diff --git a/README.md b/README.md index 11f15a39385059f2c2df5e00324c9bf0581a6306..cc51805bd7fbf8622e34ac258f1839d1f0781558 100644 --- a/README.md +++ b/README.md @@ -69,7 +69,8 @@ when batch_size is large (16, 32, 64, 128), throughput_amp > throughput_tf32. <img src="https://github.com/xuagu37/Benchmark_nnU-Net_for_PyTorch/blob/main/figures/benchmark_throughput_example.png" width="400"> -- Observation 2: The coefficient of variation of throughput for the 100 iterations is smallest when batch_size = 128. +- Observation 2: The coefficient of variation of throughput for the 100 iterations is smallest when batch_size = 128. +Benchmarking with batch_size = 128 can be used for node health check. For example, the expected throughput for dim = 2, nodes = 1, gpus = 4, batch_size = 128 would be 2500 ± 200. <img src="https://github.com/xuagu37/Benchmark_nnU-Net_for_PyTorch/blob/main/figures/benchmark_cv_example.png" width="400">