opendp.smartnoise.synthesizers.pytorch.nn package

class opendp.smartnoise.synthesizers.pytorch.nn.PATEGAN(epsilon, delta=1e-05, binary=False, latent_dim=64, batch_size=64, teacher_iters=5, student_iters=5)[source]

Bases: object

generate(n)[source]
train(data, categorical_columns=None, ordinal_columns=None, update_epsilon=None)[source]
class opendp.smartnoise.synthesizers.pytorch.nn.DPGAN(binary=False, latent_dim=64, batch_size=64, epochs=1000, delta=1e-05, epsilon=1.0)[source]

Bases: object

generate(n)[source]
train(data, categorical_columns=None, ordinal_columns=None, update_epsilon=None)[source]
class opendp.smartnoise.synthesizers.pytorch.nn.PATECTGAN(embedding_dim=128, gen_dim=(256, 256), dis_dim=(256, 256), l2scale=1e-06, epochs=300, pack=1, log_frequency=True, disabled_dp=False, target_delta=None, sigma=5, max_per_sample_grad_norm=1.0, verbose=False, loss='cross_entropy', regularization=None, binary=False, batch_size=500, teacher_iters=5, student_iters=5, sample_per_teacher=1000, epsilon=8.0, delta=1e-05, noise_multiplier=0.001, moments_order=100)[source]

Bases: ctgan.synthesizer.CTGANSynthesizer

__init__(embedding_dim=128, gen_dim=(256, 256), dis_dim=(256, 256), l2scale=1e-06, epochs=300, pack=1, log_frequency=True, disabled_dp=False, target_delta=None, sigma=5, max_per_sample_grad_norm=1.0, verbose=False, loss='cross_entropy', regularization=None, binary=False, batch_size=500, teacher_iters=5, student_iters=5, sample_per_teacher=1000, epsilon=8.0, delta=1e-05, noise_multiplier=0.001, moments_order=100)[source]

Initialize self. See help(type(self)) for accurate signature.

generate(n)[source]
train(data, categorical_columns=None, ordinal_columns=None, update_epsilon=None)[source]
w_loss(output, labels)[source]
class opendp.smartnoise.synthesizers.pytorch.nn.DPCTGAN(embedding_dim=128, gen_dim=(256, 256), dis_dim=(256, 256), l2scale=1e-06, batch_size=500, epochs=300, pack=1, log_frequency=True, disabled_dp=False, target_delta=None, sigma=5, max_per_sample_grad_norm=1.0, epsilon=1, verbose=True, loss='cross_entropy')[source]

Bases: ctgan.synthesizer.CTGANSynthesizer

__init__(embedding_dim=128, gen_dim=(256, 256), dis_dim=(256, 256), l2scale=1e-06, batch_size=500, epochs=300, pack=1, log_frequency=True, disabled_dp=False, target_delta=None, sigma=5, max_per_sample_grad_norm=1.0, epsilon=1, verbose=True, loss='cross_entropy')[source]

Differential Private Conditional Table GAN Synthesizer This code adds Differential Privacy to CTGANSynthesizer from https://github.com/sdv-dev/CTGAN

generate(n)[source]
train(data, categorical_columns=None, ordinal_columns=None, update_epsilon=None)[source]