drifter_ml.columnar_tests package

Submodules

drifter_ml.columnar_tests.columnar_tests module

class drifter_ml.columnar_tests.columnar_tests.ColumnarData(historical_data, new_data)

Bases: object

is_normal(column)
kruskal_similar_distribution(column, pvalue_threshold=0.05, num_rounds=3)
ks_2samp_similar_distribution(column, pvalue_threshold=0.05, num_rounds=3)
mann_whitney_u_similar_distribution(column, pvalue_threshold=0.05, num_rounds=3)
mean_similarity(column, tolerance=2)
median_similarity(column, tolerance=2)
pearson_similar_correlation(column, correlation_lower_bound, pvalue_threshold=0.05, num_rounds=3)
spearman_similar_correlation(column, correlation_lower_bound, pvalue_threshold=0.05, num_rounds=3)
trimean(data)
trimean_absolute_deviation(data)
trimean_similarity(column, tolerance=2)
wilcoxon_similar_distribution(column, pvalue_threshold=0.05, num_rounds=3)
class drifter_ml.columnar_tests.columnar_tests.DataSanitization(data)

Bases: object

has_completeness(column, threshold)
has_uniqueness(column, threshold)
is_complete(column)
is_in_range(column, lower_bound, upper_bound, threshold)
is_less_than(column_one, column_two)
is_non_negative(column)
is_unique(column)

Module contents

class drifter_ml.columnar_tests.DataSanitization(data)

Bases: object

has_completeness(column, threshold)
has_uniqueness(column, threshold)
is_complete(column)
is_in_range(column, lower_bound, upper_bound, threshold)
is_less_than(column_one, column_two)
is_non_negative(column)
is_unique(column)
class drifter_ml.columnar_tests.ColumnarData(historical_data, new_data)

Bases: object

is_normal(column)
kruskal_similar_distribution(column, pvalue_threshold=0.05, num_rounds=3)
ks_2samp_similar_distribution(column, pvalue_threshold=0.05, num_rounds=3)
mann_whitney_u_similar_distribution(column, pvalue_threshold=0.05, num_rounds=3)
mean_similarity(column, tolerance=2)
median_similarity(column, tolerance=2)
pearson_similar_correlation(column, correlation_lower_bound, pvalue_threshold=0.05, num_rounds=3)
spearman_similar_correlation(column, correlation_lower_bound, pvalue_threshold=0.05, num_rounds=3)
trimean(data)
trimean_absolute_deviation(data)
trimean_similarity(column, tolerance=2)
wilcoxon_similar_distribution(column, pvalue_threshold=0.05, num_rounds=3)