3.4 Benchmarks

• Deep Learning Benchmark Results: https://dawn.cs.stanford.edu//2018/04/30/dawnbench-v1-results/

• MLPerf (https://github.com/mlcommons/training, https://github.com/mlcommons/inference)

• AI-benchmark (https://pypi.org/project/ai-benchmark/) - TF-only, mainly directed towards mobile devices, but many models (pretty old, Dec, 2019)

• MLBench (https://github.com/mlbench/mlbench-core) - main focus on distributed (multi-node) algorithms

• HP DLBS (https://github.com/HewlettPackard/dlcookbook-dlbs) - TF, PyTorch, NVIDIA’s TensorRT (Jul, 2020)