[Doc] ruff format some Python examples (#26767)

Signed-off-by: DarkLight1337 <tlleungac@connect.ust.hk>
This commit is contained in:
Cyrus Leung
2025-10-14 03:21:53 -07:00
committed by GitHub
parent 70b1b330e1
commit ef9676a1f1
20 changed files with 341 additions and 290 deletions
+2 -2
View File
@@ -73,8 +73,8 @@ def forward(
self,
input_ids: torch.Tensor,
positions: torch.Tensor,
intermediate_tensors: Optional[IntermediateTensors] = None,
inputs_embeds: Optional[torch.Tensor] = None,
intermediate_tensors: IntermediateTensors | None = None,
inputs_embeds: torch.Tensor | None = None,
) -> torch.Tensor:
...
```
+23 -19
View File
@@ -16,7 +16,7 @@ Further update the model as follows:
...
@classmethod
def get_placeholder_str(cls, modality: str, i: int) -> Optional[str]:
def get_placeholder_str(cls, modality: str, i: int) -> str | None:
if modality.startswith("image"):
return "<image>"
@@ -45,14 +45,14 @@ Further update the model as follows:
...
def _process_image_input(self, image_input: YourModelImageInputs) -> torch.Tensor:
assert self.vision_encoder is not None
image_features = self.vision_encoder(image_input)
return self.multi_modal_projector(image_features)
def get_multimodal_embeddings(
self, **kwargs: object) -> Optional[MultiModalEmbeddings]:
self,
**kwargs: object,
) -> MultiModalEmbeddings | None:
# Validate the multimodal input keyword arguments
image_input = self._parse_and_validate_image_input(**kwargs)
if image_input is None:
@@ -110,7 +110,7 @@ to return the maximum number of input items for each modality supported by the m
For example, if the model supports any number of images but only one video per prompt:
```python
def get_supported_mm_limits(self) -> Mapping[str, Optional[int]]:
def get_supported_mm_limits(self) -> Mapping[str, int | None]:
return {"image": None, "video": 1}
```
@@ -258,7 +258,7 @@ Assuming that the memory usage increases with the number of tokens, the dummy in
self,
seq_len: int,
mm_counts: Mapping[str, int],
mm_options: Optional[Mapping[str, BaseDummyOptions]] = None,
mm_options: Mapping[str, BaseDummyOptions] | None = None,
) -> MultiModalDataDict:
num_images = mm_counts.get("image", 0)
@@ -421,8 +421,10 @@ Assuming that the memory usage increases with the number of tokens, the dummy in
```python
def get_image_size_with_most_features(self) -> ImageSize:
image_processor = self.get_image_processor()
return ImageSize(width=image_processor.size["width"],
height=image_processor.size["height"])
return ImageSize(
width=image_processor.size["width"],
height=image_processor.size["height"],
)
```
Fuyu does not expect image placeholders in the inputs to HF processor, so
@@ -452,10 +454,12 @@ Assuming that the memory usage increases with the number of tokens, the dummy in
return {
"image":
self._get_dummy_images(width=target_width,
height=target_height,
num_images=num_images,
overrides=image_overrides)
self._get_dummy_images(
width=target_width,
height=target_height,
num_images=num_images,
overrides=image_overrides,
)
}
```
@@ -744,8 +748,7 @@ Each [PromptUpdate][vllm.multimodal.processing.PromptUpdate] instance specifies
image_width=image_size.width,
image_height=image_size.height,
)
image_tokens = ([_IMAGE_TOKEN_ID] * ncols +
[_NEWLINE_TOKEN_ID]) * nrows
image_tokens = ([_IMAGE_TOKEN_ID] * ncols + [_NEWLINE_TOKEN_ID]) * nrows
return PromptUpdateDetails.select_token_id(
image_tokens + [bos_token_id],
@@ -781,8 +784,7 @@ Each [PromptUpdate][vllm.multimodal.processing.PromptUpdate] instance specifies
image_width=image_size.width,
image_height=image_size.height,
)
image_tokens = ([_IMAGE_TOKEN_ID] * ncols +
[_NEWLINE_TOKEN_ID]) * nrows
image_tokens = ([_IMAGE_TOKEN_ID] * ncols + [_NEWLINE_TOKEN_ID]) * nrows
return PromptUpdateDetails.select_token_id(
image_tokens + [bos_token_id],
@@ -810,9 +812,11 @@ to register them to the multi-modal registry:
from vllm.model_executor.models.interfaces import SupportsMultiModal
+ from vllm.multimodal import MULTIMODAL_REGISTRY
+ @MULTIMODAL_REGISTRY.register_processor(YourMultiModalProcessor,
+ info=YourProcessingInfo,
+ dummy_inputs=YourDummyInputsBuilder)
+ @MULTIMODAL_REGISTRY.register_processor(
+ YourMultiModalProcessor,
+ info=YourProcessingInfo,
+ dummy_inputs=YourDummyInputsBuilder,
+ )
class YourModelForImage2Seq(nn.Module, SupportsMultiModal):
```
+1 -1
View File
@@ -42,7 +42,7 @@ def register():
ModelRegistry.register_model(
"YourModelForCausalLM",
"your_code:YourModelForCausalLM"
"your_code:YourModelForCausalLM",
)
```
+11 -1
View File
@@ -15,6 +15,7 @@ Declare supported languages and capabilities:
- Set `supports_transcription_only=True` if the model should not serve text generation (eg Whisper).
??? code "supported_languages and supports_transcription_only"
```python
from typing import ClassVar, Mapping, Literal
import numpy as np
@@ -43,6 +44,7 @@ Provide an ASR configuration via [get_speech_to_text_config][vllm.model_executor
This is for controlling general behavior of the API when serving your model:
??? code "get_speech_to_text_config()"
```python
class YourASRModel(nn.Module, SupportsTranscription):
...
@@ -71,6 +73,7 @@ Implement the prompt construction via [get_generation_prompt][vllm.model_executo
Return a dict containing `multi_modal_data` with the audio, and either a `prompt` string or `prompt_token_ids`:
??? code "get_generation_prompt()"
```python
class YourASRModel(nn.Module, SupportsTranscription):
...
@@ -107,6 +110,7 @@ Return a dict containing `multi_modal_data` with the audio, and either a `prompt
Return a dict with separate `encoder_prompt` and `decoder_prompt` entries:
??? code "get_generation_prompt()"
```python
class YourASRModel(nn.Module, SupportsTranscription):
...
@@ -148,12 +152,16 @@ Language validation via [validate_language][vllm.model_executor.models.interface
If your model requires a language and you want a default, override this method (see Whisper):
??? code "validate_language()"
```python
@classmethod
def validate_language(cls, language: str | None) -> str | None:
if language is None:
logger.warning(
"Defaulting to language='en'. If you wish to transcribe audio in a different language, pass the `language` field.")
"Defaulting to language='en'. If you wish to transcribe "
"audio in a different language, pass the `language` field "
"in the TranscriptionRequest."
)
language = "en"
return super().validate_language(language)
```
@@ -165,6 +173,7 @@ Token accounting for streaming via [get_num_audio_tokens][vllm.model_executor.mo
Provide a fast duration→token estimate to improve streaming usage statistics:
??? code "get_num_audio_tokens()"
```python
class YourASRModel(nn.Module, SupportsTranscription):
...
@@ -191,6 +200,7 @@ The API server takes care of basic audio I/O and optional chunking before buildi
Relevant server logic:
??? code "_preprocess_speech_to_text()"
```python
# vllm/entrypoints/openai/speech_to_text.py
async def _preprocess_speech_to_text(...):