Benchmarker is a modular framework to automate a set of performance benchmarks, mostly for deep learning.
python3 -m benchmarker --mode=training --framework=pytorch --problem=resnet50 --problem_size=32 --batch_size=4
various devices, frameworks und underlying software stacks, network architectures etc.
Clone, install required packages for example by running
git clone --recursive https://github.com/undertherain/benchmarker.git
pip3 install [--user] -r requirements.txt
The original version was developed in 2017 by Aleksandr Drozd. Since then to the project contributed (in alphabetical order)
- Zhengyang Bai
- Kevin Brown
- Mateusz Bysiek
- Aleksandr Drozd
- Artur Podobas
- Shweta Salaria
- Emil Vatai