A high-throughput and memory-efficient inference and serving engine for LLMs
Mirrored from https://github.com/tduncan-lum/Nvidia-Optimization-Stack---VLLM.git
Synced Nov 17, 04:15 UTC
[FIX] Fix beam search test (#2930)
Zhuohan Li committed
Feb 20, 2024 at 14:37 UTC
63e2a6419dc5863311a11d1d2a95cda9fc8ef7e5
1 file changed
+1
tests/samplers/test_beam_search.py
+1
@@ -26,6 +26,7 @@ def test_beam_search_single_input(
26
max_tokens: int,
27
beam_width: int,
28
) -> None:
29
+ example_prompts = example_prompts[:1]
30
hf_model = hf_runner(model, dtype=dtype)
31
hf_outputs = hf_model.generate_beam_search(example_prompts, beam_width,
32
max_tokens)