sdv_project.py

# pip install sdv
# importing the necesary libraries
import numpy as np 
import pandas as pd 
import warnings
warnings.filterwarnings('ignore')

# import all 4 sdv models under the single table scenario
from sdv.tabular import GaussianCopula
from sdv.tabular import CTGAN
from sdv.tabular import CopulaGAN
from sdv.tabular import TVAE

# load the models 
model_Gaussian = GaussianCopula(primary_key='CLIENTNUM')
model_CTGAN = CTGAN(primary_key='CLIENTNUM')
model_CopulaGAN = CopulaGAN(primary_key='CLIENTNUM')
model_TVAE = TVAE(primary_key='CLIENTNUM')

# fit the models
model_Gaussian.fit(data)
model_CTGAN.fit(data)
model_CopulaGAN.fit(data)
model_TVAE.fit(data)

# create synthetic data with each fitted model
new_data_model_Gaussian = model_Gaussian.sample(10000)
new_data_model_CTGAN = model_CTGAN.sample(10000)
new_data_model_CopulaGAN = model_CopulaGAN.sample(10000)
new_data_model_TVAE = model_TVAE.sample(10000)

# save each fitted model into a pkl file
model_Gaussian.save('model_Gaussian.pkl')
model_CTGAN.save('model_CTGAN.pkl')
model_CopulaGAN.save('model_CopulaGAN.pkl')
model_TVAE.save('model_TVAE.pkl')