* DayZSynthesizer

*SDV Enterprise Feature. This feature is only available for licensed, enterprise users. To learn more about the SDV Enterprise features and purchasing a license, visit our website.

The Day Z Synthesizer produces synthetic data from scratch using only the metadata. This allows you start generating synthetic data from day zero: no real data or machine learning required!

from sdv.single_table import DayZSynthesizer

synthesizer = DayZSynthesizer(metadata)
synthetic_data = synthesizer.sample(num_rows=10)

Creating a synthesizer

When creating your synthesizer, you are required to pass in a SingleTableMetadata object as the first argument. Other parameters are optional.

Parameter Reference

locales: A list of locale strings. Any PII columns will correspond to the locales that you provide.

synthesizer = DayZSynthesizer(
    metadata,
    locales=['en_US', 'en_CA', 'fr_CA']
)

Making the data more realistic

By default, this synthesizer will randomly generate data that conforms to your metadata specification. If you'd like to generate more realistic data, you can use the methods below to add guidance.

add_numerical_bounds

Use this method to set lower and upper bounds for numerical columns

Parameters

  • (required) column_name: A string with the name of the column. This must be a numerical column referenced in your metadata.

  • (required) min_value: A float or int representing the minimum value.

  • (required) max_value: A float or int representing the max value

Output (None) The sampled synthetic data will follow the min and max bounds

synthesizer.add_numerical_bounds(
    column_name='room_rate',
    min_value=30.00,
    max_value=5000.00
)

add_datetime_bounds

Use this method to set lower and upper bounds for datetime columns

Parameters

  • (required) column_name: A string with the name of the column. This must be a datetime column referenced in your metadata.

  • (required) start_timestamp: A string representing the earliest allowed datetime. The string must be in the same datetime format as referenced in your metadata.

  • (required) end_timestamp: A string representing the latest allowed datetime. The string must be in the same datetime format as referenced in your metadata.

Output (None) The sampled synthetic data will follow start and end bounds

synthesizer.add_datetime_bounds(
    column_name='checkin_date',
    start_timestamp='01 Jan 2020',
    end_timestamp='31 Dec 2020'
)

set_category_values

Use this method to set the different values that are possible for categorical columns.

Parameters

  • (required) column_name: A string with the name of the column. This must be a categorical column referenced in your metadata.

  • (required) category_values: A list of strings representing the different unique category values that are possible. (If missing values are allowed, use the set_missing_values method instead of listing it here.)

Output (None) The sampled synthetic data will include the category values

synthesizer.set_category_values(
    column_name='room_type',
    category_values=['BASIC', 'DELUXE', 'SUITE'],
)

set_missing_values

Use this method to set the proportion of missing values to generate in a column

Parameters

  • (required) column_name: A string representing the name of the column

  • (required) missing_values_proportion: A float representing the proportion of missing values

    • Any float between 0.0 and 1.0: Randomly create this proportion of missing values in the column

synthesizer.set_missing_values(
    column_name='room_type',
    missing_values_proportion=0.1
)

get_parameters

Use this method to get a dictionary of all the parameters used to make synthetic data -- those you have provided as well as the default ones.

Parameters

  • output_filepath: A string representing the name of the file to write the parameters to. We recommend storing this as a JSON file. Defaults to None, meaning that no output filepath is written.

Output A dictionary representing all the custom parameters added to the synthesizer.

{
    'locales': ['en_US', 'en_CA', 'fr_CA'],
    'columns': {
        'room_rate': {
            'min_value': 30.00,
            'max_value': 500.00
        },
        'checkin_date': {
            'start_timestamp':'01 Jan 2020',
            'end_timestamp':'31 Dec 2020'
        },
        'room_type': {
            'category_values': ['BASIC', 'DELUXE', 'SUITE'],
            'missing_values_proportion': 0.1
        }
    },
    ...
}

Saving your synthesizer

Save your synthesizer for future use

save

Use this function to save your synthesizer as a Python pickle file.

Parameters

  • (required) filepath: A string describing the filepath where you want to save your synthesizer. Make sure this ends in .pkl

Output (None) The file will be saved at the desired location

synthesizer.save(
    filepath='my_synthesizer.pkl'
)

DayZSynthesizer.load

Use this function to load a synthesizer from a Python pickle file

Parameters

  • (required) filepath: A string describing the filepath of your saved synthesizer

Output Your synthesizer, as a DayZSynthesizer object

from sdv.single_table import DayZSynthesizer

synthesizer = DayZSynthesizer.load(
    filepath='my_synthesizer.pkl'
)

Sample synthetic data

Sample any amount of synthetic data

sample

Use this method to sample synthetic data

Parameters

  • (required) num_rows: An integer >0 that specifies the number of rows to synthesize

Output A pandas DataFrame object with synthetic data

What's next?

This synthesizer has limited functionality. It is not compatible with conditional sampling or constraints

If you wish to use these features, we recommend using real data and machine learning to train a GaussianCopulaSynthesizer.

Last updated

Copyright (c) 2023, DataCebo, Inc.