Simplifying Deep Learning (DL) and Neural Networks (NN) for Embedded Processing
Deep Learning (DL), a subset of Machine Learning, is quickly becoming a crucial technology within vehicles: from vision processing to automated driving —the DL market is expected to reach USD 18.16 Billion by 2023. DL offers better accuracy and maintainability in tasks such as object detection and classification over “traditional” computer vision algorithms, but the barriers to full implementation bring complexity and steep costs.
This webinar will show how to implement and configure the NXP eIQ™ Auto DL toolkit for optimizing and implementing DL without the need for customized hardware expertise. The eIQ Auto toolkit quantizes, prunes, and compresses Neural Networks (NN) by partitioning workload and selecting the optimum hardware to compute engines on the MPU.
Attendees will learn:
- How eIQ Auto DL toolkit simplifies DL and NN for Embedded Processing
- To connect to TensorFlow, Pytorch, ONNX, and Caffe leading training frameworks
- How the eIQ Auto DL toolkit partitions networks to improve performance and reduce DL complexity using on-chip computation engines like the Arm® Cortex® A, Arm Neon™, and APEX vision accelerators