Microsoft has made SynapseML, a framework that allows you to create scalable machine learning pipelines, publicly available. In SynapseML, several existing ML platforms and new Microsoft algorithms are combined into a single API that can be used in Python, R, Scala and Java.
With SynapseML, developers can create scalable and intelligent systems to solve problems in the following areas:
- anomaly detection,
- computer vision,
- deep learning,
- pose and face recognition,
- gradient boosting,
- reinforcement learning,
- search and extract,
- speech processing,
- text analysis,
Many SynapseML tools do not require large marked-up datesets. Instead, SynapseML provides APIs for off-the-shelf services, such as Azure Cognitive Services, to quickly solve large-scale AI tasks.
SynapseML enables developers to implement more than 45 ML services directly into their systems and databases. For example, you can use already trained models to analyze documents, transcribe conversations in real time and translate text into more than 100 languages.
SynapseML allows you not only to use existing models and services, but also to create and train your own. The framework is available on Github.