CircuitSimulator

A quantum circuit simulator based on http://arxiv.org/abs/1601.07601. Features:

• A simple programming language for quantum circuits in the Clifford+T gate set
• An algorithm for sampling from the output distribution of a circuit
• An algorithm for calculating the probability of a particular output for a circuit

This application is written in python, a cross-platform programming language. A C implementation also is available for the slow part of both algorithms. The C implementation is used by default, but a python-only version is also available.

See the documentation for a complete description, once it is available.

Installation

1. Install python

Python 3 is recommended, although the code should still work in python 2. Download from https://www.python.org/downloads/ or use your favorite package manager.

2. Install numpy

Download numpy from https://pypi.python.org/pypi/numpyor use pip. Scipy is not required.

3. (Optional) Install matplotlib

The circuits/hiddenshift.py test script generates a plot similar to the one in http://arxiv.org/abs/1601.07601 using matplotlib. You can disable this but setting plot = False on line 30, or you can download and install matplotlib from http://matplotlib.org/users/installing.html.

4. Download the code

Press the clone or download button above and press "Download ZIP" or open up a terminal and type

$ git clone [email protected]:patrickrall/CircuitSimulator.git

5. Compile the code C implementation

On Mac OS X and Linux simply:

1. Open a terminal
2. Change into the root directory: $ cd CircuitSimulator
3. Call make: $ make

On Windows, a precompiled executable is available at libcirc/sample.exe. Tell the main.py executable to use it by passing cpath=libcirc/sample.exe as an option. If that does not work, try the -py option to use the python back-end.

Usage

There are two ways to use CircuitSimulator:

1. Use the main.py handle to compile a circuit and run an algorithm.

Examples:

python main.py circuits/toffoli.circ MMM
python main.py circuits/HTstack.circ 0

A full usage statement is available:

python main.py -h

Or just read the documentation.

2. Import the libcirc/probability.py file and use the probability() and sampleQubits() functions.

This is useful for automatically generated circuits that need to be evaluated. The circuits/hiddenshift.py file does exactly this: it randomly generates a circuit implementing the hidden shift algorithm and evaluates it immediately. Read circuits/hiddenshift.py for guidance when writing your own code and/or consult the documentation.