using mechanisms individually

[canoztas]

i have checked the documentation and could not find how to use mechanisms on a dataset without ML. is there such a feature? for instance, I just want to add gaussian mechanism to my dataset then check the rows.

[naoise-h]

You can use the mechanisms in isolation by initialising them and then using the randomise() method to randomise the input. If you want to run a mechanism on an input vector or array, you can use Numpy's vectorize function as follows:

import diffprivlib as dp
import numpy as np

mech = dp.mechanisms.Gaussian(epsilon=0.5, delta=1e-5, sensitivity=1)
vfunc = np.vectorize(mech.randomise)

and then run it on your array, vfunc(array).

[canoztas]

thank you, besides that, how can i add data like noise to my original data ?