mljar

Automated Machine Learning
Perform Mode

The perfect mode for production-ready ML pipelines

Perform Mode

The mljar AutoML framework can work in the Perform mode.
The Perform mode is perfect to get the ML pipeline ready for production.


Compete Mode

Cross-Validation

The Perform mode is using 5-fold Cross-Validation. It provides good balance between speed of training and accuracy of predictions.

Model Evaluation

ML Pipeline

The Perform mode was designed to create production-ready ML pipeline. The models are tuned but not as heavy as in Compete mode.

Algorithm Selection

Save time

With Perform mode you will save your time and deliver ready ML solution on schedule.

Machine Learning Algorithms

The Perform mode uses the following Machine Learning algorithms:
Linear Model, Random Forest, Xgboost, LightGBM, CatBoost, Neural Network, Ensemble.

Linear Model Random Forest Xgboost LightGBM CatBoost Neural Network Ensemble

The AutoML Code

Just set mode="Perform" during AutoML initialization and you are set! 

"""AutoML Perform code """

from supervised import AutoML

# Initialize AutoML in Perform Mode
automl = AutoML(mode="Perform")
automl.fit(X, y)

Check more mljar features

Golden Features

K-Means Features

Model Ensembling

ML Explainability

Automatic Documentation

AutoML Github Code