Generated CNN Class Hierarchy

The generated CNN code has the following class hierarchy. The Layer class and the generated Network class have three important methods:

1. setup(), which allocates memory and system resources for each layer.

2. predict(), which performs forward inference in the execution loop.

3. cleanup(), which releases all memory and system resources.

Related Topics

• Supported Networks, Layers, and Classes
• Load Pretrained Networks for Code Generation
• Code Generation for Deep Learning Networks by Using cuDNN
• Code Generation for Deep Learning Networks by Using TensorRT
• Code Generation for Deep Learning Networks Targeting ARM Mali GPUs