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

Fix signature for stop parameter

Browse Source
This commit is contained in:
Andrei Betlen
2023-04-14 09:59:08 -04:00
parent 7dc0838fff
commit 26cc4ee029
1 changed files with 6 additions and 4 deletions
Download Patch File Download Diff File Expand all files Collapse all files

10
llama_cpp/llama.py
View File

@@ -323,7 +323,7 @@ class Llama:
top_p: float = 0.95, top_p: float = 0.95,
logprobs: Optional[int] = None, logprobs: Optional[int] = None,
echo: bool = False, echo: bool = False,
stop: List[str] = [], stop: Optional[List[str]] = [],
repeat_penalty: float = 1.1, repeat_penalty: float = 1.1,
top_k: int = 40, top_k: int = 40,
stream: bool = False, stream: bool = False,
@@ -336,6 +336,7 @@ class Llama:
prompt_tokens = self.tokenize(b" " + prompt.encode("utf-8")) prompt_tokens = self.tokenize(b" " + prompt.encode("utf-8"))
text = b"" text = b""
returned_characters = 0 returned_characters = 0
stop = stop if not None else []
if self.verbose: if self.verbose:
llama_cpp.llama_reset_timings(self.ctx) llama_cpp.llama_reset_timings(self.ctx)
@@ -537,7 +538,7 @@ class Llama:
top_p: float = 0.95, top_p: float = 0.95,
logprobs: Optional[int] = None, logprobs: Optional[int] = None,
echo: bool = False, echo: bool = False,
stop: List[str] = [], stop: Optional[List[str]] = [],
repeat_penalty: float = 1.1, repeat_penalty: float = 1.1,
top_k: int = 40, top_k: int = 40,
stream: bool = False, stream: bool = False,
@@ -592,7 +593,7 @@ class Llama:
top_p: float = 0.95, top_p: float = 0.95,
logprobs: Optional[int] = None, logprobs: Optional[int] = None,
echo: bool = False, echo: bool = False,
stop: List[str] = [], stop: Optional[List[str]] = [],
repeat_penalty: float = 1.1, repeat_penalty: float = 1.1,
top_k: int = 40, top_k: int = 40,
stream: bool = False, stream: bool = False,
@@ -698,7 +699,7 @@ class Llama:
top_p: float = 0.95, top_p: float = 0.95,
top_k: int = 40, top_k: int = 40,
stream: bool = False, stream: bool = False,
stop: List[str] = [], stop: Optional[List[str]] = [],
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]]:
@@ -717,6 +718,7 @@ class Llama:
Returns: Returns:
Generated chat completion or a stream of chat completion chunks. Generated chat completion or a stream of chat completion chunks.
""" """
stop = stop if not None else []
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"]}'