Overview | Key features | Installation | Usage | Free software | Contribute | Help us

Overview

Multi-modal learning, where diverse data types are integrated and analyzed together, has emerged as a critical field in artificial intelligence. However, most algorithms assume fully observed data, an assumption that is often unrealistic in real-world scenarios. To address this gap, we have developed iMML, a Python package designed for multi-modal learning with incomplete data.

Overview of iMML for multi-modal learning with incomplete data.

Key features

The key features of this package are:

• Coverage: More than 25 methods for integrating, processing, and analyzing incomplete multi-modal datasets implemented as a single, user-friendly interface.
• Comprehensive: Designed to be compatible with widely-used machine learning and data analysis tools, allowing use with minimal programming effort.
• Extensible: A unified framework where researchers can contribute and integrate new approaches, serving as a community platform for hosting new methods.

Installation

Run the following command to install the most recent release of iMML using pip:

pip install imml

Or if you prefer uv, use:

uv pip install imml

Some features of iMML rely on optional dependencies. To enable these additional features, ensure you install the required packages as described in our documentation: https://imml.readthedocs.io/stable/main/installation.html.

Usage

For this example, we will generate a random multi-modal dataset, that we have called Xs:

import numpy as np
Xs = [np.random.random((10,5)) for i in range(3)] # or your multi-modal dataset

You can use any other complete or incomplete multi-modal dataset. Once you have your dataset ready, you can leverage the iMML library for a wide range of machine learning tasks, such as:

• Decompose a multi-modal dataset using MOFA to capture joint information.

from imml.decomposition import MOFA
transformed_Xs = MOFA().fit_transform(Xs)

• Cluster samples from a multi-modal dataset using NEMO to find hidden groups.

from imml.cluster import NEMO
labels = NEMO().fit_predict(Xs)

• Simulate incomplete multi-modal datasets for evaluation and testing purposes using Amputer.

from imml.ampute import Amputer
transformed_Xs = Amputer(p=0.8).fit_transform(Xs)

Free software

iMML is free software; you can redistribute it and/or modify it under the terms of the BSD 3-Clause License.

Contribute

Our vision is to establish iMML as a leading and reliable library for multi-modal learning across research and applied settings. Our priorities include to broaden algorithmic coverage, improve performance and scalability, strengthen interoperability, and grow a healthy contributor community. Therefore, we welcome practitioners, researchers, and the open-source community to contribute to the iMML project, and in doing so, helping us extend and refine the library for the community. Such a community-wide effort will make iMML more versatile, sustainable, powerful, and accessible to the machine learning community across many domains.

For the full contributing guide, please see:

• In-repo: https://github.com/ocbe-uio/imml/tree/main?tab=contributing-ov-file
• Documentation: https://imml.readthedocs.io/stable/development/contributing.html

Help us grow

How you can help iMML grow:

• 🔥 Try it out and share your feedback.
• 🤝 Contribute if you are interested in building with us.
• 🗣️ Share this project with colleagues who deal with multi-modal data.
• 🌟 And of course… give the repo a star to support the project!