Installation

The synthyverse uses optional installation extras for generator-specific and evaluation-specific dependencies. The core package installs the shared tabular dependencies used by the base generator and preprocessing APIs.

Install multiple extras to use multiple modules in one environment, for example a generator plus the evaluation module.

We strongly advise installing only the extras required for a specific run. Some generator dependencies are heavy or may conflict with each other, so separate virtual environments are recommended when combining many extras.

Available Installation Templates

The following installation templates are available:

Template Name	Category	Installation Command
arf	Generator	pip install synthyverse[arf]
bn	Generator	pip install synthyverse[bn]
cdtd	Generator	pip install synthyverse[cdtd]
ctgan	Generator	pip install synthyverse[ctgan]
smote	Generator	pip install synthyverse[smote]
tabargn	Generator	pip install synthyverse[tabargn]
tabddpm	Generator	pip install synthyverse[tabddpm]
tabsyn	Generator	pip install synthyverse[tabsyn]
tvae	Generator	pip install synthyverse[tvae]
univariate	Generator	pip install synthyverse[univariate]
base	Generator	pip install synthyverse[base]
eval	Evaluation	pip install synthyverse[eval]
full	All	pip install synthyverse[full]

Note: You can install multiple templates by separating them with commas, e.g., pip install synthyverse[ctgan,eval]

General Installation Template

pip install synthyverse[template]

Installation Examples

pip install synthyverse[ctgan]

pip install synthyverse[arf,bn,ctgan,tvae]

pip install synthyverse[ctgan,eval]

First Synthetic Dataset

The high-level SynthyverseGenerator wrapper combines shared preprocessing with a low-level generator.

from synthyverse.generators import SynthyverseGenerator

generator = SynthyverseGenerator(
    "ctgan",
    generator_params={"epochs": 300},
    missing_imputation_method="median",
    random_state=42,
)

generator.fit(X, discrete_features=["category", "target"])
X_syn = generator.generate(1000)

For manual preprocessing, direct generator usage, evaluation metrics, and benchmarking examples, continue to the in-depth usage guide.