We read every piece of feedback, and take your input very seriously.
To see all available qualifiers, see our documentation.
To run some micro-benchmarks using synthetic data to measure the performance of ROCm PyTorch, use the below repo: https://github.com/ROCmSoftwarePlatform/pytorch-micro-benchmarking