Synthetic data with SDV and CTGAN

Updated 2024-03-06 (0533e64)

import pandas as pd
import warnings

warnings.filterwarnings('ignore')

data = pd.read_csv("data/svm-hyperparameters-train-features.csv")

data.head()

	Pclass	Sex	Age	SibSp	Fare
0	3	1	22.0	1	7.2500
1	1	0	38.0	1	71.2833
2	3	0	26.0	0	7.9250
3	1	0	35.0	1	53.1000
4	3	1	35.0	0	8.0500

data.describe(include='all')

	Pclass	Sex	Age	SibSp	Parch	Fare
count	891.000000	891.000000	891.000000	891.000000	891.000000	891.000000
mean	2.308642	0.647587	29.758889	0.523008	0.381594	32.204208
std	0.836071	0.477990	13.002570	1.102743	0.806057	49.693429
min	1.000000	0.000000	0.420000	0.000000	0.000000	0.000000
25%	2.000000	0.000000	22.000000	0.000000	0.000000	7.910400
50%	3.000000	1.000000	30.000000	0.000000	0.000000	14.454200
75%	3.000000	1.000000	35.000000	1.000000	0.000000	31.000000
max	3.000000	1.000000	80.000000	8.000000	6.000000	512.329200

from sdv.tabular import CTGAN

model = CTGAN()

model.fit(data)

new_data = model.sample(200)

new_data.head()

	Pclass	Sex	Age	SibSp	Parch	Fare
0	3	1	31.40	1	1	24.7142
1	3	1	63.44	1	1	0.6418
2	3	0	32.73	0	0	2.8117
3	1	0	27.53	0	0	40.4747
4	2	1	46.31	2	1	104.2955

new_data.describe(include='all')

	Pclass	Sex	Age	SibSp	Parch	Fare
count	200.000000	200.000000	200.000000	200.000000	200.000000	200.000000
mean	2.605000	0.535000	37.052600	0.740000	0.450000	32.947807
std	0.715215	0.500025	15.763543	1.212332	0.692748	56.156945
min	1.000000	0.000000	2.870000	0.000000	0.000000	0.000000
25%	2.000000	0.000000	27.217500	0.000000	0.000000	12.590775
50%	3.000000	1.000000	33.100000	0.000000	0.000000	18.435050
75%	3.000000	1.000000	46.692500	1.000000	1.000000	24.083125
max	3.000000	1.000000	80.000000	6.000000	2.000000	380.969200

from sdv.evaluation import evaluate

evaluate(new_data, data)

0.5513349938501996

model = CTGAN(
    epochs=500,
    batch_size=100,
    generator_dim=(256, 256, 256),
    discriminator_dim=(256, 256, 256)
)

model.fit(data)

new_data = model.sample(200)

new_data.head()

	Pclass	Sex	Age	SibSp	Fare
0	3	1	30.69	0	13.5556
1	2	0	0.42	1	33.1753
2	3	1	30.86	0	34.4404
3	1	1	11.07	0	29.0068
4	2	0	54.40	0	36.6160

new_data.describe(include='all')

	Pclass	Sex	Age	SibSp	Parch	Fare
count	200.000000	200.000000	200.000000	200.000000	200.000000	200.000000
mean	2.350000	0.540000	36.034900	0.365000	0.395000	34.128625
std	0.699964	0.499648	16.254839	0.809352	0.762813	42.191675
min	1.000000	0.000000	0.420000	0.000000	0.000000	2.339800
25%	2.000000	0.000000	28.737500	0.000000	0.000000	10.917300
50%	2.000000	1.000000	31.645000	0.000000	0.000000	19.874400
75%	3.000000	1.000000	49.312500	1.000000	0.000000	35.050750
max	3.000000	1.000000	80.000000	5.000000	2.000000	269.388000

evaluate(new_data, data)

0.6065965175235917