From 11ad8fe4d3e2bc324dac1d42ed47d3fc78085025 Mon Sep 17 00:00:00 2001 From: yuanheng-zhao Date: Tue, 5 Sep 2023 17:43:25 +0800 Subject: [PATCH] trivial - uncomment for testing --- examples/inference/bench_bloom.py | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/examples/inference/bench_bloom.py b/examples/inference/bench_bloom.py index dbd60d103c34..ce4396b11ba5 100644 --- a/examples/inference/bench_bloom.py +++ b/examples/inference/bench_bloom.py @@ -81,7 +81,7 @@ def bench_bloom(test_config): outputs = infer_engine.generate(input_tokens, generate_kwargs) torch.cuda.synchronize() end = time.time() - # infer_engine.cache_manager.free_all() + infer_engine.cache_manager.free_all() out_len = outputs.shape[1] print(f" iter {i}: out len {str(out_len)}, generation time {str(end - start)} s") times.append((end - start) / (out_len - input_len))