Uncover every layer of understanding Deep learning from SAS

Uncover every layer of understanding

With deep learning from SAS

From algorithmic improvements to new classes of neural networks to higher computational power – deep learning has advanced by leaps and bounds. The dynamic nature of deep learning methods – their ability to continuously improve and adapt to changes – presents countless new opportunities.

By combining the power of the SAS^® with open source technologies, you can take deep learning to the next level.

Deep Learning + Python = DLPy

Learn how using Python and SAS^® Viya^® has changed the way you utilize AI solutions like NLP and computer vision, opening the door to new deployment strategies, more accurate models and better results.

Read the SAS deep learning with Python made easy blog

Read blog

Deep learning demos

Image Classification With Convolutional Neural Networks

Convolutional neural networks extract features from images and are often more efficient than traditional neural networks.

Import and Export Deep Learning Models With ONNX

With ONNX, models can be deployed wherever work needs to be done – including iPhones and Android phones.

Object Detection With Tiny YOLOv2 Model

With “You Only Look Once” (YOLO) algorithms, systems identify common objects that can be recognized in a single glance.

Text Classification and Text Generation With Recurrent Neural Networks

RNN models can identify and “remember” important words in a sentence easily, such as entities, verbs and adjectives.

Time Series Forecasting With Recurrent Neural Networks

By applying neural networks, you can improve speed and accuracy of forecasts versus using traditional time series models.

Deep learning in action

Fun fact: SAS HQ has two on-site solar farms. Check out this blog and the below demo to see how we’ve built a recurrent neural network to forecast solar energy output.