Monotonicity Analysis

mlbugdetection.monotonic.check_monotonicity_multiple_samples(model, samples, feature, start, stop, step=1)

Monotonicity Analysis for multiple examples

Parameters

• model (sklearn model or str) – Model that will be used to make predictions. Could be a model object or a path to a model file.

• samples (pandas.DataFrame) – Pandas DataFrame containing two or more rows that will be used as base point.

• feature (str) – Name of the feature being analysed.

• start (int) – The starting value of the feature’s interval.

• stop (int) – The end value of the feature’s interval.

• step (float, default=1) – Size of the step between ranges “start” and “stop”. Ex: step = 0.1 between ranges 0 and 1 will result in [0 0.1 0.2 0.3 0.4 0.5 0.6 0.7 0.8 0.9]

Returns

• AnalysisReport object with following attributes – For more information: >>> from mlbugdetection.analysis_report import AnalysisReport >>> help(AnalysisReport)

• model_name (str) – Name of the model being analysed.

• analysed_feature (str) – Name of the feature being analysed.

• feature_range (tuple) – Range of values of the feature being analysed: (start, stop).

• metrics (dictionary) – Dictionary with all the calculated metrics, such as:

  ’monotonic’bool

  If the list of values is monotonic.

  ’monotonic_mse’: float

  MSE between the list of values and it`s closest monotonic aproximation.

  ’monotonic_means_std’: float

  Standard deviation of the means of the predictions probabilities.

• graphs (List) – List of all the figures created.

mlbugdetection.monotonic.check_monotonicity_single_sample(model, sample, feature, start, stop, step=1)

Monotonicity Analysis for a single example

Parameters

• model (sklearn model or str) – Model that will be used to make predictions. Could be a model object or a path to a model file.

• sample (pandas.DataFrame) – Pandas DataFrame containing one row that will be used as base point.

• feature (str) – Name of the feature being analysed.

• start (int) – The starting value of the feature’s interval.

• stop (int) – The end value of the feature’s interval.

• step (float, default=1) – Size of the step between ranges “start” and “stop”. Ex: step = 0.1 between ranges 0 and 1 will result in [0 0.1 0.2 0.3 0.4 0.5 0.6 0.7 0.8 0.9]

Returns

• AnalysisReport object with following attributes – For more information: >>> from mlbugdetection.analysis_report import AnalysisReport >>> help(AnalysisReport)

• model_name (str) – Name of the model being analysed.

• analysed_feature (str) – Name of the feature being analysed.

• feature_range (tuple) – Range of values of the feature being analysed: (start, stop).

• metrics (dictionary) – Dictionary with all the calculated metrics, such as:

  ’monotonic’bool

  If the list of values is monotonic.

  ’monotonic_mse’: float

  MSE between the list of values and it`s closest monotonic aproximation.

• graphs (List) – List of all the figures created.

mlbugdetection.monotonic.monotonicity_mse(predictions)

Monotonicity Mean Square Error

Calculates the MSE between a list of prediction brobabilities and the closest monotonic version of this list.

Parameters

predictions (List) – List of prediction probabilities calculated on the check_monotonicity function.

Returns

• desc | asc (List) – List of closest monotonic version of “predictions”.

• mse_desc | mse_as (int) – MSE between “predictions” and desc/asc.