From c9cb10436140a564b0f2fcb66d387310b8b26b23 Mon Sep 17 00:00:00 2001 From: supa-thibaud Date: Wed, 11 Sep 2024 06:38:24 +0200 Subject: [PATCH] dry --- src/engine.py | 20 +++++++++++++++----- 1 file changed, 15 insertions(+), 5 deletions(-) diff --git a/src/engine.py b/src/engine.py index 28e3861..cdd86e5 100644 --- a/src/engine.py +++ b/src/engine.py @@ -89,7 +89,7 @@ def wait_for_server(self, timeout=300, interval=5): return True except requests.RequestException: pass - time.sleep(interval) + time.sleep(interval) raise TimeoutError("Server failed to start within the timeout period.") def shutdown(self): @@ -102,7 +102,7 @@ class OpenAIRequest: def __init__(self, base_url="http://0.0.0.0:30000/v1", api_key="EMPTY"): self.client = openai.Client(base_url=base_url, api_key=api_key) - async def request_chat_completions(self, model="default", messages=None, max_tokens=100, stream=False, frequency_penalty=0.0, n=1, stop=None, temperature=1.0, top_p=1.0): + async def request_chat_completions(self, model="default", messages=None, max_tokens=100, stream=False, frequency_penalty=0.0, n=1, stop=None, temperature=1.0, top_p=1.0, dry_multiplier=0, dry_base=1.75, dry_allowed_length=2, dry_penalty_last_n=0, dry_sequence_breakers='"\\n", ":", "\\"", "*"'): if messages is None: messages = [ {"role": "system", "content": "You are a helpful AI assistant"}, @@ -118,7 +118,12 @@ async def request_chat_completions(self, model="default", messages=None, max_tok n=n, stop=stop, temperature=temperature, - top_p=top_p + top_p=top_p, + dry_multiplier=dry_multiplier, + dry_base=dry_base, + dry_allowed_length=dry_allowed_length, + dry_penalty_last_n=dry_penalty_last_n, + dry_sequence_breakers=dry_sequence_breakers, ) if stream: @@ -127,7 +132,7 @@ async def request_chat_completions(self, model="default", messages=None, max_tok else: yield response.to_dict() - async def request_completions(self, model="default", prompt="The capital of France is", max_tokens=100, stream=False, frequency_penalty=0.0, n=1, stop=None, temperature=1.0, top_p=1.0): + async def request_completions(self, model="default", prompt="The capital of France is", max_tokens=100, stream=False, frequency_penalty=0.0, n=1, stop=None, temperature=1.0, top_p=1.0, dry_multiplier=0, dry_base=1.75, dry_allowed_length=2, dry_penalty_last_n=0, dry_sequence_breakers='"\\n", ":", "\\"", "*"'): response = self.client.completions.create( model=model, prompt=prompt, @@ -137,7 +142,12 @@ async def request_completions(self, model="default", prompt="The capital of Fran n=n, stop=stop, temperature=temperature, - top_p=top_p + top_p=top_p, + dry_multiplier=dry_multiplier, + dry_base=dry_base, + dry_allowed_length=dry_allowed_length, + dry_penalty_last_n=dry_penalty_last_n, + dry_sequence_breakers=dry_sequence_breakers, ) if stream: