yardstick: precision – R documentation

Become an expert in R — Interactive courses, Cheat Sheets, certificates and more!

precision

Precision

Description

These functions calculate the precision() of a measurement system for finding relevant documents compared to reference results (the truth regarding relevance). Highly related functions are recall() and f_meas().

Usage

precision(data, ...)

## S3 method for class 'data.frame'
precision(
  data,
  truth,
  estimate,
  estimator = NULL,
  na_rm = TRUE,
  event_level = yardstick_event_level(),
  ...
)

precision_vec(
  truth,
  estimate,
  estimator = NULL,
  na_rm = TRUE,
  event_level = yardstick_event_level(),
  ...
)

Arguments

`data`	Either a `data.frame` containing the `truth` and `estimate` columns, or a `table`/`matrix` where the true class results should be in the columns of the table.
`...`	Not currently used.
`truth`	The column identifier for the true class results (that is a `factor`). This should be an unquoted column name although this argument is passed by expression and supports quasiquotation (you can unquote column names). For `_vec()` functions, a `factor` vector.
`estimate`	The column identifier for the predicted class results (that is also `factor`). As with `truth` this can be specified different ways but the primary method is to use an unquoted variable name. For `_vec()` functions, a `factor` vector.
`estimator`	One of: `"binary"`, `"macro"`, `"macro_weighted"`, or `"micro"` to specify the type of averaging to be done. `"binary"` is only relevant for the two class case. The other three are general methods for calculating multiclass metrics. The default will automatically choose `"binary"` or `"macro"` based on `estimate`.
`na_rm`	A `logical` value indicating whether `NA` values should be stripped before the computation proceeds.
`event_level`	A single string. Either `"first"` or `"second"` to specify which level of `truth` to consider as the "event". This argument is only applicable when `estimator = "binary"`. The default uses an internal helper that generally defaults to `"first"`, however, if the deprecated global option `yardstick.event_first` is set, that will be used instead with a warning.

Details

The precision is the percentage of predicted truly relevant results of the total number of predicted relevant results and characterizes the "purity in retrieval performance" (Buckland and Gey, 1994).

When the denominator of the calculation is 0, precision is undefined. This happens when both # true_positive = 0 and # false_positive = 0 are true, which mean that there were no predicted events. When computing binary precision, a NA value will be returned with a warning. When computing multiclass precision, the individual NA values will be removed, and the computation will procede, with a warning.

Value

A tibble with columns .metric, .estimator, and .estimate and 1 row of values.

For grouped data frames, the number of rows returned will be the same as the number of groups.

For precision_vec(), a single numeric value (or NA).

Relevant Level

There is no common convention on which factor level should automatically be considered the "event" or "positive" result when computing binary classification metrics. In yardstick, the default is to use the first level. To alter this, change the argument event_level to "second" to consider the last level of the factor the level of interest. For multiclass extensions involving one-vs-all comparisons (such as macro averaging), this option is ignored and the "one" level is always the relevant result.

Multiclass

Macro, micro, and macro-weighted averaging is available for this metric. The default is to select macro averaging if a truth factor with more than 2 levels is provided. Otherwise, a standard binary calculation is done. See vignette("multiclass", "yardstick") for more information.

Implementation

Suppose a 2x2 table with notation:

	Reference
Predicted	Relevant	Irrelevant
Relevant	A	B
Irrelevant	C	D

The formulas used here are:

recall = A/(A+C)

precision = A/(A+B)

F_{meas} = (1+β^2) * precision * recall/((β^2 * precision)+recall)

See the references for discussions of the statistics.

Author(s)

Max Kuhn

References

Buckland, M., & Gey, F. (1994). The relationship between Recall and Precision. Journal of the American Society for Information Science, 45(1), 12-19.

Powers, D. (2007). Evaluation: From Precision, Recall and F Factor to ROC, Informedness, Markedness and Correlation. Technical Report SIE-07-001, Flinders University

Examples

# Two class
data("two_class_example")
precision(two_class_example, truth, predicted)

# Multiclass
library(dplyr)
data(hpc_cv)

hpc_cv %>%
  filter(Resample == "Fold01") %>%
  precision(obs, pred)

# Groups are respected
hpc_cv %>%
  group_by(Resample) %>%
  precision(obs, pred)

# Weighted macro averaging
hpc_cv %>%
  group_by(Resample) %>%
  precision(obs, pred, estimator = "macro_weighted")

# Vector version
precision_vec(
  two_class_example$truth,
  two_class_example$predicted
)

# Making Class2 the "relevant" level
precision_vec(
  two_class_example$truth,
  two_class_example$predicted,
  event_level = "second"
)

yardstick

Tidy Characterizations of Model Performance

v0.0.8

MIT + file LICENSE

Authors

Max Kuhn [aut], Davis Vaughan [aut, cre], RStudio [cph]

Initial release