LearnKit

v0.1.6

LearnKit v0.1.6 LearnKit.NaiveBayes.Gaussian View Source

Module for Gaussian NB algorithm

Link to this section Summary

Types

data_set()
feature()
features()
fit_data()
fit_feature()
fit_features()
label()
point()
prediction()
predictions()

Functions

add_train_data(gaussian, arg)

Add train data to classifier

fit(gaussian)

Fit train data

new()

Creates classifier with empty data_set

new(data_set)

Creates classifier with data_set

normalize_train_data(gaussian, type \\ "none")

Normalize train data

predict(gaussian, feature)

Return exact prediction for the feature

predict_proba(gaussian, feature)

Return probability estimates for the feature

score(gaussian)

Returns the mean accuracy on the given test data and labels

Link to this section Types

Link to this type data_set() View Source 
data_set() :: [{label(), features()}]
Link to this type feature() View Source 
feature() :: [integer()]
Link to this type features() View Source 
features() :: [feature()]
Link to this type fit_data() View Source 
fit_data() :: [{label(), fit_features()}]
Link to this type fit_feature() View Source 
fit_feature() :: %{
  mean: float(),
  standard_deviation: float(),
  variance: float()
}
Link to this type fit_features() View Source 
fit_features() :: [fit_feature()]
Link to this type prediction() View Source 
prediction() :: {label(), number()}
Link to this type predictions() View Source 
predictions() :: [prediction()]

Link to this section Functions

Link to this function add_train_data(gaussian, arg) View Source 
add_train_data(
  %LearnKit.NaiveBayes.Gaussian{data_set: data_set(), fit_data: term()},
  point()
) :: %LearnKit.NaiveBayes.Gaussian{data_set: data_set(), fit_data: term()}

Add train data to classifier

Parameters

  • classifier: %LearnKit.NaiveBayes.Gaussian{}
  • train data: tuple with label and feature

Examples 

iex> classifier = classifier |> LearnKit.NaiveBayes.Gaussian.add_train_data({:a1, [-1, -1]})
%LearnKit.NaiveBayes.Gaussian{data_set: [a1: [[-1, -1]]], fit_data: []}
Link to this function fit(gaussian) View Source 
fit(%LearnKit.NaiveBayes.Gaussian{data_set: data_set(), fit_data: term()}) ::
  %LearnKit.NaiveBayes.Gaussian{data_set: data_set(), fit_data: fit_data()}

Fit train data

Parameters

  • classifier: %LearnKit.NaiveBayes.Gaussian{}

Examples 

iex> classifier = classifier |> LearnKit.NaiveBayes.Gaussian.fit
%LearnKit.NaiveBayes.Gaussian{
  data_set: [a1: [[-1, -1]]],
  fit_data: [
    a1: [
      %{mean: -1.0, standard_deviation: 0.0, variance: 0.0},
      %{mean: -1.0, standard_deviation: 0.0, variance: 0.0}
    ]
  ]
}
Link to this function new() View Source 
new() :: %LearnKit.NaiveBayes.Gaussian{data_set: [], fit_data: term()}

Creates classifier with empty data_set

Examples 

iex> classifier = LearnKit.NaiveBayes.Gaussian.new
%LearnKit.NaiveBayes.Gaussian{data_set: [], fit_data: []}
Link to this function new(data_set) View Source 
new(data_set()) :: %LearnKit.NaiveBayes.Gaussian{
  data_set: data_set(),
  fit_data: term()
}

Creates classifier with data_set

Parameters

  • data_set: Keyword list with labels and features in tuples

Examples 

iex> classifier = LearnKit.NaiveBayes.Gaussian.new([{:a1, [[1, 2], [2, 3]]}, {:b1, [[-1, -2]]}])
%LearnKit.NaiveBayes.Gaussian{data_set: [a1: [[1, 2], [2, 3]], b1: [[-1, -2]]], fit_data: []}
Link to this function normalize_train_data(gaussian, type \\ "none") View Source 
normalize_train_data(
  %LearnKit.NaiveBayes.Gaussian{data_set: data_set(), fit_data: term()},
  String.t()
) :: %LearnKit.NaiveBayes.Gaussian{data_set: data_set(), fit_data: fit_data()}

Normalize train data

Parameters

  • classifier: %LearnKit.NaiveBayes.Gaussian{}
  • type: none/minimax/z_normalization, default is none, optional

Examples 

iex> classifier = classifier |> LearnKit.NaiveBayes.Gaussian.normalize_train_data("minimax")
%LearnKit.NaiveBayes.Gaussian{
  data_set: [a1: [[0.6666666666666666, 0.8], [1.0, 1.0]], b1: [[0.0, 0.0]]],
  fit_data: []
}
Link to this function predict(gaussian, feature) View Source 
predict(
  %LearnKit.NaiveBayes.Gaussian{data_set: term(), fit_data: fit_data()},
  feature()
) :: {:ok, prediction()}

Return exact prediction for the feature

Parameters

  • classifier: %LearnKit.NaiveBayes.Gaussian{}

Examples 

iex> classifier |> LearnKit.NaiveBayes.Gaussian.predict([1, 2])
{:ok, {:a1, 0.334545454}}
Link to this function predict_proba(gaussian, feature) View Source 
predict_proba(
  %LearnKit.NaiveBayes.Gaussian{data_set: term(), fit_data: fit_data()},
  feature()
) :: {:ok, predictions()}

Return probability estimates for the feature

Parameters

  • classifier: %LearnKit.NaiveBayes.Gaussian{}

Examples 

iex> classifier |> LearnKit.NaiveBayes.Gaussian.predict_proba([1, 2])
{:ok, [a1: 0.0359, a2: 0.0039]}
Link to this function score(gaussian) View Source 
score(%LearnKit.NaiveBayes.Gaussian{data_set: data_set(), fit_data: fit_data()}) ::
  {:ok, number()}

Returns the mean accuracy on the given test data and labels

Parameters

  • classifier: %LearnKit.NaiveBayes.Gaussian{}

Examples 

iex> classifier |> LearnKit.NaiveBayes.Gaussian.score
{:ok, 0.857143}