Catapult AI NN provides automation of Python-to-RTL for neural network (NN) hardware designs. The integration of hls4ml for generating synthesizable C++ Neural Network models from Python, along with enhanced Catapult features, produces high-performance ML inference hardware accelerators at the edge. Now Convolutional Neural Network designers and implementers can rapidly explore latency, throughput, and frequency performance, power and area with practically any CNN architecture and quantization by automatically generating the RTL implementation from common AI/ML frameworks.
Read the Press Release.
