Member-only story
Tech Trends
Best OpenSource AutoML frameworks in 2021
A curated list of OpenSource AutoML frameworks.
What is the AutoML framework?
AutoML is machine learning that has been automated to some extent, reducing the effort required from human coders. AutoML may include the following:
- Automated data preparation (e.g., imputation of missing values, feature scaling, feature selection, etc.)
- Automated grid search and hyperparameter optimization
- Automatic evaluation of multiple algorithms
- Automated ensembling of models (e.g., ensemble selection and stacking)
Ludwig is a toolbox that permits users to train and test deep learning models without writing code, built on top of TensorFlow.
To train a model, we need to provide is a file containing data, a list of columns as an input, and a list of columns as an output; Ludwig will perform the rest for us.
The Linux Foundation hosts Ludwig as part of the LF AI & Data Foundation.
MLBox is an influential Automated Machine Learning python library.
It delivers the below capabilities:
- Fast scanning and distributed data preprocessing, cleaning and formatting.
- Exceedingly robust feature selection and leak detection.
- Precise hyper-parameter optimization in the high-dimensional space.
- Futuristic predictive models for classification and regression (Deep Learning, Stacking, LightGBM).
- Prediction with models interpretation.
TPOT means Tree-based Pipeline Optimization Tool, a Python-based Automated Machine Learning tool that optimizes machine learning pipelines using genetic programming. TPOT is an Assistant to a Data Scientist.
- TPOT will automate the most tedious machine learning by intelligently exploring thousands of possible pipelines to find the best one for your data.
- TPOT is built beside the scikit-learn framework, so all of the code it generates should look familiar if you’re familiar with scikit-learn.
