An example of using Nvidia's TensorRT library for popular Deep Learning architectures (in progress)
The CNN model implemented is adapted from this tensorflow tutorial. The source code from which it was adapted can be found here.
The model contains two convolutional layers coupled with max pooling and normalization. Following the convolutions are two fully connected layers, concluding with a softmax layer for classification.
Data used was the Cifar10 dataset. Details on how to download it and read it can be found in the previously mentioned tutorial.