dqm package

Subpackages

• dqm.completeness package
  • Submodules
  • dqm.completeness.metric module
    • completeness_tabular()
    • data_completion()
    • DataCompleteness
      • DataCompleteness.completeness_tabular()
      • DataCompleteness.data_completion()
      • DataCompleteness.completeness_tabular()
      • DataCompleteness.data_completion()
  • Module contents
• dqm.diversity package
  • Submodules
  • dqm.diversity.diversity module
    • DiversityCalculator
      • DiversityCalculator.compute_diversity()
      • DiversityCalculator.compute_diversity()
      • DiversityCalculator.validate_inputs()
  • dqm.diversity.metric module
    • DiversityIndexCalculator
      • DiversityIndexCalculator.num()
      • DiversityIndexCalculator.simpson()
      • DiversityIndexCalculator.prob()
      • DiversityIndexCalculator.gini()
      • DiversityIndexCalculator.RD()
      • DiversityIndexCalculator.gini()
      • DiversityIndexCalculator.num()
      • DiversityIndexCalculator.prob()
      • DiversityIndexCalculator.simpson()
  • Module contents
• dqm.domain_gap package
  • Submodules
  • dqm.domain_gap.custom_datasets module
    • ImagesFromFolderDataset
    • ImagesWithLabelsDataset
  • dqm.domain_gap.metrics module
    • CMD
      • CMD.compute()
    • FID
      • FID.calculate_statistics()
      • FID.compute_image_distance()
    • KLMVN
      • KLMVN.calculate_statistics()
      • KLMVN.compute_image_distance()
      • KLMVN.klmvn()
      • KLMVN.regularize_covariance()
    • MMD
      • MMD.compute()
    • Metric
      • Metric.compute()
    • ProxyADistance
      • ProxyADistance.adapt_format_like_pred()
      • ProxyADistance.compute_image_distance()
      • ProxyADistance.function_pad()
    • RMSELoss
      • RMSELoss.mse
      • RMSELoss.eps
      • RMSELoss.forward()
      • RMSELoss.forward()
    • Wasserstein
      • Wasserstein.compute_1D_distance()
      • Wasserstein.compute_cov_matrix()
      • Wasserstein.compute_slice_wasserstein_distance()
  • dqm.domain_gap.utils module
    • compute_features()
    • construct_dataloader()
    • display_resume()
    • extract_nth_layer_feature()
    • generate_transform()
    • load_config()
    • load_model()
  • Module contents
• dqm.representativeness package
  • Submodules
  • dqm.representativeness.metric module
    • DistributionAnalyzer
      • DistributionAnalyzer.chisquare_test()
      • DistributionAnalyzer.kolmogorov()
      • DistributionAnalyzer.shannon_entropy()
      • DistributionAnalyzer.grte()
      • DistributionAnalyzer.chisquare_test()
      • DistributionAnalyzer.grte()
      • DistributionAnalyzer.kolmogorov()
      • DistributionAnalyzer.shannon_entropy()
  • dqm.representativeness.utils module
    • DiscretisationParams
      • DiscretisationParams.__init__()
      • DiscretisationParams.to_dict()
      • DiscretisationParams.get_data()
      • DiscretisationParams.get_data()
      • DiscretisationParams.to_dict()
    • VariableAnalysis
      • VariableAnalysis.variable_counting()
      • VariableAnalysis.countplot()
      • VariableAnalysis.discretisation()
      • VariableAnalysis.normal_discretization()
      • VariableAnalysis.data_processing_for_chisqure_test()
      • VariableAnalysis.uniform_discretization()
      • VariableAnalysis.discretisation_intervals()
      • VariableAnalysis.delete_na()
      • VariableAnalysis.expected()
      • VariableAnalysis.expected_hist()
      • VariableAnalysis.observed_hist()
      • VariableAnalysis.countplot()
      • VariableAnalysis.data_processing_for_chisqure_test()
      • VariableAnalysis.delete_na()
      • VariableAnalysis.discretisation()
      • VariableAnalysis.discretisation_intervals()
      • VariableAnalysis.expected()
      • VariableAnalysis.normal_discretization()
      • VariableAnalysis.observed_hist()
      • VariableAnalysis.uniform_discretization()
      • VariableAnalysis.variable_counting()
  • Module contents
• dqm.utils package
  • Submodules
  • dqm.utils.twe_logger module
    • get_logger()
    • log_str_to_level()
  • Module contents

Submodules

dqm.main module

This script is the entry point for using DQM with command line and docker

dqm.main.load_dataframe(config_dict)

This function loads a pandas dataframe from the config dict passed as input. This config dict comes from a pipeline configuration: An example of such pipeline is present in examples/ folder

Parameters:

config_dict (dict) – Dict containing a metric configuration

dqm.main.load_raw_data(file, separator)

This function load a raw data file content as a pandas dataframe

Parameters:

• file (str) – Path of the file to load

• separator (str) – Separator to use when processing csv and txt format file

Returns:

Output dataframe

Return type:

df (pandas.DataFrame)

dqm.main.main()

Main script of DQM component:

Parameters:

• pipeline_config_path (str) – Path to the pipeline definition you want to apply

• result_file_path – (str): Path the output YAML file where all computed metrics scores are stored

Module contents