Benchmarking: Deep Learning

The QuickStart guide is 9 steps (out of which only 7 are focused on training) and the details can be found here: https://ngc.nvidia.com/catalog/model-scripts/nvidia:resnet_50_v1_5_for_mxnet/quickStartGuide

Clone the repository.

git clone https://github.com/NVIDIA/DeepLearningExamples
cd DeepLearningExamples/MxNet/Classification/RN50v1.5

Build the ResNet50 MXNet NGC container.

After Docker is setup, you can build the ResNet50 image with:

docker build . -t nvidia_rn50_mx

Start an interactive session in the NGC container to run preprocessing/training/inference.

nvidia-docker run --rm -it --ipc=host <path to dataset>:/data/imagenet/train-val-recordio-passthrough nvidia_rn50_mx

Download and preprocess the data.

• Download the images from http://image-net.org/download-images.

• Extract the training and validation data:

	  mkdir train && mv ILSVRC2012_img_train.tar train/ && cd train
	  tar -xvf ILSVRC2012_img_train.tar && rm -f ILSVRC2012_img_train.tar
	  find . -name "*.tar" | while read NAME ; do mkdir -p "${NAME%.tar}"; tar -xvf "${NAME}" -C "${NAME%.tar}"; rm -f "${NAME}"; done
  cd ..

Extract the validation data and move the images to subfolders.

mkdir val && mv ILSVRC2012_img_val.tar val/ && cd val && tar -xvf ILSVRC2012_img_val.tar
wget -qO- https://raw.githubusercontent.com/soumith/imagenetloader.torch/master/valprep.sh | bash

Preprocess the dataset.

./scripts/prepare_imagenet.sh <path to raw imagenet> <path where processed dataset will be created>

Start training.

./runner -n <number of gpus> -b <batch size per GPU (default 192)>