Choco

Installation

You can install the virtual environment in two version, The CPU version is for simulation quantum circuit with CPU and the GPU version simulate quantum circuit by GPU.

CPU version

1. Install the envirnoment by conda

conda env create -f environment_cpu.yml

2. activate the installed virtual environment.

conda activate choco_cpu

If the installation is interrupted or fails, you need to delete the environment and reinstall:

conda remove -n choco_cpu --all

3. install chocoq

pip install .

GPU version （need the support of CUDA12）

1. Install the envirnoment by conda

conda env create -f environment_gpu.yml

2. activate the installed virtual environment.

conda activate choco_gpu

If the nextwork is poor, the installation time for GPU will be longer, please be patient. When the installation is interrupted or fails, you need to delete the environment and reinstall:

conda remove -n choco_gpu --all

3. install chocoq

pip install .

Test for installation

Run corresponding test files according to the installation version:

python testbed_cpu.py

for CPU version or

python testbed_gpu.py

for GPU version.

If you see "Environment configuration is successful!" it means the installation is successful.

If the installation fails, you may consider:

1. Ensure that the correct conda environment is activated in the terminal.
2. After switching environments, execute pip install . under Choco-Q/ to install the chocoq package.
3. Make sure the Python execution environment is set to the corresponding conda environment. you may disable the user site by

 export PYTHONNOUSERSITE=1

Reproduce the experiments by Notebooks

1. implementations/0_test.ipynb - Custom optimization problem and solve it.

2. implementations/1_table.ipynb - Table 1 in the paper.