@tduncan / Nvidia-Optimization-Stack-VLLM Mirror · read-only / commits / f0cc0e68e3
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, 23:15 UTC

[Misc] Remove import from transformers logging (#5625)

Chang Su committed Jun 18, 2024 at 05:12 UTC f0cc0e68e3ceef6fe43f78bf36df88e6cad28766
1 file changed -3
vllm/model_executor/models/phi3v.py
-3
@@ -18,7 +18,6 @@ from typing import Iterable, List, Literal, Optional, Tuple, TypedDict
18 import torch
19 import torch.nn as nn
20 from transformers import CLIPVisionConfig, CLIPVisionModel, PretrainedConfig
21 -from transformers.utils import logging
21
22 from vllm.attention import AttentionMetadata
23 from vllm.config import CacheConfig, VisionLanguageConfig
@@ -35,8 +34,6 @@ from vllm.multimodal import MULTIMODAL_REGISTRY
34 from vllm.multimodal.image import get_dummy_image_data
35 from vllm.sequence import SamplerOutput
36
38 -logger = logging.get_logger(__name__)
39 -
37 _KEYS_TO_MODIFY_MAPPING = {
38 "model.vision_embed_tokens": "vision_embed_tokens",
39 }