Bases: CustomLLM
Cleanlab TLM.
Examples:
pip install llama-index-llms-cleanlab
from llama_index.llms.cleanlab import CleanlabTLM
llm = CleanlabTLM(api_key=api_key, quality_preset="best", options={"log": ["explanation"]})
resp = llm.complete("Who is Paul Graham?")
print(resp)
Arguments:
quality_preset and options are configuration settings you can optionally specify to improve latency or accuracy.
More information can be found here
https://help.cleanlab.ai/tlm/
Source code in llama-index-integrations/llms/llama-index-llms-cleanlab/llama_index/llms/cleanlab/base.py
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
61
62
63
64
65
66
67
68
69
70
71
72
73
74
75
76
77
78
79
80
81
82
83
84
85
86
87
88
89
90
91
92
93
94
95
96
97
98
99
100
101
102
103
104
105
106
107
108
109
110
111
112
113
114
115
116
117
118
119
120
121
122
123 | class CleanlabTLM(CustomLLM):
"""
Cleanlab TLM.
Examples:
`pip install llama-index-llms-cleanlab`
```python
from llama_index.llms.cleanlab import CleanlabTLM
llm = CleanlabTLM(api_key=api_key, quality_preset="best", options={"log": ["explanation"]})
resp = llm.complete("Who is Paul Graham?")
print(resp)
```
Arguments:
`quality_preset` and `options` are configuration settings you can optionally specify to improve latency or accuracy.
More information can be found here:
https://help.cleanlab.ai/tlm/
"""
model: str = Field(
default=DEFAULT_MODEL,
description="Base LLM to use with TLM.",
)
max_tokens: int = Field(
default=DEFAULT_MAX_TOKENS,
description="The maximum number of tokens to generate in TLM response.",
)
_client: Any = PrivateAttr()
def __init__(
self,
api_key: Optional[str] = None,
quality_preset: Optional[str] = DEFAULT_QUALITY_PRESET,
options: Optional[Dict] = None,
callback_manager: Optional[CallbackManager] = None,
additional_kwargs: Optional[Dict[str, Any]] = None,
) -> None:
super().__init__(
additional_kwargs=additional_kwargs or {},
callback_manager=callback_manager,
)
self.max_tokens = (
options.get("max_tokens")
if options and "max_tokens" in options
else DEFAULT_MAX_TOKENS
)
api_key = get_from_param_or_env("api_key", api_key, "CLEANLAB_API_KEY")
self._client = TLM(api_key=api_key, quality_preset=quality_preset, options=options)
self.model = self._client.get_model_name()
@classmethod
def class_name(cls) -> str:
return "CleanlabTLM"
@property
def metadata(self) -> LLMMetadata:
"""Get LLM metadata."""
return LLMMetadata(
context_window=get_default_context_limit(),
num_output=self.max_tokens,
model_name=self.model,
)
def _parse_response(self, response: Dict) -> CompletionResponse:
"""Parse the response from TLM and return a CompletionResponse object."""
try:
text = response["response"]
trust_score = response["trustworthiness_score"]
except KeyError as e:
raise ValueError(f"Missing expected key in response: {e}")
additional_data = {"trustworthiness_score": trust_score}
if "log" in response and "explanation" in response["log"]:
additional_data["explanation"] = response["log"]["explanation"]
return CompletionResponse(text=text, additional_kwargs=additional_data)
@llm_completion_callback()
def complete(self, prompt: str, **kwargs: Any) -> CompletionResponse:
response = self._client.prompt(prompt)
return self._parse_response(response)
@llm_completion_callback()
def stream_complete(self, prompt: str, **kwargs: Any) -> CompletionResponseGen:
# Raise implementation error since TLM doesn't support native streaming
raise NotImplementedError(
"Streaming is not supported in TLM. Instead stream in the response from the LLM and subsequently use TLM to score its trustworthiness."
)
|
