They also define the recommended objects from the dials package that can be used to generate new parameter values and other characteristics.

Usage

## S3 method for class 'step_adasyn'
tunable(x, ...)

## S3 method for class 'step_bsmote'
tunable(x, ...)

## S3 method for class 'step_downsample'
tunable(x, ...)

## S3 method for class 'step_nearmiss'
tunable(x, ...)

## S3 method for class 'step_rose'
tunable(x, ...)

## S3 method for class 'step_smote'
tunable(x, ...)

## S3 method for class 'step_smotenc'
tunable(x, ...)

## S3 method for class 'step_upsample'
tunable(x, ...)

Arguments

Value

A tibble object.