I am using tasks on the NVDLA Virtual Platform (emulator for NVDLA) using the
nvdla_runtime. The nvdla runtime is invoked as :
./nvdla_runtime --loadable lenet_mnist.nvdla --image seven.pgm --rawdump --instance 0
This however runs only a single image. Is there a way to run a stream/batch of images with the NVDLA runtime? Invoking the runtime per image is prohibitively expensive, so I assume there must be a better way. Guidance is appreciated. Thanks!