[X] I have searched exisisting GitHub issues to make sure the issue does not already exist.
Expected behavior
I expect the application of Gaussian operations on state in the Fock representation to apply suitable cutoffs based on the AUTOSHAPE settings
Actual behavior
If I apply any Gaussian operation on a Fock state, the cutoff is expanded to the value given by AUTOSHAPE_MAX, regardless of the value of AUTOSHAPE_PROBABILITY
Reproduces how often
100 %
System information
Mr Mustard: a differentiable bridge between phase space and Fock space.
Copyright 2021 Xanadu Quantum Technologies Inc.
Python version: 3.10.14
Platform info: Windows-10-10.0.19045-SP0
Installation path:
c:\Users\jacob\miniconda3\envs\mrmustard_dev\lib\site-packages\mrmustard
Mr Mustard version: 0.7.3
Numpy version: 1.23.5
Numba version: 0.59.1
Scipy version: 1.14.1
The Walrus version: 0.21.0
TensorFlow version: 2.17.0
Source code
from mrmustard import settings
import mrmustard.lab_dev as mm
settings.AUTOSHAPE_PROBABILITY = 0.9
settings.AUTOSHAPE_MAX = 10
state = mm.Ket.from_fock([0], [0, 1])
state >> mm.Dgate([0])
Before posting a bug report
Expected behavior
I expect the application of Gaussian operations on state in the Fock representation to apply suitable cutoffs based on the AUTOSHAPE settings
Actual behavior
If I apply any Gaussian operation on a Fock state, the cutoff is expanded to the value given by AUTOSHAPE_MAX, regardless of the value of AUTOSHAPE_PROBABILITY
Reproduces how often
100 %
System information
Source code
Tracebacks
No response
Additional information
No response