Training sequential models with Q#, using multiple simulators in parallel

This sample uses Q# and the Microsoft.Quantum.MachineLearning library to train a simple sequential model. The model is trained on a half-moons dataset, loaded in C# using the System.Text.Json package, then preprocessed using Q#.

In this sample, the training loop is parallelized over model start points, with each model using its own instance of the full-state quantum simulator. Parallelizing in this way can lead to significantly improved performance, especially when using a large number of cores on a small number of qubits. The actual distribution of parallel tasks is performed using Parallel LINQ (PLINQ) from a C# host.

Prerequisites

Running the Sample

This sample can be run in a number of different ways, depending on your preferred environment.

C# in Visual Studio Code or the Command Line

At a terminal, run the following command:

dotnet run

C# in Visual Studio 2019

Open the folder containing this sample in Visual Studio ("Open a local folder" from the Getting Started screen or "File → Open → Folder..." from the menu bar) and set HalfMoons.csproj as the startup project. Press Start in Visual Studio to run the sample.

Manifest

  • Training.qs: Q# code implementing quantum operations for this sample.
  • Host.cs: C# code to load data, and to interact with and print out results of the Q# operations for this sample.
  • ParallelHalfMoons.csproj: Main C# project for the sample.
  • data.json: JSON-serialized training and validation data used by this sample.