CIFY is a framework for computational intelligence algorithms written in Python. The goal of the project is to create a framework that allows users to easily and reliably implement nature-inspired metaheuristics. The framework provides a set of very simple abstractions for implementing metaheuristics, objective functions, running experiments and collecting results.

The guiding principles of the project are:

Low barrier of entry. Reading the tutorial or looking at the examples is all you should need to start working with CIFY.
Reproducibility. Experiments with stochastic operations should be easily reproducible.
Speed. Computational time should be kept to a minimum.
Tests and documentation. All code should be thoroughly tested and documented.

Quickstart#

Install cify using pip

pip install cify

or clone the repo and install cify to your environment using

cd CIFY
pip install -e .

Using the following example to get started or refer to the documentation for more details.

import numpy as np

from cify import ObjectiveFunction, Optimization, Task, set_seed
from cify.ga import GA

# 1. Set internal seed for stochastic operations.
set_seed(0)

# 2. Define an objective function.
bounds = [0.0, 1.0]
dim = 10
sphere = ObjectiveFunction(lambda vector: np.sum(vector ** 2),
                           bounds,
                           dim,
                           Optimization.Min,
                           "sphere")

# 3. Create an algorithm.
ga = GA(30, sphere)

# 4. Define a performance metric
def metric(ga: GA, f: ObjectiveFunction) -> float:
    return sorted(ga.individuals)[0]

# 5. Create a task to run the algorithm.
task = Task(ga,
            sphere,
            max_iterations=1000,
            log_iterations=100,
            metrics=[("best_of_value", metric)])

# 6. Execute
task.run()
print(task.results["best_of_value"][-1])

Contributing#

CIFY is an open-source project, and it depends its users to continually improve. We are very pleased that you would like to help!

If you notice a bug, have an idea for a feature, or have a question about the code, please don’t hesitate to open an issue. Many features come about simply by the asking of a question, or requesting a feature directly.

If you are unsure as to where to start take a look at the existing issues or reach out directly. Improving documentation, adding examples and adding tests are all good (and highly appreciated) ways to start contributing to CIFY.

If your contribution has been derived from or inspired by other work, please state this in its docstring and provide proper attribution. When possible, include the original authors’ names and a link to the original work.

To setup your development environment for CIFY, please follow the commands below.:

git clone https://github.com/KyleErwin/cify
cd cify
python3 -m venv venv
source venv/bin/activate
pip install -r dev-requirements.txt
pre-commit install

To build the docs, please install pandoc.

Lastly, please run pytest to ensure all of your changes pass the tests.

Quickstart#

Contributing#

Contents#