Skip to content

Latest commit

 

History

History
15 lines (10 loc) · 611 Bytes

File metadata and controls

15 lines (10 loc) · 611 Bytes

#Benchmark scripts

These scripts will test the efficiency of SINGA by training benchmark models pecified in convnet-benchmarks over different devices (e.g., CPU and GPU).

To run them, create a python pip virtualenv or anaconda virtual environment as guided by this article. Then, execute the run.py as

$ python run.py

Different models and devices could be tested, please refer to the command line help message,

$ python run.py -h