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 16, 02:00 UTC
[DOC] Fix typo in docstring and assert message (#12194)
Signed-off-by: Yuan Tang <terrytangyuan@gmail.com>
Yuan Tang committed
Jan 20, 2025 at 01:58 UTC
c5c06209ec1d90146dd12095d7bff3326aa6dd15
1 file changed
+2
-2
vllm/engine/output_processor/single_step.py
+2
-2
@@ -102,9 +102,9 @@ class SingleStepOutputProcessor(SequenceGroupOutputProcessor):
102
103
Args:
104
seq_group: the output is associated with this :class:`SequenceGroup`
105
- output: the :class:`SequenceGroupOutput` for a single scheduler step
105
+ outputs: the :class:`SequenceGroupOutput` for a single scheduler step
106
"""
107
- assert len(outputs) == 1, ("Single step should only has 1 output.")
107
+ assert len(outputs) == 1, "Single step should only have 1 output."
108
output = outputs[0]
109
assert isinstance(output, CompletionSequenceGroupOutput)
110
single_step_process_prompt_logprob(self, seq_group, output)