[5/N] pass the whole config to model (#9983)

Signed-off-by: youkaichao <youkaichao@gmail.com>
This commit is contained in:
youkaichao
2024-11-09 14:17:28 +08:00
committed by GitHub
parent 49d2a41a86
commit 1a95f10ee7
75 changed files with 583 additions and 654 deletions
+11 -10
View File
@@ -26,16 +26,14 @@ import librosa
import numpy as np
import torch
import torch.nn as nn
from transformers import Qwen2AudioConfig, Qwen2AudioEncoder
from transformers import Qwen2AudioEncoder
from vllm.attention import AttentionMetadata
from vllm.config import CacheConfig, MultiModalConfig
from vllm.config import VllmConfig
from vllm.inputs import (INPUT_REGISTRY, DecoderOnlyInputs, DummyData,
InputContext, token_inputs)
from vllm.logger import init_logger
from vllm.model_executor.layers.logits_processor import LogitsProcessor
from vllm.model_executor.layers.quantization.base_config import (
QuantizationConfig)
from vllm.model_executor.layers.sampler import SamplerOutput, get_sampler
from vllm.model_executor.layers.vocab_parallel_embedding import ParallelLMHead
from vllm.model_executor.model_loader.weight_utils import (
@@ -266,13 +264,16 @@ def input_mapper_for_qwen2_audio(
class Qwen2AudioForConditionalGeneration(nn.Module, SupportsMultiModal,
SupportsPP):
def __init__(self,
config: Qwen2AudioConfig,
multimodal_config: MultiModalConfig,
cache_config: Optional[CacheConfig] = None,
quant_config: Optional[QuantizationConfig] = None) -> None:
def __init__(
self,
vllm_config: VllmConfig,
prefix: str = "",
) -> None:
super().__init__()
config = vllm_config.model_config.hf_config
cache_config = vllm_config.cache_config
quant_config = vllm_config.quant_config
multimodal_config = vllm_config.model_config.multimodal_config
self.config = config
self.multimodal_config = multimodal_config