Compile Neural Networks developed in common development frameworks, such as TensorFlow, Caffe or Keras, for implementation onto Lattice CNN and compact CNN Accelerator IP cores.
Capabilities of NN Compiler
Analyze networks for fit in the chosen number of engines and allocated memory. After compilation, simulate networks for functionality and performance prior to testing in hardware. Graphical display of networks supports analysis and understanding.
*By clicking on the "Notify Me of Changes" button, you agree to receive notifications on changes to the document(s) you selected.