Implementation:Sdv dev SDV BaseSynthesizer Add Constraints

Knowledge Sources	SDV SDV Documentation
Domains	Data_Quality, Synthetic_Data
Last Updated	2026-02-14 00:00 GMT

Overview

Concrete tool for registering constraint objects with a synthesizer, provided by the SDV library.

Description

The BaseSynthesizer.add_constraints method accepts a list of constraint objects, classifies each as chained or reject-sampling, wraps ProgrammableConstraint instances, and updates the synthesizer's metadata and DataProcessor.

Usage

Call on a synthesizer before fitting with a list of constraint objects.

Code Reference

Source Location

• Repository: SDV
• File: sdv/single_table/base.py
• Lines: L450-483

Signature

def add_constraints(self, constraints):
    """Add constraint-augmented generation constraints to the synthesizer.

    Args:
        constraints (list): List of constraint objects.
    """

Import

from sdv.single_table import GaussianCopulaSynthesizer
# add_constraints is called as: synthesizer.add_constraints([constraint1, constraint2])

I/O Contract

Inputs

Name	Type	Required	Description
constraints	list	Yes	List of BaseConstraint or ProgrammableConstraint instances

Outputs

Name	Type	Description
(mutates self)	None	Registers constraints; updates metadata and DataProcessor

Usage Examples

from sdv.cag import Inequality, FixedCombinations

constraints = [
    Inequality(low_column_name='start_date', high_column_name='end_date'),
    FixedCombinations(column_names=['country', 'city']),
]

synthesizer.add_constraints(constraints)
synthesizer.fit(data)

Related Pages

Implements Principle

• Principle:Sdv_dev_SDV_Constraint_Integration

Requires Environment

• Environment:Sdv_dev_SDV_Python_Runtime