Device Benchmark Worker
A worker script to benchmark the training performance of a deep neural network on CPU and GPU devices. This script is intended to be run as a subprocess by the main device selection module, which will provide the necessary configuration and data paths.
- seawrd._device_benchmark_worker.main()
The main function to run the benchmark worker. It reads the worker configuration from a JSON file specified in the command line arguments, loads the training and validation data, and runs the benchmark training for both CPU and GPU devices. The results are printed as a JSON object to stdout.