OutlierIdentifiers 0.1.2

Outlier identifiers functions package.

OutlierIdentifiers

In brief

This a Python package for 1D outlier identifier functions. If follows closely the Wolfram Language (WL) paclet [AAp1], the R package [AAp2], and the Raku package [AAp3].

Here is a Jupyter notebook with usage examples: "OutlierIdentifiers-guide.ipynb"; (Markdown version).

---

Installation

From PyPI.org:

python3 -m pip install OutlierIdentifiers

From GitHub:

python3 -m pip install git+https://github.com/antononcube/Python-packages.git#egg=OutlierIdentifiers\&subdirectory=OutlierIdentifiers

---

Usage examples

Load packages:

import numpy as np
import plotly.graph_objects as go

from OutlierIdentifiers import *

Generate a vector with random numbers:

np.random.seed(14)
vec = np.random.normal(loc=10, scale=20, size=30)
print(vec)

[ 41.02678223  11.58372049  13.47953057   8.55326868 -30.086588
  12.89355626 -20.02337245  14.22218902  -1.16410111  31.6905813
   6.27421752  10.2932275  -11.51138939  22.84504148   6.39326577
  22.40600507  26.21948669  25.55871733   5.25020644 -27.83824691
 -13.44243588  26.72413943  30.18546801  35.86198722  -0.98662331
  -9.6342573   28.29345516  27.46140757  10.44222283   9.91712833]

Plot the vector:

# Create a scatter plot with markers
fig = go.Figure(data=go.Scatter(y=vec, mode='markers'))

# Add labels and title
fig.update_layout(title='Vector of Numbers', xaxis_title='Index', yaxis_title='Value', template = "plotly_dark")

# Display the plot
fig.show()

Find outlier positions:

outlier_identifier(vec, identifier=hampel_identifier_parameters)

array([ True, False, False, False,  True, False,  True, False, False,
        True, False, False,  True, False, False, False, False, False,
       False,  True,  True, False, False,  True, False,  True, False,
       False, False, False])

Find outlier values:

outlier_identifier(vec, identifier=hampel_identifier_parameters, value = True)

array([ 41.02678223, -30.086588  , -20.02337245,  31.6905813 ,
       -11.51138939, -27.83824691, -13.44243588,  35.86198722,
        -9.6342573 ])

Find top outlier positions and values:

outlier_identifier(vec, identifier = lambda v: top_outliers(hampel_identifier_parameters(v)))

array([ True, False, False, False, False, False, False, False, False,
        True, False, False, False, False, False, False, False, False,
       False, False, False, False, False,  True, False, False, False,
       False, False, False])

outlier_identifier(vec, identifier = lambda v: top_outliers(hampel_identifier_parameters(v)), value=True)

array([41.02678223, 31.6905813 , 35.86198722])

Find bottom outlier positions and values (using quartiles-based identifier):

outlier_identifier(vec, identifier = lambda v: bottom_outliers(quartile_identifier_parameters(v)))

array([False, False, False, False,  True, False,  True, False, False,
       False, False, False, False, False, False, False, False, False,
       False,  True, False, False, False, False, False, False, False,
       False, False, False])

outlier_identifier(vec, identifier = lambda v: bottom_outliers(quartile_identifier_parameters(v)), value=True)

array([-30.086588  , -20.02337245, -27.83824691])

Here is another way to get the outlier values:

vec[pred]

array([-30.086588  , -20.02337245, -27.83824691])

The available outlier parameters functions are:

• hampel_identifier_parameters
• splus_quartile_identifier_parameters
• quartile_identifier_parameters

[ f(vec) for f in (hampel_identifier_parameters, splus_quartile_identifier_parameters, quartile_identifier_parameters)]

[(-8.796653643076334, 30.822596969354976),
 (-37.649981209714, 64.27685968784428),
 (-14.46873856125025, 36.49468188752889)]

---

References

[AA1] Anton Antonov, "Outlier detection in a list of numbers", (2013), MathematicaForPrediction at WordPress.

[AAp1] Anton Antonov, OutlierIdentifiers WL paclet, (2023), Wolfram Language Paclet Repository.

[AAp2] Anton Antonov, OutlierIdentifiers R package, (2019), R-packages at GitHub/antononcube.

[AAp3] Anton Antonov, OutlierIdentifiers Raku package, (2022), GitHub/antononcube.