[Frontend] Refine the entrypoint class's inheritance hierarchy. (#47498)
Signed-off-by: wang.yuqi <yuqi.wang@daocloud.io>
This commit is contained in:
@@ -16,13 +16,13 @@ from transformers import PreTrainedTokenizerBase
|
||||
|
||||
import vllm.envs as envs
|
||||
from vllm.engine.protocol import EngineClient
|
||||
from vllm.entrypoints.generate.base.serving import GenerateBaseServing
|
||||
from vllm.entrypoints.openai.engine.protocol import (
|
||||
DeltaMessage,
|
||||
ErrorResponse,
|
||||
RequestResponseMetadata,
|
||||
UsageInfo,
|
||||
)
|
||||
from vllm.entrypoints.openai.engine.serving import OpenAIServing
|
||||
from vllm.entrypoints.openai.models.serving import OpenAIServingModels
|
||||
from vllm.entrypoints.serve.engine.typing import SpeechToTextRequest
|
||||
from vllm.entrypoints.serve.utils.api_utils import get_max_tokens
|
||||
@@ -87,7 +87,7 @@ def asr_inter_chunk_separator(
|
||||
return "" if language and language.lower() in no_space_languages else " "
|
||||
|
||||
|
||||
class OpenAISpeechToText(OpenAIServing):
|
||||
class SpeechToTextBaseServing(GenerateBaseServing):
|
||||
"""Base class for speech-to-text operations like transcription and
|
||||
translation."""
|
||||
|
||||
@@ -475,7 +475,7 @@ class OpenAISpeechToText(OpenAIServing):
|
||||
list_result_generator: list[AsyncGenerator[RequestOutput, None]] | None = None
|
||||
|
||||
input_len = (
|
||||
OpenAISpeechToText._get_decoder_prompt_len(engine_inputs)
|
||||
SpeechToTextBaseServing._get_decoder_prompt_len(engine_inputs)
|
||||
if request.use_beam_search
|
||||
else 0
|
||||
)
|
||||
|
||||
@@ -9,7 +9,7 @@ from typing import Literal, cast
|
||||
import numpy as np
|
||||
|
||||
from vllm.engine.protocol import EngineClient, StreamingInput
|
||||
from vllm.entrypoints.openai.engine.serving import OpenAIServing
|
||||
from vllm.entrypoints.generate.base.serving import GenerateBaseServing
|
||||
from vllm.entrypoints.openai.models.serving import OpenAIServingModels
|
||||
from vllm.entrypoints.serve.utils.request_logger import RequestLogger
|
||||
from vllm.inputs import PromptType
|
||||
@@ -20,7 +20,7 @@ from vllm.renderers.inputs.preprocess import parse_model_prompt
|
||||
logger = init_logger(__name__)
|
||||
|
||||
|
||||
class OpenAIServingRealtime(OpenAIServing):
|
||||
class OpenAIServingRealtime(GenerateBaseServing):
|
||||
"""Realtime audio transcription service via WebSocket streaming.
|
||||
|
||||
Provides streaming audio-to-text transcription by transforming audio chunks
|
||||
|
||||
@@ -14,7 +14,7 @@ from vllm.entrypoints.serve.utils.request_logger import RequestLogger
|
||||
from vllm.logger import init_logger
|
||||
from vllm.outputs import RequestOutput
|
||||
|
||||
from ..base.serving import OpenAISpeechToText
|
||||
from ..base.serving import SpeechToTextBaseServing
|
||||
from .protocol import (
|
||||
TranscriptionRequest,
|
||||
TranscriptionResponse,
|
||||
@@ -26,7 +26,7 @@ from .protocol import (
|
||||
logger = init_logger(__name__)
|
||||
|
||||
|
||||
class OpenAIServingTranscription(OpenAISpeechToText):
|
||||
class OpenAIServingTranscription(SpeechToTextBaseServing):
|
||||
"""Handles transcription requests."""
|
||||
|
||||
def __init__(
|
||||
|
||||
@@ -14,7 +14,7 @@ from vllm.entrypoints.serve.utils.request_logger import RequestLogger
|
||||
from vllm.logger import init_logger
|
||||
from vllm.outputs import RequestOutput
|
||||
|
||||
from ..base.serving import OpenAISpeechToText
|
||||
from ..base.serving import SpeechToTextBaseServing
|
||||
from .protocol import (
|
||||
TranslationRequest,
|
||||
TranslationResponse,
|
||||
@@ -26,7 +26,7 @@ from .protocol import (
|
||||
logger = init_logger(__name__)
|
||||
|
||||
|
||||
class OpenAIServingTranslation(OpenAISpeechToText):
|
||||
class OpenAIServingTranslation(SpeechToTextBaseServing):
|
||||
"""Handles translation requests."""
|
||||
|
||||
def __init__(
|
||||
|
||||
Reference in New Issue
Block a user