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)¶