qsensoropt 1.1.0

https://arxiv.org/abs/2312.16985

qsensoropt

qsensoropt is a library based on Tensorflow 2 for the automatic optimization of adaptive and non-adaptive controls in quantum metrology tasks.

Docs

Documentation for this project is available on gitlab pages.

Along with the description of all the modules, the documentation contains many examples of use on which the user can base new applications.

Installation

The following instructions are written for a Linux system. It is advisable to install qsensoropt in a conda environment created for the purpose.

1. If conda is not available on you machine, install it with the following commands

   curl https://repo.anaconda.com/miniconda/Miniconda3-latest-Linux-x86_64.sh -o Miniconda3-latest-Linux-x86_64.sh
   bash Miniconda3-latest-Linux-x86_64.sh
   

2. Resource the bash configuration with

   source ~/.bashrc
   

3. Create a new conda environment named qsensoropt

   conda create --name qsensoropt python==3.10
   

4. Activate the environment just created and upgrade pip

   conda activate qsensoropt
   pip install --upgrade pip
   

5. qsensoropt is based on Machine Learning and a such it is best used with a GPU. If a GPU is available on your system you can use it by installing CUDA and cuDNN

   conda install -c conda-forge cudatoolkit=11.8
   pip install nvidia-cudnn-cu11==8.6.0.163
   conda install -c nvidia cuda-nvcc
   

   Configure the system path with

   mkdir -p $CONDA_PREFIX/etc/conda/activate.d
   echo 'CUDNN_PATH=$(dirname $(python -c "import nvidia.cudnn;print(nvidia.cudnn.__file__)"))' >> $CONDA_PREFIX/etc/conda/activate.d/env_vars.sh
   echo 'export LD_LIBRARY_PATH=$CONDA_PREFIX/lib/:$CONDA_PREFIX/bin:$CUDNN_PATH/lib:$LD_LIBRARY_PATH' >> $CONDA_PREFIX/etc/conda/activate.d/env_vars.sh
   

   followed by

   echo 'export XLA_FLAGS=--xla_gpu_cuda_data_dir=/path/to/cuda/' >> $CONDA_PREFIX/etc/conda/activate.d/env_vars.sh
   

   where in the place of /path/to/cuda/ the complete path to cuda on the machine must be inserted. In most cases, while using conda it is sufficient to specify the directory of the environment qsensoropt.

   Then restart the environment with

   conda deactivate
   conda activate qsensoropt
   

6. Clone the repository and enter in the qsensoropt directory

   git clone https://gitlab.com/federico.belliardo/qsensoropt.git
   cd qsensoropt
   

7. Install the qsensoropt library and all its dependencies

   pip install -e .
   

Congratulation! You can now optimize your quantum sensor!