alihan/llama-cpp-python
1
0
Fork 0
mirror of https://github.com/abetlen/llama-cpp-python.git synced 2023-09-07 17:34:22 +03:00
Code Issues Packages Projects Releases Wiki Activity

Add docstring for create_chat_completion

Browse Source
This commit is contained in:
Andrei Betlen
2023-04-03 20:24:20 -04:00
parent 4615f1e520
commit 4aa349d777
1 changed files with 15 additions and 0 deletions
Download Patch File Download Diff File Expand all files Collapse all files

15
llama_cpp/llama.py
View File

@@ -586,6 +586,21 @@ class Llama:
max_tokens: int = 128, max_tokens: int = 128,
repeat_penalty: float = 1.1, repeat_penalty: float = 1.1,
) -> Union[ChatCompletion, Iterator[ChatCompletionChunk]]: ) -> Union[ChatCompletion, Iterator[ChatCompletionChunk]]:
"""Generate a chat completion from a list of messages.
Args:
messages: A list of messages to generate a response for.
temperature: The temperature to use for sampling.
top_p: The top-p value to use for sampling.
top_k: The top-k value to use for sampling.
stream: Whether to stream the results.
stop: A list of strings to stop generation when encountered.
max_tokens: The maximum number of tokens to generate.
repeat_penalty: The penalty to apply to repeated tokens.
Returns:
Generated chat completion or a stream of chat completion chunks.
"""
instructions = """Complete the following chat conversation between the user and the assistant. System messages should be strictly followed as additional instructions.""" instructions = """Complete the following chat conversation between the user and the assistant. System messages should be strictly followed as additional instructions."""
chat_history = "\n".join( chat_history = "\n".join(
f'{message["role"]} {message.get("user", "")}: {message["content"]}' f'{message["role"]} {message.get("user", "")}: {message["content"]}'