top of page

AutoML with Sparkflows 

AutoML in Sparkflows offers guided automation by providing a flexible and easy web interface. Using AutoML citizen Data Scientists can quickly build multiple models of different flavours and decide on the top model for production deployment. Data Scientists can use the tool to quickly check the outcome of their hypothesis and let them shift their focus towards more complex issues requiring human expertise and domain knowledge.

Empower participation of non-technical users in solving

data-oriented problems and make machine learning attainable

to your organization using the power of Sparkflows AutoML.

Democratize the Model Selection Process

Empower your organizations business analysts, product owners, citizen data scientists to easily build machine learning models. Use AutoML to implement your data science project with or without programming expertise, save time and resource by being able to quickly work in an agile problem-solving approach. Sparkflows AutoML automatically takes care of missing values, converts categorical data using encoding techniques, among other techniques and hence reduces the effort required by the team to start building the ML model.

Remove the manual coding machine learning process and take advantage of the Sparkflows AutoML engine to quickly scale up your ML Model building capacity.

Frame 22.png

Classification and Regression

Tackle your organizations classification or regression problems by training models with Sparkflows AutoML using the inbuilt open-source H2O and PyCaret engines. Sparkflows AutoML guides the end-user in every step of the AutoML selection process and provides a clean and simple-to-understand interface that can be used by non-experts too. Both H2O and PyCaret solves supervised and unsupervised machine learning problems and gives out one/multiple trained machine learning model.

With the support of over 25+ automatic data cleaning techniques, automatic hyperparameter tuning, automatic model selection, and experiment logging accelerate the time it takes to get

production-ready ML models with ease and efficiency.


Easily iterate through the AutoML experiments and view the leaderboard of current as well as past experiments. The leaderboard lists out the models with the metrics in a tabular format and one can choose the model of that fits the best. Within the AutoML experiments, select the best model based on performance metrics like R2, MAE, RMSE, MSE etc for regression and metrics like F1 , LOG_LOSS, precision, accuracy, ROC_AUC etc for classification.

The model performing best on the leaderboard might not be the best model for a particular problem/industry. For example, in a regulated industry, even though a Stacked ensemble model might be the best performer, it cannot be used and hence one can choose the best performing linear model from the leaderboard. All these options are available to the user when using AutoML in Sparkflows.

bottom of page