@@ -21,11 +20,11 @@ The ${BENCH} name should match the directory name of the respective benchmark un
The benchmarks are built under: `/hpvm/hpvm/build/tools/hpvm/test/dnn_benchmarks/`
Before running benchmarks set up CUDA/CuDNN paths using `hpvm/hpvm/setup_paths.sh`
Before running benchmarks set up CUDA/CuDNN paths using `hpvm/hpvm/set_paths.sh`
Modify the setup script to point to your local installations.
The accuracy of the benchmark is dumped to a file named `final_accuracy`
The accuracy of the benchmark (averaged across batches) is dumped to a file named `final_accuracy`
Ensure that this matches the accuracy of corresponding benchmarks here: https://gitlab.engr.illinois.edu/llvm/hpvm/-/tree/approx_hpvm_reorg/hpvm/projects/keras
