@tduncan / Nvidia-Optimization-Stack-VLLM Mirror · read-only / commits / d05f0a9db2
Fork 0

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, 17:00 UTC

[Bugfix] Fix import error in Phi-3.5-MoE (#8052)

Cyrus Leung committed Aug 31, 2024 at 13:26 UTC d05f0a9db2c32528f4aff7e741ff6caf21dd0802
1 file changed +2 -2
vllm/model_executor/models/phimoe.py
+2 -2
@@ -38,13 +38,13 @@ from vllm.model_executor.layers.logits_processor import LogitsProcessor
38 from vllm.model_executor.layers.quantization.base_config import (
39 QuantizationConfig)
40 from vllm.model_executor.layers.rotary_embedding import get_rope
41 -from vllm.model_executor.layers.sampler import Sampler
41 +from vllm.model_executor.layers.sampler import Sampler, SamplerOutput
42 from vllm.model_executor.layers.vocab_parallel_embedding import (
43 DEFAULT_VOCAB_PADDING_SIZE, ParallelLMHead, VocabParallelEmbedding)
44 from vllm.model_executor.model_loader.weight_utils import (
45 default_weight_loader, maybe_remap_kv_scale_name)
46 from vllm.model_executor.sampling_metadata import SamplingMetadata
47 -from vllm.sequence import IntermediateTensors, SamplerOutput
47 +from vllm.sequence import IntermediateTensors
48
49 from .interfaces import SupportsLoRA
50