AQMLator

A package for auto (quantum machine learning)-izing your experiments!

Requirements

Python version 3.11 is required. Necessary packages are provided in the respective requirements*.txt files.

Database

The Postgres database is used to store the Optuna trials data. It has to be installed prior to running the package, so that the psycopg2 package can be installed properly.

Documentation

The documentation is built using Sphinx. Building PDF version of the documentation requires latex distribution (we used miktex) and perl (we used strawberry perl). To generate the pdf version of documentation, run

make latexpdf

and to generate the html version of documentation, run

make html

Both command should be run from the docs directory.

Installation

The package is available on pip, and can be installed using

pip install aqmlator

To install the package from the sources, run

pip install .

To develop the package, run

pip install -e .

To install the packages required for development, run

pip install -r requirements -r requirements-dev.txt

Setup

To fully set up the package, one has to add the database url to the aqmlator_database_url environment variable. The example of the database url is

postgresql://user:password@localhost/mydb

where user is the database user, password is the database password, localhost is the database host, and mydb is the database name.

Tests

We use tox to run full package tests. To do so, simply call

tox

in the project folder.

Access

To access the Optuna trials data use optuna-dashboard application. By default, it can be run using the following command

optuna-dashboard postgresql://user:password@localhost/mydb

while the (PostgreSQL) database is running.

Alternatively, one can use aqmlator.database.export_data_to_sqlite_database to export the data to the SQLite database, and handle it as one pleases.