This repo is no longer in use or maintained, the detection benchmark in MLCommons was updated to use RetinaNet instead of SSD: https://github.com/mlcommons/training
This is an experimental repository that is based on Nvidia's SSD-RN50 and is used to test and evaluate the model as the new object detection benchmark for MLPerf training and inference.
build and launch the container:
./scripts/docker/build.sh
./scripts/docker/launch_local.sh # you might want to change the dataset mount location
If necessary, Download MS-COCO dataset
./scripts/download_dataset.sh
To train the model, use any of the training scripts in scripts/train