Machine Learning
in your browser!

MLJAR is a human-first platform for machine learning.
It provides a service for prototyping, development and deploying pattern recognition algorithms.
It makes algorithm search and tuning painless!


Some examples how it can be used:

Stock prediction

Predict stock market on AI tournament.

Credit scoring

Credit scoring with Machine Learning.

Employee analytics

Predict employee attrition with Machine Learning.

High energy physics

Analysis of data from high energy physics experiments.
Coming soon ...

How it works?

You need only 3 steps to build the most accurate Machine Learning models

Upload data

Upload your data with our secure connection.

Train & tune

Train and tune many Machine Learning algorithms and select best match to your data.

Predict & share

Use the best models for predictions share your results with others.

See MLJAR in action!

Beautiful python API!

This example code train and tune for you: Xgboost, LightGBM and Neural Networks models.

from mljar import Mljar


# create MLJAR project and experiment

model = Mljar(project = 'Project 1', experiment = 'Start!')


# train models

model.fit(X, y)


All in one JAR

MLJAR integrations:

Python
Nvidia Cuda
dmlc XGBoost
Keras
TensorFlow
scikit learn
DMTK
RGF

Features

one interface for many ML algorithms

built-in hyper-parameters search

train models in parallel

full reproducibility

beatiful interface

easily deploy models in the cloud or locally

MLJAR stats

So far we have:

Project

4,838

projects created

Experiment

5,668

experiments created

Dataset

12,772

datasets uploaded

Model

122,568

models trained

Sign up for newsletter

Enter your email and sign up for newsletter