Scribd
Upload
36 views3 pages

Import SDV

The document discusses loading multi-table demo data using SDV, generating a synthetic dataset from the demo data using an HMASynthesizer, and evaluating the quality of the synthetic data.

Uploaded by

Hari Sudhan
Copyright
© © All Rights Reserved
We take content rights seriously. If you suspect this is your content, claim it here.
Available Formats
Download as TXT, PDF, TXT or read online on Scribd
36 views3 pages

Import SDV

The document discusses loading multi-table demo data using SDV, generating a synthetic dataset from the demo data using an HMASynthesizer, and evaluating the quality of the synthetic data.

Uploaded by

Hari Sudhan
Copyright
© © All Rights Reserved
We take content rights seriously. If you suspect this is your content, claim it here.
Available Formats
Download as TXT, PDF, TXT or read online on Scribd
You are on page 1/ 3

import sdv

from sdv.datasets.local import load_csvs


from sdv.metadata import MultiTableMetadata
metadata = MultiTableMetadata()

from sdv.datasets.local import load_csvs

# assume that my_folder contains many CSV files including


# guests.csv, hotels.csv
datasets = load_csvs(folder_name='my_folder/')

guests_table = data['guests']
hotels_table = data['hotels']

/*Imports the SDV Package*/


import sdv

/*Import all the available demos*/


from sdv.datasets.demo import get_available_demos

/*List the available demos*/


get_available_demos(modality='multi_table')

/*Import download utility*/


from sdv.datasets.demo import download_demo

/*define demo package*/


data, metadata = download_demo(
modality='multi_table',
dataset_name='Chess_v1',
output_folder_name='C:/Users/Haric/Documents/basket/Chess_new/'
)

aunt = data['aunt']
aunt1 = data['aunt1']
aunt2 = data['aunt2']
aunt3 = data['aunt3']
aunt4 = data['aunt4']
aunt5 = data['aunt5']
brother = data['brother']
brother1 = data['brother1']
brother2 = data['brother2']
brother3 = data['brother3']
brother4 = data['brother4']
brother5 = data['brother5']
daughter = data['daughter']
daughter1 = data['daughter1']
daughter2 = data['daughter2']
daughter3 = data['daughter3']
daughter4 = data['daughter4']
daughter5 = data['daughter5']
father = data['father']
father1 = data['father1']
father2 = data['father2']
father3 = data['father3']
father4 = data['father4']
father5 = data['father5']
mother = data['mother']
mother1 = data['mother1']
mother2 = data['mother2']
mother3 = data['mother3']
mother4 = data['mother4']
mother5 = data['mother5']
husband = data['husband']
husband1 = data['husband1']
husband2 = data['husband2']
husband3 = data['husband3']
husband4 = data['husband4']
husband5 = data['husband5']
nephew = data['nephew']
nephew1 = data['nephew1']
nephew2 = data['nephew2']
nephew3 = data['nephew3']
nephew4 = data['nephew4']
nephew5 = data['nephew5']
niece = data['niece']
niece1 = data['niece1']
niece2 = data['niece2']
niece3 = data['niece3']
niece4 = data['niece4']
niece5 = data['niece5']
sister = data['sister']
sister1 = data['sister1']
sister2 = data['sister2']
sister3 = data['sister3']
sister4 = data['sister4']
sister5 = data['sister5']
son = data['son']
son1 = data['son1']
son2 = data['son2']
son3 = data['son3']
son4 = data['son4']
son5 = data['son5']
uncle = data['uncle']
uncle1 = data['uncle1']
uncle2 = data['uncle2']
uncle3 = data['uncle3']
uncle4 = data['uncle4']
uncle5 = data['uncle5']
wife = data['wife']
person = data['person']

python_dict = metadata.to_dict()

metadata.visualize(
show_table_details=True,
show_relationship_labels=True,
output_filepath='C:/Users/Haric/Documents/basket/mooney/my_metadata.png'
)
metadata.validate()

from sdv.metadata import MultiTableMetadata

metadata = MultiTableMetadata.load_from_json(filepath='C:/Users/Haric/Documents/
basket/mooney/metadata.json')

from sdv.multi_table import HMASynthesizer

synthesizer = HMASynthesizer(metadata)

synthesizer.fit(data)

synthetic_data = synthesizer.sample(scale=1)

synthetic_data['aunt'].head(3)

synthetic_data['game'].to_csv('C:/Users/Haric/Documents/basket/Chess/game_syn.csv')

from sdv.evaluation.multi_table import evaluate_quality

quality_report = evaluate_quality(
data,
synthetic_data,
metadata,
verbose=False
)

quality_report.get_visualization('Column Shapes', table_name='game')

from sdv.evaluation.multi_table import get_column_plot

fig = get_column_plot(
real_data=data,
synthetic_data=synthetic_data,
column_name='name',
table_name='person',
metadata=metadata
)

fig.show()

You might also like