privacy

privacy copied to clipboard

ValueError: Dimension size must be evenly divisible by 10 but is 1 for '{{node training/Reshape}} = Reshape[T=DT_FLOAT, Tshape=DT_INT32](loss/mul, training/Reshape/shape)' with input shapes: [], [2] and with input tensors computed as partial shapes: input[1] = [10,?]

4

Hello, I am trying to train with differential privacy using tensorflow_privacy.DPKerasSGDOptimizer() here is my code E = 1 B = 100 N = 2 dataset = pickle.load( open(path, "rb" )...

jaalmoes

Combining the guide and tutorial sections into one for RAI docs.

copybara-service[bot]

RANCE-sama

Bolton DP is not running on Tensorflow 1.14.0

1

Runing this : https://github.com/tensorflow/privacy/blob/master/tutorials/bolton_tutorial.py Traceback (most recent call last): File "/Users/shachatt1/Desktop/sharmi/books/My_book_responsible_ai/python_code/chapter4/privacy-master/tutorials/bolton_tutorial.py", line 18, in from tensorflow_privacy.privacy.bolt_on import losses # pylint: disable=wrong-import-position File "/opt/miniconda3/lib/python3.7/site-packages/tensorflow_privacy/__init__.py", line 30, in from tensorflow_privacy import v1...

sharmi1206

I use the tensorflow=1.11.0, what tensorflow-privacy version is adapt to the tensorflow?

chensquan

There are dpSGD, dpAdam, dpAdagrad but no dpRMSprop provided by the code. Could someone please offer some intuition into this?

allhailjustice

the sensitivity of DP-SGD

1

I read the pepar and the code, I find the default cropping norm is 1 where sensitivity is 1 , and if I crop norm is not 1 ,how can...

JeffffffFu

Hello community! PATE's data-dependent analysis is based on teacher predictions. Is it possible to extend PATE to multi-label tasks such as part-of-speech tagging or named entity recognition, where the length...

donarni

DP accountant and composition

1

Hi, I read some of paper (both DP and RDP) such as [1,2], they mentioned that the composition of epsilon is _O_(\sqrt(T)) when all the iterations have the same epsilon...

XRC99

An op outside of the function building code is being passed a "Graph" tensor

2

Hello, I'm trying to make a differentially private GAN, however when computing the gradients for my discriminator I get the error message ``` TypeError: An op outside of the function...

samwincott