13e8e75143
### What this PR does / why we need it? we notice that `patch_main` is never used. Usually the patch is for all version. And if it's for specified version, we can use `vllm_version_is` instead. So let's remove the useless sub folder in patch module to make it clear. - vLLM version: v0.11.0rc3 - vLLM main: https://github.com/vllm-project/vllm/commit/v0.11.0 Signed-off-by: wangxiyuan <wangxiyuan1007@gmail.com>
27 lines
1.4 KiB
Python
27 lines
1.4 KiB
Python
import torch
|
|
import vllm
|
|
from vllm._custom_ops import apply_repetition_penalties_torch
|
|
|
|
|
|
def apply_repetition_penalties(logits: torch.Tensor, prompt_mask: torch.Tensor,
|
|
output_mask: torch.Tensor,
|
|
repetition_penalties: torch.Tensor) -> None:
|
|
"""Apply repetition penalties to logits in-place.
|
|
|
|
Args:
|
|
logits: The logits tensor of shape [num_seqs, vocab_size].
|
|
prompt_mask: A boolean tensor indicating which tokens appear in the prompt.
|
|
output_mask: A boolean tensor indicating which tokens appear in the output.
|
|
repetition_penalties: The repetition penalties of shape (num_seqs, ).
|
|
"""
|
|
apply_repetition_penalties_torch(logits, prompt_mask, output_mask,
|
|
repetition_penalties)
|
|
|
|
|
|
# NPU device type tensors have attributes is_cuda=True and is_npu=True, according to its implementation in
|
|
# https://github.com/Ascend/pytorch/blob/863b9071cbdf47023c12c246e3efa9c6e2285fc6/torch_npu/npu/_stream_check.py#L74
|
|
# This causes that vLLM's apply_repetition_penalties function will run into the branch of "if logits.is_cuda" and
|
|
# call the custom op implemented in CUDA, which is not compatible with NPU.
|
|
# Reference: https://github.com/vllm-project/vllm/blob/f66673a39d9f364194c249f28098cad8a5584ccb/vllm/_custom_ops.py#L314
|
|
vllm._custom_ops.apply_repetition_penalties = apply_repetition_penalties
|