Skip to content

Commit

Permalink
Merge branch 'dev/header' of github.com:narugo1992/huggingface_hub in…
Browse files Browse the repository at this point in the history 
…to dev/header
  • Loading branch information
@hanouticelina
hanouticelina committed Feb 4, 2025
2 parents 62c9307 + 67692f5 commit fe5211f
Show file tree
Hide file tree
Showing 5 changed files with 84 additions and 36 deletions.
56 changes: 40 additions & 16 deletions src/huggingface_hub/inference/_client.py
View file
Original file line number Diff line number Diff line change
Expand Up @@ -463,6 +463,7 @@ def automatic_speech_recognition(
audio: ContentT,
*,
model: Optional[str] = None,
extra_body: Optional[Dict] = None,
) -> AutomaticSpeechRecognitionOutput:
"""
Perform automatic speech recognition (ASR or audio-to-text) on the given audio content.
Expand All @@ -473,6 +474,9 @@ def automatic_speech_recognition(
model (`str`, *optional*):
The model to use for ASR. Can be a model ID hosted on the Hugging Face Hub or a URL to a deployed
Inference Endpoint. If not provided, the default recommended model for ASR will be used.
extra_body (`Dict`, *optional*):
Additional provider-specific parameters to pass to the model. Refer to the provider's documentation
for supported parameters.
Returns:
[`AutomaticSpeechRecognitionOutput`]: An item containing the transcribed text and optionally the timestamp chunks.
Expand All @@ -493,7 +497,7 @@ def automatic_speech_recognition(
provider_helper = get_provider_helper(self.provider, task="automatic-speech-recognition")
request_parameters = provider_helper.prepare_request(
inputs=audio,
parameters={},
parameters={**(extra_body or {})},
headers=self.headers,
model=model or self.model,
api_key=self.token,
Expand Down Expand Up @@ -524,6 +528,7 @@ def chat_completion( # type: ignore
tools: Optional[List[ChatCompletionInputTool]] = None,
top_logprobs: Optional[int] = None,
top_p: Optional[float] = None,
extra_body: Optional[Dict] = None,
) -> ChatCompletionOutput: ...

@overload
Expand All @@ -549,6 +554,7 @@ def chat_completion( # type: ignore
tools: Optional[List[ChatCompletionInputTool]] = None,
top_logprobs: Optional[int] = None,
top_p: Optional[float] = None,
extra_body: Optional[Dict] = None,
) -> Iterable[ChatCompletionStreamOutput]: ...

@overload
Expand All @@ -574,6 +580,7 @@ def chat_completion(
tools: Optional[List[ChatCompletionInputTool]] = None,
top_logprobs: Optional[int] = None,
top_p: Optional[float] = None,
extra_body: Optional[Dict] = None,
) -> Union[ChatCompletionOutput, Iterable[ChatCompletionStreamOutput]]: ...

def chat_completion(
Expand All @@ -599,6 +606,7 @@ def chat_completion(
tools: Optional[List[ChatCompletionInputTool]] = None,
top_logprobs: Optional[int] = None,
top_p: Optional[float] = None,
extra_body: Optional[Dict] = None,
) -> Union[ChatCompletionOutput, Iterable[ChatCompletionStreamOutput]]:
"""
A method for completing conversations using a specified language model.
Expand All @@ -613,7 +621,7 @@ def chat_completion(
</Tip>
<Tip>
Some parameters might not be supported by some providers.
You can pass provider-specific parameters to the model by using the `extra_body` argument.
</Tip>
Args:
Expand Down Expand Up @@ -668,7 +676,9 @@ def chat_completion(
tools (List of [`ChatCompletionInputTool`], *optional*):
A list of tools the model may call. Currently, only functions are supported as a tool. Use this to
provide a list of functions the model may generate JSON inputs for.
extra_body (`Dict`, *optional*):
Additional provider-specific parameters to pass to the model. Refer to the provider's documentation
for supported parameters.
Returns:
[`ChatCompletionOutput`] or Iterable of [`ChatCompletionStreamOutput`]:
Generated text returned from the server:
Expand Down Expand Up @@ -753,7 +763,7 @@ def chat_completion(
print(chunk.choices[0].delta.content)
```
Example using a third-party provider directly. Usage will be billed on your Together AI account.
Example using a third-party provider directly with extra (provider-specific) parameters. Usage will be billed on your Together AI account.
```py
>>> from huggingface_hub import InferenceClient
>>> client = InferenceClient(
Expand All @@ -763,6 +773,7 @@ def chat_completion(
>>> client.chat_completion(
... model="meta-llama/Meta-Llama-3-8B-Instruct",
... messages=[{"role": "user", "content": "What is the capital of France?"}],
... extra_body={"safety_model": "Meta-Llama/Llama-Guard-7b"},
... )
```
Expand Down Expand Up @@ -956,6 +967,7 @@ def chat_completion(
"top_p": top_p,
"stream": stream,
"stream_options": stream_options,
**(extra_body or {}),
}
request_parameters = provider_helper.prepare_request(
inputs=messages,
Expand Down Expand Up @@ -2390,7 +2402,7 @@ def text_to_image(
model: Optional[str] = None,
scheduler: Optional[str] = None,
seed: Optional[int] = None,
extra_parameters: Optional[Dict[str, Any]] = None,
extra_body: Optional[Dict[str, Any]] = None,
) -> "Image":
"""
Generate an image based on a given text using a specified model.
Expand All @@ -2401,6 +2413,10 @@ def text_to_image(
</Tip>
<Tip>
You can pass provider-specific parameters to the model by using the `extra_body` argument.
</Tip>
Args:
prompt (`str`):
The prompt to generate an image from.
Expand All @@ -2424,7 +2440,7 @@ def text_to_image(
Override the scheduler with a compatible one.
seed (`int`, *optional*):
Seed for the random number generator.
extra_parameters (`Dict[str, Any]`, *optional*):
extra_body (`Dict[str, Any]`, *optional*):
Additional provider-specific parameters to pass to the model. Refer to the provider's documentation
for supported parameters.
Expand Down Expand Up @@ -2490,7 +2506,7 @@ def text_to_image(
>>> image = client.text_to_image(
... "An astronaut riding a horse on the moon.",
... model="black-forest-labs/FLUX.1-schnell",
... extra_parameters={"output_quality": 100},
... extra_body={"output_quality": 100},
... )
>>> image.save("astronaut.png")
```
Expand All @@ -2506,7 +2522,7 @@ def text_to_image(
"guidance_scale": guidance_scale,
"scheduler": scheduler,
"seed": seed,
**(extra_parameters or {}),
**(extra_body or {}),
},
headers=self.headers,
model=model or self.model,
Expand All @@ -2526,11 +2542,15 @@ def text_to_video(
num_frames: Optional[float] = None,
num_inference_steps: Optional[int] = None,
seed: Optional[int] = None,
extra_parameters: Optional[Dict[str, Any]] = None,
extra_body: Optional[Dict[str, Any]] = None,
) -> bytes:
"""
Generate a video based on a given text.
<Tip>
You can pass provider-specific parameters to the model by using the `extra_body` argument.
</Tip>
Args:
prompt (`str`):
The prompt to generate a video from.
Expand All @@ -2550,7 +2570,7 @@ def text_to_video(
expense of slower inference.
seed (`int`, *optional*):
Seed for the random number generator.
extra_parameters (`Dict[str, Any]`, *optional*):
extra_body (`Dict[str, Any]`, *optional*):
Additional provider-specific parameters to pass to the model. Refer to the provider's documentation
for supported parameters.
Expand Down Expand Up @@ -2598,7 +2618,7 @@ def text_to_video(
"num_frames": num_frames,
"num_inference_steps": num_inference_steps,
"seed": seed,
**(extra_parameters or {}),
**(extra_body or {}),
},
headers=self.headers,
model=model or self.model,
Expand Down Expand Up @@ -2629,11 +2649,15 @@ def text_to_speech(
top_p: Optional[float] = None,
typical_p: Optional[float] = None,
use_cache: Optional[bool] = None,
extra_parameters: Optional[Dict[str, Any]] = None,
extra_body: Optional[Dict[str, Any]] = None,
) -> bytes:
"""
Synthesize an audio of a voice pronouncing a given text.
<Tip>
You can pass provider-specific parameters to the model by using the `extra_body` argument.
</Tip>
Args:
text (`str`):
The text to synthesize.
Expand Down Expand Up @@ -2687,7 +2711,7 @@ def text_to_speech(
paper](https://hf.co/papers/2202.00666) for more details.
use_cache (`bool`, *optional*):
Whether the model should use the past last key/values attentions to speed up decoding
extra_parameters (`Dict[str, Any]`, *optional*):
extra_body (`Dict[str, Any]`, *optional*):
Additional provider-specific parameters to pass to the model. Refer to the provider's documentation
for supported parameters.
Returns:
Expand Down Expand Up @@ -2746,7 +2770,7 @@ def text_to_speech(
>>> audio = client.text_to_speech(
... "Hello, my name is Kororo, an awesome text-to-speech model.",
... model="hexgrad/Kokoro-82M",
... extra_parameters={"voice": "af_nicole"},
... extra_body={"voice": "af_nicole"},
... )
>>> Path("hello.flac").write_bytes(audio)
```
Expand Down Expand Up @@ -2777,7 +2801,7 @@ def text_to_speech(
... model="m-a-p/YuE-s1-7B-anneal-en-cot",
... api_key=...,
... )
>>> audio = client.text_to_speech(lyrics, extra_parameters={"genres": genres})
>>> audio = client.text_to_speech(lyrics, extra_body={"genres": genres})
>>> with open("output.mp3", "wb") as f:
... f.write(audio)
```
Expand All @@ -2802,7 +2826,7 @@ def text_to_speech(
"top_p": top_p,
"typical_p": typical_p,
"use_cache": use_cache,
**(extra_parameters or {}),
**(extra_body or {}),
},
headers=self.headers,
model=model or self.model,
Expand Down
Loading

0 comments on commit fe5211f

Please sign in to comment.