The MapR Distributed Deep Learning QSS leverages expertise from implementing advanced machine learning environments for MapR customers. The new offering features access to distributed deep learning libraries (e.g. TensorFlow, Caffe, mxnet, etc.), a framework that intelligently switches storage and workflow between CPU and GPUs, and the stability, scale and performance of the Converged Data Platform to form the basis for advanced, data-driven applications. Use cases for distributed deep learning technologies include:
- Extracting insights from images/video: Improve business outcomes from processing and analyzing images and video, such as ultrasounds, dashboard cameras, drone computer vision, satellite images, surveillance footage, etc.
- Understanding and predicting sequence of events: Predicting behaviors or understanding patterns based on analysis of sequenced audio files, language models (natural language processing), written texts/social media posts, and analysis of time series data will allow businesses to stay one step ahead of expected outcomes. Businesses can use this knowledge to better predict how customers will react to a certain event, offer customers more relevant recommendations, streamline operations etc.
- Classification and forecasting: Increase accuracy and reliability of existing statistical and predictive models with more complex algorithms and advanced modeling. Applying deep neural networks provides more reliable forecasts, and delivers better analysis from unsupervised learning
Deep learning algorithms can provide profound transformational opportunities for an organization,” said Anil Gadre, chief product officer, MapR Technologies. “Our expertise in advance machine learning deployments coupled with the unique design of the MapR Platform form the foundation for our new offering. The QSS will enable companies to quickly take advantage of modern GPU-based architectures and set them on the right path for scaling their deep learning efforts.”