From e9e89c399450c527f436f716a16363b493e22f7d Mon Sep 17 00:00:00 2001 From: George S <113141689+gs-olive@users.noreply.github.com> Date: Thu, 9 Nov 2023 13:18:22 -0800 Subject: [PATCH] minor fix: Update Benchmark values (#2453) --- tools/perf/benchmark.sh | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/tools/perf/benchmark.sh b/tools/perf/benchmark.sh index 319c779bf5..fd2b8732e2 100644 --- a/tools/perf/benchmark.sh +++ b/tools/perf/benchmark.sh @@ -92,7 +92,7 @@ echo "Benchmarking SD UNet model" for bs in ${large_model_batch_sizes[@]} do python perf_run.py --model_torch sd_unet \ - --precision fp32,fp16 --inputs="(${bs}, 4, 128, 128)@fp16;(${bs})@fp16;(${bs}, 1, 768)@fp16" \ + --precision fp32,fp16 --inputs="(${bs}, 4, 64, 64)@fp16;(${bs})@fp16;(${bs}, 1, 768)@fp16" \ --batch_size ${bs} \ --backends torch,dynamo,torch_compile,inductor \ --truncate \