fock 0.0.4

A continuous variable quantum neural network simulating framework written in PyTorch.

Fock

A continuous variable quantum neural network simulating framework written in PyTorch.

The name of the package comes from the fact that the Fock basis representation is used as the munderlining mathematics.

It is forked from strawberryfields==0.10.0.

Development Notes (Don't look)

torch.einsum('i,j->ij', x, y)

$|x\rangle \otimes |y \rangle = x_i y_j$

torch.einsum: Multiply each component of each operand then sum if possible.

Tensor product genralizes outer product.

Todo

test_circuit.py

How to debug

test env named torch

numpy==1.22.3

tensorflow-macos==2.8.0

torch==1.11.0

install fock from repo in editible mode

in the same directory as setup.cfg

pip install -e .

migrate

migrate strawberryfields==0.10.0 to tf2.0 then run comparison test to fock

test

write test logic

 cd tests/
 python3 test_ops.py

Push to Github

in the same directory as setup.cfg

git add .
git commit -m ''
git push

Publish to PyPI

change version number in setup.cfg

in the same directory as setup.cfg

python3 -m pip install --upgrade build

python3 -m build

python3 -m pip install --upgrade twine

python3 -m twine upload dist/*

References

https://packaging.python.org/en/latest/tutorials/packaging-projects/

https://stackoverflow.com/questions/5341006/where-should-i-put-tests-when-packaging-python-modules