ML Manager

Splice ML Manager is an integrated machine learning platform that minimizes data movement and enables enterprises to deliver better decisions faster by continuously training the models on the most updated available data.

With Splice ML Manager, data science teams are able to produce a higher number of more predictive models as they are empowered to:

• Experiment frequently using diverse parameters to compare model effectiveness
• Leverage updated operational data to concurrently train the model
• Minimize the movement of data by running the models in the database
• Compress the time from model deployment to action

End-to-End Lifecycle Management for Your ML Models

Splice ML Manager provides end-to-end lifecycle management for your ML models, thereby streamlining and accelerating the design and deployment of intelligent applications using real-time data.

With our ML Manager platform, based on MLFlow, we have enabled a closed-loop machine learning lifecycle. Our improved API makes it quicker and easier to manage your ML development, from bulk logging of model parameters and metrics to full visibility into pipeline stages and feature transformations. With just a few added lines of code, data engineers can recreate any ML pipeline in seconds. Direct access to the training and testing tables allows data scientists to guarantee new models are evaluated on the same data as currently deployed ones.

Traditional enterprise data infrastructure consisting of separate transactional, analytical and data science platforms does not provide a viable foundation to power mission-critical machine learning (ML) applications. This architecture has latency built-in at multiple levels.

First, the model development and training phase requires data to be continuously extracted from transactional enterprise applications. Once the model is built and trained, it requires expensive transformations and aggregations to operationalize the features before any predictions can be made. This infrastructure is also not agile enough to trigger the right action in real-time especially where data attributes change rapidly and require the model to be continuously trained on updated data.