reference.md

Reference

Prompts

client.prompts.log(...)

📝 Description

Log to a Prompt.

You can use query parameters version_id, or environment, to target an existing version of the Prompt. Otherwise, the default deployed version will be chosen.

Instead of targeting an existing version explicitly, you can instead pass in Prompt details in the request body. In this case, we will check if the details correspond to an existing version of the Prompt. If they do not, we will create a new version. This is helpful in the case where you are storing or deriving your Prompt details in code.

🔌 Usage

import datetime

from humanloop import Humanloop

client = Humanloop(
    api_key="YOUR_API_KEY",
)
client.prompts.log(
    path="persona",
    prompt={
        "model": "gpt-4",
        "template": [
            {
                "role": "system",
                "content": "You are {{person}}. Answer questions as this person. Do not break character.",
            }
        ],
    },
    messages=[{"role": "user", "content": "What really happened at Roswell?"}],
    inputs={"person": "Trump"},
    created_at=datetime.datetime.fromisoformat(
        "2024-07-19 00:29:35.178000+00:00",
    ),
    provider_latency=6.5931549072265625,
    output_message={
        "content": "Well, you know, there is so much secrecy involved in government, folks, it's unbelievable. They don't want to tell you everything. They don't tell me everything! But about Roswell, it’s a very popular question. I know, I just know, that something very, very peculiar happened there. Was it a weather balloon? Maybe. Was it something extraterrestrial? Could be. I'd love to go down and open up all the classified documents, believe me, I would. But they don't let that happen. The Deep State, folks, the Deep State. They’re unbelievable. They want to keep everything a secret. But whatever the truth is, I can tell you this: it’s something big, very very big. Tremendous, in fact.",
        "role": "assistant",
    },
    prompt_tokens=100,
    output_tokens=220,
    prompt_cost=1e-05,
    output_cost=0.0002,
    finish_reason="stop",
)

⚙️ Parameters

version_id: typing.Optional[str] — A specific Version ID of the Prompt to log to.

environment: typing.Optional[str] — Name of the Environment identifying a deployed version to log to.

run_id: typing.Optional[str] — Unique identifier for the Run to associate the Log to.

path: typing.Optional[str] — Path of the Prompt, including the name. This locates the Prompt in the Humanloop filesystem and is used as as a unique identifier. For example: folder/name or just name.

id: typing.Optional[str] — ID for an existing Prompt.

output_message: typing.Optional[ChatMessageParams] — The message returned by the provider.

prompt_tokens: typing.Optional[int] — Number of tokens in the prompt used to generate the output.

output_tokens: typing.Optional[int] — Number of tokens in the output generated by the model.

prompt_cost: typing.Optional[float] — Cost in dollars associated to the tokens in the prompt.

output_cost: typing.Optional[float] — Cost in dollars associated to the tokens in the output.

finish_reason: typing.Optional[str] — Reason the generation finished.

messages: typing.Optional[typing.Sequence[ChatMessageParams]] — The messages passed to the to provider chat endpoint.

tool_choice: typing.Optional[PromptLogRequestToolChoiceParams]

Controls how the model uses tools. The following options are supported:

'none' means the model will not call any tool and instead generates a message; this is the default when no tools are provided as part of the Prompt.
'auto' means the model can decide to call one or more of the provided tools; this is the default when tools are provided as part of the Prompt.
'required' means the model can decide to call one or more of the provided tools.
{'type': 'function', 'function': {name': <TOOL_NAME>}} forces the model to use the named function.

prompt: typing.Optional[PromptKernelRequestParams] — Details of your Prompt. A new Prompt version will be created if the provided details are new.

start_time: typing.Optional[dt.datetime] — When the logged event started.

end_time: typing.Optional[dt.datetime] — When the logged event ended.

output: typing.Optional[str] — Generated output from your model for the provided inputs. Can be None if logging an error, or if creating a parent Log with the intention to populate it later.

created_at: typing.Optional[dt.datetime] — User defined timestamp for when the log was created.

error: typing.Optional[str] — Error message if the log is an error.

provider_latency: typing.Optional[float] — Duration of the logged event in seconds.

stdout: typing.Optional[str] — Captured log and debug statements.

provider_request: typing.Optional[typing.Dict[str, typing.Optional[typing.Any]]] — Raw request sent to provider.

provider_response: typing.Optional[typing.Dict[str, typing.Optional[typing.Any]]] — Raw response received the provider.

inputs: typing.Optional[typing.Dict[str, typing.Optional[typing.Any]]] — The inputs passed to the prompt template.

source: typing.Optional[str] — Identifies where the model was called from.

metadata: typing.Optional[typing.Dict[str, typing.Optional[typing.Any]]] — Any additional metadata to record.

source_datapoint_id: typing.Optional[str] — Unique identifier for the Datapoint that this Log is derived from. This can be used by Humanloop to associate Logs to Evaluations. If provided, Humanloop will automatically associate this Log to Evaluations that require a Log for this Datapoint-Version pair.

trace_parent_id: typing.Optional[str] — The ID of the parent Log to nest this Log under in a Trace.

user: typing.Optional[str] — End-user ID related to the Log.

prompt_log_request_environment: typing.Optional[str] — The name of the Environment the Log is associated to.

save: typing.Optional[bool] — Whether the request/response payloads will be stored on Humanloop.

request_options: typing.Optional[RequestOptions] — Request-specific configuration.

client.prompts.update_log(...)

📝 Description

Update a Log.

Update the details of a Log with the given ID.