Skip to content

Commit

Permalink
document VLM
Browse files Browse the repository at this point in the history
  • Loading branch information
@Wauplin
Wauplin committed Sep 20, 2024
1 parent e0d39b7 commit 94ac490
Show file tree
Hide file tree
Showing 2 changed files with 81 additions and 12 deletions.
46 changes: 40 additions & 6 deletions src/huggingface_hub/inference/_client.py
View file
Original file line number Diff line number Diff line change
Expand Up @@ -453,7 +453,7 @@ def automatic_speech_recognition(
@overload
def chat_completion( # type: ignore
self,
messages: List[Dict[str, str]],
messages: List[Dict],
*,
model: Optional[str] = None,
stream: Literal[False] = False,
Expand All @@ -478,7 +478,7 @@ def chat_completion( # type: ignore
@overload
def chat_completion( # type: ignore
self,
messages: List[Dict[str, str]],
messages: List[Dict],
*,
model: Optional[str] = None,
stream: Literal[True] = True,
Expand All @@ -503,7 +503,7 @@ def chat_completion( # type: ignore
@overload
def chat_completion(
self,
messages: List[Dict[str, str]],
messages: List[Dict],
*,
model: Optional[str] = None,
stream: bool = False,
Expand All @@ -527,7 +527,7 @@ def chat_completion(

def chat_completion(
self,
messages: List[Dict[str, str]],
messages: List[Dict],
*,
model: Optional[str] = None,
stream: bool = False,
Expand Down Expand Up @@ -562,7 +562,7 @@ def chat_completion(
</Tip>
Args:
messages (List[Union[`SystemMessage`, `UserMessage`, `AssistantMessage`]]):
messages (List of [`ChatCompletionInputMessage`]):
Conversation history consisting of roles and content pairs.
model (`str`, *optional*):
The model to use for chat-completion. Can be a model ID hosted on the Hugging Face Hub or a URL to a deployed
Expand Down Expand Up @@ -665,7 +665,7 @@ def chat_completion(
)
```
Example (stream=True):
Example using streaming:
```py
>>> from huggingface_hub import InferenceClient
>>> messages = [{"role": "user", "content": "What is the capital of France?"}]
Expand Down Expand Up @@ -703,6 +703,40 @@ def chat_completion(
print(chunk.choices[0].delta.content)
```
Example using Image + Text as input:
```py
>>> from huggingface_hub import InferenceClient
# provide a remote URL
>>> image_url ="https://cdn.britannica.com/61/93061-050-99147DCE/Statue-of-Liberty-Island-New-York-Bay.jpg"
# or a base64-encoded image
>>> image_path = "/path/to/image.jpeg"
>>> with open(image_path, "rb") as f:
... base64_image = base64.b64encode(f.read()).decode("utf-8")
>>> image_url = f"data:image/jpeg;base64,{base64_image}"
>>> client = InferenceClient("mistralai/Mistral-Nemo-Instruct-2407")
>>> output = client.chat.completions.create(
... messages=[
... {
... "role": "user",
... "content": [
... {
... "type": "image_url",
... "image_url": {"url": image_url},
... },
... {
... "type": "text",
... "text": "Describe this image in one sentence.",
... },
... ],
... },
... ],
... )
>>> output
A determine figure of Lady Liberty stands tall, holding a torch aloft, atop a pedestal on an island.
```
Example using tools:
```py
>>> client = InferenceClient("meta-llama/Meta-Llama-3-70B-Instruct")
Expand Down
47 changes: 41 additions & 6 deletions src/huggingface_hub/inference/_generated/_async_client.py
View file
Original file line number Diff line number Diff line change
Expand Up @@ -488,7 +488,7 @@ async def automatic_speech_recognition(
@overload
async def chat_completion( # type: ignore
self,
messages: List[Dict[str, str]],
messages: List[Dict],
*,
model: Optional[str] = None,
stream: Literal[False] = False,
Expand All @@ -513,7 +513,7 @@ async def chat_completion( # type: ignore
@overload
async def chat_completion( # type: ignore
self,
messages: List[Dict[str, str]],
messages: List[Dict],
*,
model: Optional[str] = None,
stream: Literal[True] = True,
Expand All @@ -538,7 +538,7 @@ async def chat_completion( # type: ignore
@overload
async def chat_completion(
self,
messages: List[Dict[str, str]],
messages: List[Dict],
*,
model: Optional[str] = None,
stream: bool = False,
Expand All @@ -562,7 +562,7 @@ async def chat_completion(

async def chat_completion(
self,
messages: List[Dict[str, str]],
messages: List[Dict],
*,
model: Optional[str] = None,
stream: bool = False,
Expand Down Expand Up @@ -597,7 +597,7 @@ async def chat_completion(
</Tip>
Args:
messages (List[Union[`SystemMessage`, `UserMessage`, `AssistantMessage`]]):
messages (List of [`ChatCompletionInputMessage`]):
Conversation history consisting of roles and content pairs.
model (`str`, *optional*):
The model to use for chat-completion. Can be a model ID hosted on the Hugging Face Hub or a URL to a deployed
Expand Down Expand Up @@ -701,7 +701,7 @@ async def chat_completion(
)
```
Example (stream=True):
Example using streaming:
```py
# Must be run in an async context
>>> from huggingface_hub import AsyncInferenceClient
Expand Down Expand Up @@ -741,6 +741,41 @@ async def chat_completion(
print(chunk.choices[0].delta.content)
```
Example using Image + Text as input:
```py
# Must be run in an async context
>>> from huggingface_hub import AsyncInferenceClient
# provide a remote URL
>>> image_url ="https://cdn.britannica.com/61/93061-050-99147DCE/Statue-of-Liberty-Island-New-York-Bay.jpg"
# or a base64-encoded image
>>> image_path = "/path/to/image.jpeg"
>>> with open(image_path, "rb") as f:
... base64_image = base64.b64encode(f.read()).decode("utf-8")
>>> image_url = f"data:image/jpeg;base64,{base64_image}"
>>> client = AsyncInferenceClient("mistralai/Mistral-Nemo-Instruct-2407")
>>> output = await client.chat.completions.create(
... messages=[
... {
... "role": "user",
... "content": [
... {
... "type": "image_url",
... "image_url": {"url": image_url},
... },
... {
... "type": "text",
... "text": "Describe this image in one sentence.",
... },
... ],
... },
... ],
... )
>>> output
A determine figure of Lady Liberty stands tall, holding a torch aloft, atop a pedestal on an island.
```
Example using tools:
```py
# Must be run in an async context
Expand Down

0 comments on commit 94ac490

Please sign in to comment.