The first ML framework for
feature learning
business data
multivariate time series
relational data
automated feature engineering

Build predictive analytics models on
relational data & time series blazingly fast.

Load Data


From Pandas, Dict or JSON

Database connectors

Unified import interface for PostgreSQL, MySQL, MariaDB, SQLite3, Greenplum or from any other ODBC compliant database

File storage

Import your data from CSV or AWS S3 buckets

Machine Learning

Data wrangling

Multirel & Relboost for feature learning from relational data and time series


Predict with XGB Regressor, XGB Classifier, logistics & linear or bring your own algorithm

Hyperparameter optimization

Tune hyperparameters on a latin hypercube or using a gaussian search

Evaluate & Deploy

Train pipelines

Wrap feature learner ensembles and predictors in end-to-end ML pipelines


Benchmark models & insights through features


Use python, or deploy models behind a HTTP model server to serve predictions or feature transforms

You stay in control

getML is for devs & data scientists. We don’t believe in black box magic. getML’s algorithms automate the tiresome part of your job to allow you to focus on bringing ML into your business.

Automate data wrangling

No more SQL! Stop writing hundreds of statements in SQL (or Python, or R) to extract features from your relational data - getMLs algorithms cover that. getML brings the concept of deep learning to relational data.

Focus on statistics

Data scientists are experts on statistics and machine learning. To build accurate ML models, they need to acquire difficult-to-get domain knowledge for data wrangling. getML helps you to stay focused on machine learning.

Intuitive API

Enjoy an easy-to-use Python API that follows patterns familiar from popular open source ML frameworks like scikit-lear n or TensorFlow. After completing our getting started guide you will be ready to tackle your own use case.

Hyperparameter optimization

getML comes with state-of-the-art AutoML capabilities. For tweaking the hyperparameters of your trained models getML includes a latin hypercube search or Gaussian hyperparameter optimization.


You don’t need to worry about data privacy: getML is a stand-alone machine learning framework. getML runs locally or in your private cloud. We don’t get access to your private data, ever.