quingo 0.1.3

Quingo Runtime System

Quingo Runtime System

Along with quingo compilers, the Quingo runtime system which provides users the capability to program and simulate Quingo programs.

Installation

Overview

The Quingo installation comprises of two main steps:

Runtime system and simulator

Install Quingo runtime system with required simulators using the following command:

pip install quingo

Upon success, it will automatically install the Quingo runtime system (this package), the PyQCAS simulator and the PyQCISim simulator.

The Quingo compiler

Since the Quingo runtime is a framework integrating and managing quantum and classical computational resources, it does not contain the quantum compiler by default. The Quingo compiler should be downloaded separately.

Two versions of Quingo compiler has been developed:

1. the xtext-based compiler, which appears as a java executable, and
2. the mlir-based compiler, which presents as a binary executable.

The xtext compiler can generate eQASM instructions which can be simulated by PyQCAS and the mlir compiler can generate QCIS instructions which can be simulated by PyQCISim.

After downloading the binary, you need to call specify the compiler path for once in python using the following command:

import quingo
# for xtext compiler
quingo.quingo_interface.set_xtext_compiler_path(<path-to-quingo.jar>)
# or for mlir compiler
quingo.quingo_interface.set_mlir_compiler_path(<path-to-quingoc>)

For the mlir compiler, you could also put it in a directory so that your terminal can find it, like /usr/loca/bin/. In this case, you no longer need to call set_mlir_compiler_path to specify its path.

Special Care for Linux

Since the compiler executable quingoc depends on a number of libraries under linux, quingoc may not work well if you only download this executable binary. To cease the difficulty in running Quingo programs under linux, we have prepared a docker image (around 400MB) which prepares everything in ready. You can install it using the following command:

docker pull xsu1989/quingo:beta
docker run -it xsu1989/quingo:beta
cd examples && python3 host.py

In this case, you do not need to install anything else to run Quingo programs.

Usage

A simple example can be found in the directory src/examples. You can simply run the bell_state example by running:

cd src/examples/bell_state
python host.py

If everything runs correctly, you should see the following output:

connecting pyqcisim_quantumsim...
num_qubits:  2
The result of bell_state is:
(['q0', 'q1'], {'00': 504, '01': 0, '10': 0, '11': 496})

APIs of the Quingo runtime system

The Quingo_interface class expose the following methods:

• set_log_level(<log_level>): <log_level> can be one of DEBUG, INFO, WARNING, ERROR, or CRITICAL.
• connect_backend(<backend>): <backend> currently can be 'pyqcas_quantumsim' or 'pyqcisim_quantumsim'.
• get_backend_name(): return the name of the backend that is being used. An empty string will be returned if no backend has been set.
• set_compiler(<compiler_name>): <compiler_name> can be 'mlir' or 'xtext'.
• get_last_qasm(): get the qasm code generated by the last execution.
• config_execution(<mode>, <num_shots>):
  • Configure the execution mode to 'one_shot' or 'state_vector'.
  • When the execution mode is 'one_shot', the number of times to run the uploaded quantum circuit can be configured using the parameter num_shots at the same time.
• call_quingo(<qg_filename>, <qg_func_name>, *args):
  • the main entry to call Quingo operation.
  • <qg_filename (str)> : the name of the Qingo file which contains the quantum function called by the host program.
  • <qg_func_name (str)> : the name of the quantum function
  • <args (dict)>: a variable length of parameters used to call the Quingo operation in the form qg_func_name(<args>).
• read_result(): read the computation result from the quantum kernel.
  • For eQASM-based backend, the result is a binary block which encodes the quantum computation result.
  • For QCIS-based backend, the result format is defined by PyQCISim. Please refer to the docstring of quingo.if_backend.non_arch_backend.pyqcisim_quantumsim.PyQCISim_quantumsim::execute()

Quingo programming tutorial

Please refer to the Quingo User Document.