So far the best solution we’ve ever used for machine learning

What are the best aspects of this product?

With lots of tools it has to offer, it's possible to use this solution to develop an end-to-end machine learning platform. It has a plethora of machine learning algorithms like decision tree, logistic regression, support vector machines, linear discriminant analysis, and other clustering algorithms as well as boosting algorithms. There are also available pre-processing tools as well as hyperparameter tuning tools including GrindSearchCV and RandomSearchCV. It also offers different metric types to tune the model for precision, accuracy, etc. On top of that, this software works well with motplotlib, pandas, and other Python libraries.

What aspects are problematic or could work better?

It doesn't contain more advanced algoritms like LightGBM, XGBoost, and Catboost. Also, hyperparameter tuning is quite time consuming. Facilitating GPU may solve this problem.

What specific problems in your company were solved by this product?

Specifically, this solution is helping me clean data, test the baseline models, and tune and finalize them. It also gives me an opportunity to try various algorithms.

Are you a current user of this product?

Yes