huggingface/text-generation-inference
5
0
Fork 0
mirror of https://github.com/huggingface/text-generation-inference.git synced 2025-09-11 12:24:53 +00:00
Code Issues Packages Projects Releases Wiki Activity

fix: add chat docs to client

Browse Source
This commit is contained in:
drbh 2024-02-21 18:25:01 -05:00
parent c8f2081171
commit 1aa2126206
1 changed files with 70 additions and 3 deletions
Show Stats Download Patch File Download Diff File Expand all files Collapse all files

73
clients/python/text_generation/client.py
View File

@ -79,7 +79,41 @@ class Client:
top_p: Optional[float] = None, top_p: Optional[float] = None,
tools: Optional[List[Tool]] = None, tools: Optional[List[Tool]] = None,
): ):
""" """ """
Given a list of messages, generate a response asynchronously
Args:
messages (`List[Message]`):
List of messages
frequency_penalty (`float`):
The parameter for frequency penalty. 0.0 means no penalty. See [this
paper](https://arxiv.org/pdf/1909.05858.pdf) for more details.
logit_bias (`List[float]`):
Adjust the likelihood of specified tokens
logprobs (`bool`):
Include log probabilities in the response
top_logprobs (`int`):
Include the `n` most likely tokens at each step
max_tokens (`int`):
Maximum number of generated tokens
n (`int`):
Generate `n` completions
presence_penalty (`float`):
The parameter for presence penalty. 0.0 means no penalty. See [this
paper](https://arxiv.org/pdf/1909.05858.pdf) for more details.
stream (`bool`):
Stream the response
seed (`int`):
Random sampling seed
temperature (`float`):
The value used to module the logits distribution.
top_p (`float`):
If set to < 1, only the smallest set of most probable tokens with probabilities that add up to `top_p` or
higher are kept for generation
tools (`List[Tool]`):
List of tools to use
"""
request = ChatRequest( request = ChatRequest(
model="tgi", model="tgi",
messages=messages, messages=messages,
@ -379,8 +413,41 @@ class AsyncClient:
top_p: Optional[float] = None, top_p: Optional[float] = None,
tools: Optional[List[Tool]] = None, tools: Optional[List[Tool]] = None,
): ):
""" """ """
print("chat") Given a list of messages, generate a response asynchronously
Args:
messages (`List[Message]`):
List of messages
frequency_penalty (`float`):
The parameter for frequency penalty. 0.0 means no penalty. See [this
paper](https://arxiv.org/pdf/1909.05858.pdf) for more details.
logit_bias (`List[float]`):
Adjust the likelihood of specified tokens
logprobs (`bool`):
Include log probabilities in the response
top_logprobs (`int`):
Include the `n` most likely tokens at each step
max_tokens (`int`):
Maximum number of generated tokens
n (`int`):
Generate `n` completions
presence_penalty (`float`):
The parameter for presence penalty. 0.0 means no penalty. See [this
paper](https://arxiv.org/pdf/1909.05858.pdf) for more details.
stream (`bool`):
Stream the response
seed (`int`):
Random sampling seed
temperature (`float`):
The value used to module the logits distribution.
top_p (`float`):
If set to < 1, only the smallest set of most probable tokens with probabilities that add up to `top_p` or
higher are kept for generation
tools (`List[Tool]`):
List of tools to use
"""
request = ChatRequest( request = ChatRequest(
model="tgi", model="tgi",
messages=messages, messages=messages,