ML Workbench

Novopay’s Machine Learning(ML) Workbench is a data science tool for the non-data-scientist. A credit officer who wants to create a new credit model based on previous loan book(s), can use the ML Workbench to create a new predictive model and deploy it onto the cloud. Novopay’s ML Workbench guides you through the process of creating an accurate predictive model including - Data Ingestion, Exploratory Data Analysis, Model Building, Model Validation, Model Deployment, Model Monitoring and Model Prediction.


Download the Digital Transformation Solutions brochure for Banks, NBFCs and Fintechs.

Binoculars icon
Exploratory Data Analysis (EDA)
Choose from a library of pre-built exploratory data analysis algorithms and find patterns in your data using visual methods.Validate hypotheses, spot outliers and get insight into the complexity of subsequent Machine Learning algorithms required to achieve your desired outcome.Exploratory data analysis a precursor to building Machine Learning models and when done right can bring about improvements in Machine Learning models by an order of magnitude
Binoculars icon
Feature Engineering
One of the key reasons why AI lends itself to the future of Banking is the ability to process a large number of parameters about the customer in order to generate a more accurate credit decision. When there are lots of parameters or features about the customer in your training data, one needs to figure out which features are important to making say the credit decision. This process is called feature engineering. Novopay’s ML Workbench has tools that help in undertaking feature engineering.
Binoculars icon
Model Building
Choose from our pre-built library of supervised or unsupervised Machine Learning models to meet your regression, classification or clustering requirements. Set up hyper parameters for your algorithms and train your models at a click of a button without having to write a single line of code.Our library of supervised and unsupervised models includes algorithms ranging from logistic regression to deep neural networks to non linear dimensionality reduction techniques.
Binoculars icon
Model Validation
View accuracy measures for your classification and regression models on an intuitive dashboard to validate that the Machine Learning models have indeed “learnt” a meaningful pattern and have the ability to produce reliable classification or regression results.In addition to the standard accuracy metrics for classification and regression, the NovoAI Machine Learning Workbench also provides a visual way of inspecting model accuracy and performance.
Binoculars icon
One Click Model Deployment
Deploy your Machine Learning models with a click to your production system.Free yourself from building prototypes, converting code from Data Science specific languages into Java code or the like to deploy to your production system. With the click of a button your Machine Learning model is deployed to production and exposed as a REST API as well.The NovoAI Machine Learning Workbench also provides the ability to combine multiple Machine Learning models into a single Machine Learning model.Model output can be viewed through either the NovoAI Machine Learning workbench or by accessing the Machine Learning model API.
Binoculars icon
Model Prediction APIs
The NovoAI Machine Learning workbench enables users of the platform to create Machine Learning workflows either through a user interface or through an application programming interface. All actions, metrics and model scoring interfaces of the NovoAI Machine Learning Workbench are exposed as API endpoints enabling developers as well as data scientists to use the NovoAI Machine Learning Workbench in a flexible way.