I installed smartnoise-synth 1.0.1 and I'm trying to run:
from snsynth.pytorch.nn import DPGAN
from snsynth.pytorch import PytorchDPSynthesizer
pums = pd.read_csv("PUMS.csv")
synth = PytorchDPSynthesizer(epsilon=3.0, gan=DPGAN())
synth.fit(pums, preprocessor_eps=1.0)
pums_synth = synth.sample(1000)
The error message:
...\lib\site-packages\opacus\privacy_engine.py:638: UserWarning: The sample rate will be defined from batch_size and sample_size.The returned privacy budget will be incorrect.
Is there a way to correct this? Or how can I circumvent this error?
My environment (Windows 11): Python 3.8.17 ipykernel 6.19.2 ipython 8.12.0 numpy 1.24.4 opacus 0.14.0 opendp 0.7.0 pac-synth 0.0.8 smartnoise-sql 1.0.1 smartnoise-synth 1.0.1 torch 1.13.1
I installed smartnoise-synth 1.0.1 and I'm trying to run: from snsynth.pytorch.nn import DPGAN from snsynth.pytorch import PytorchDPSynthesizer pums = pd.read_csv("PUMS.csv") synth = PytorchDPSynthesizer(epsilon=3.0, gan=DPGAN()) synth.fit(pums, preprocessor_eps=1.0) pums_synth = synth.sample(1000)
The error message: ...\lib\site-packages\opacus\privacy_engine.py:638: UserWarning: The sample rate will be defined from
batch_sizeandsample_size.The returned privacy budget will be incorrect.Is there a way to correct this? Or how can I circumvent this error?
Thanks in advance for your help and time.