modular-pipelines

Modular Pipelines

Fast composition of Machine Learning Pipelines
Explore the docs » . View Demo . Check Examples

---

About Modular Pipelines (mopi)

Modular Pipelines is a lightweight and extensible library to create complex multi-model and multi-modal pipelines.

You can easily build:

Ensemble Pipelines:

Ensemble Learning → Ensemble learning is a general meta approach to machine learning that seeks better predictive performance by combining the predictions from multiple models. See more on wikipedia Wiki - Ensemble Learning or A Gentle Introduction to Ensemble Learning Algorithms

Meta-Learning Pipelines:

Meta Learning → in machine learning refers to learning algorithms that learn from other learning algorithms. Most commonly, this means the use of machine learning algorithms that learn how to best combine the predictions from other machine learning algorithms in the field of ensemble learning. Meta Modelling and What Is Meta-Learning in Machine Learning?

Multi-objective Pipelines:

(back to top)

Compatible With

(back to top)

---

Installation

The project was entire built in python

Prerequisites

• conda, python >= 3.7 and pip
• pytorch and wandb

Install mopi (default)

1. Clone the project by running

   git clone https://github.com/applied-exploration/modular-pipelines.git
   

2. Navigate to the project root directory

3. Build the project by executing

   pip install -e .
   

Clone project

1. The repo comes with a conda env file, enter mamba env create to install all dependencies

2. We use isort + black for formatting

To be able to use Huggingface and Wandb:

3. Enter your Wandb and Huggingface API in and .env file

    HF_HUB_TOKEN = [ENTER YOUR API]
    WANDB_API_KEY = [ENTER YOUR API]

(back to top)

Contact

Project Link: https://github.com/applied-exploration/modular-pipelines

(back to top)