search

JinseongP / DPTrainer

Official PyTorch implementation of "In-distribution Public Data Synthesis with Diffusion Models for Differentially Private Image Classification", CVPR 2024.
MIT License
3 stars 1 forks source link

Dataset Preparation for MNIST and Fashion-MNIST #1

Open M-Ghorbel opened 2 weeks ago

M-Ghorbel commented 2 weeks ago

Hi,

I would like to evaluate your method on the MNIST and Fashion-MNIST datasets. However, I couldn't locate the part of the code where the dataset is prepared according to your strategy, as you have directly provided the modified CIFAR-10 and CIFAR-100 datasets.

Could you please point me to the relevant section?

Thank you.

uzn36 commented 1 week ago

We've uploaded the data preparation code to examples/prepare_train_data_cifar10.ipynb. You need to zip the output directory and place it in the same directory as the train.py file of EDM.