Microsoft AI Open-Sources ‘SynapseML’ For Developing Scalable Machine Learning Pipelines

Microsoft has announced the release of SynapseML, an open-source library that simplifies and speeds up the creation of machine learning (ML) pipelines. SynapseML can be used for building scalable and intelligent systems to solve various types of challenges, including anomaly detection, computer vision, deep learning, form and face recognition, Gradient boosting, microservice orchestration, model interpretability, reinforcement learning, and personalization, search and retrieval, speech processing, text analytics, and translation.

SynapseML is a powerful platform for building production-ready distributed machine learning pipelines. It bridges the gap between several existing ML frameworks and Microsoft algorithms in order to create one scalable API that works across Python, R Language-based platforms like Scala or Java.

In order to build a machine learning pipeline, you need more than just coding skills. In fact, many developers find composing tools from different ecosystems requires considerable code, and frameworks aren’t designed for the task at hand-building servers in this case.

The growing pressure on data science teams to get more machine learning models into use and the fact that many companies still find themselves deploying AI within a reasonable amount of time despite these rising trends should not go unheeded.

SynapseML eliminates the hassle of working with multiple different ML learning frameworks by providing a single API that is scalable, data-agnostic, and language-neutral. It’s designed to help developers focus on high-level structures in their datasets instead of having them get bogged down trying to implement all these individual networks one at a time for every possible task or application type imaginable.

With the integration of a unified API, many tools are standardized, including frameworks and algorithms. This streamlines distributed machine learning experience for all users. It enables developers to write ML frameworks for use cases that require more than one framework. These features make it easy and quick, which is perfect for quick web supervised learning or search engine creation. It can also train models on a single node and scalable cluster of computers without wasting resources-this will help them quickly scale up their work with minimal overhead costs. The API can be used in a variety of programming languages to abstract over database access, file systems, and cloud data stores.

Github: https://github.com/microsoft/SynapseML

Tutorial: https://docs.microsoft.com/en-us/azure/synapse-analytics/machine-learning/tutorial-build-applications-use-mmlspark

Related Paper: https://arxiv.org/pdf/2009.08044.pdf

Reference: https://www.microsoft.com/en-us/research/blog/synapseml-a-simple-multilingual-and-massively-parallel-machine-learning-library/

Platform: https://microsoft.github.io/SynapseML/

Suggested