nip.code_validation.agents._ParsedChatCompletion#
- class nip.code_validation.agents._ParsedChatCompletion(next_messages: OrderedDict[str, str] | None, raw_message: str, prompt: list[PromptMessage], discrete_decision: Literal[0, 1, 2, 3], continuous_decision: float, raw_decision: str, retry_count: int, warning: Literal['max_tokens', 'unparsable_response'] | None)[source]#
A parsed chat completion output.
- next_messages#
A dictionary mapping channel names to next messages, ordered by channel order. If the model has made a decision, this will be None.
- prompt#
The prompt used to generate the message.
- Type:
- discrete_decision#
The discrete decision from the verifier model, with the following meanings: - 0: reject - 1: accept - 3: neither accept nor reject
- Type:
Literal[0, 1, 3]
- continuous_decision#
The continuous decision from the verifier model. This is a number between -1 and 1, where -1 is “reject” and 1 is “accept”. This is a more fine-grained version of
discrete_decision
.- Type:
- raw_decision#
The raw decision text from the verifier model. This is the text which appears after “Decision: “ in the completion text.
- Type:
- warning#
The warning message from the generation, if any. One of:
“max_tokens”: If the generation was stopped because the maximum number of tokens was reached.
“unparsable_response”: If the generation could not be parsed.
- Type:
Literal[“max_tokens”, “unparsable_response”] | None
Methods Summary
__eq__
(other)Return self==value.
__init__
(next_messages, raw_message, prompt, ...)__repr__
()Return repr(self).
Attributes
retry_count
Methods
- __eq__(other)#
Return self==value.
- __init__(next_messages: OrderedDict[str, str] | None, raw_message: str, prompt: list[PromptMessage], discrete_decision: Literal[0, 1, 2, 3], continuous_decision: float, raw_decision: str, retry_count: int, warning: Literal['max_tokens', 'unparsable_response'] | None) None #
- __repr__()#
Return repr(self).