humanloop 0.5.0a2

Client for Humanloop API

humanloop@0.5.0a2

Requirements

Python >=3.7

Installing

pip install humanloop==0.5.0a2

Getting Started

from pprint import pprint
from humanloop import Humanloop, ApiException

humanloop = Humanloop(
    api_key="YOUR_API_KEY",
    openai_api_key="YOUR_OPENAI_API_KEY",
    ai21_api_key="YOUR_AI21_API_KEY",
    mock_api_key="YOUR_MOCK_API_KEY",
    anthropic_api_key="YOUR_ANTHROPIC_API_KEY",
)

try:
    # Chat
    chat_response = humanloop.chat(
        project="sdk-example",
        messages=[
            {
                "role": "user",
                "content": "Explain asynchronous programming.",
            }
        ],
        model_config={
            "model": "gpt-3.5-turbo",
            "max_tokens": -1,
            "temperature": 0.7,
            "chat_template": [
                {
                    "role": "system",
                    "content": "You are a helpful assistant who replies in the style of {{persona}}.",
                },
            ],
        },
        inputs={
            "persona": "the pirate Blackbeard",
        },
        stream=False,
    )
    pprint(chat_response.body)
    pprint(chat_response.body["project_id"])
    pprint(chat_response.body["data"][0])
    pprint(chat_response.body["provider_responses"])
    pprint(chat_response.headers)
    pprint(chat_response.status)
    pprint(chat_response.round_trip_time)
except ApiException as e:
    print("Exception when calling .chat: %s\n" % e)
    pprint(e.body)
    if e.status == 422:
        pprint(e.body["detail"])
    pprint(e.headers)
    pprint(e.status)
    pprint(e.reason)
    pprint(e.round_trip_time)

try:
    # Complete
    complete_response = humanloop.complete(
        project="sdk-example",
        inputs={
            "text": "Llamas that are well-socialized and trained to halter and lead after weaning and are very friendly and pleasant to be around. They are extremely curious and most will approach people easily. However, llamas that are bottle-fed or over-socialized and over-handled as youth will become extremely difficult to handle when mature, when they will begin to treat humans as they treat each other, which is characterized by bouts of spitting, kicking and neck wrestling.[33]",
        },
        model_config={
            "model": "gpt-3.5-turbo",
            "max_tokens": -1,
            "temperature": 0.7,
            "prompt_template": "Summarize this for a second-grade student:\n\nText:\n{{text}}\n\nSummary:\n",
        },
        stream=False,
    )
    pprint(complete_response.body)
    pprint(complete_response.body["project_id"])
    pprint(complete_response.body["data"][0])
    pprint(complete_response.body["provider_responses"])
    pprint(complete_response.headers)
    pprint(complete_response.status)
    pprint(complete_response.round_trip_time)
except ApiException as e:
    print("Exception when calling .complete: %s\n" % e)
    pprint(e.body)
    if e.status == 422:
        pprint(e.body["detail"])
    pprint(e.headers)
    pprint(e.status)
    pprint(e.reason)
    pprint(e.round_trip_time)

try:
    # Feedback
    feedback_response = humanloop.feedback(
        type="rating",
        value="good",
        data_id="data_[...]",
        user="[e-mail]",
    )
    pprint(feedback_response.body)
    pprint(feedback_response.headers)
    pprint(feedback_response.status)
    pprint(feedback_response.round_trip_time)
except ApiException as e:
    print("Exception when calling .feedback: %s\n" % e)
    pprint(e.body)
    if e.status == 422:
        pprint(e.body["detail"])
    pprint(e.headers)
    pprint(e.status)
    pprint(e.reason)
    pprint(e.round_trip_time)

try:
    # Log
    log_response = humanloop.log(
        project="sdk-example",
        inputs={
            "text": "Llamas that are well-socialized and trained to halter and lead after weaning and are very friendly and pleasant to be around. They are extremely curious and most will approach people easily. However, llamas that are bottle-fed or over-socialized and over-handled as youth will become extremely difficult to handle when mature, when they will begin to treat humans as they treat each other, which is characterized by bouts of spitting, kicking and neck wrestling.[33]",
        },
        output="Llamas can be friendly and curious if they are trained to be around people, but if they are treated too much like pets when they are young, they can become difficult to handle when they grow up. This means they might spit, kick, and wrestle with their necks.",
        source="sdk",
        config={
            "model": "gpt-3.5-turbo",
            "max_tokens": -1,
            "temperature": 0.7,
            "prompt_template": "Summarize this for a second-grade student:\n\nText:\n{{text}}\n\nSummary:\n",
            "type": "model",
        },
    )
    pprint(log_response.body)
    pprint(log_response.headers)
    pprint(log_response.status)
    pprint(log_response.round_trip_time)
except ApiException as e:
    print("Exception when calling .log: %s\n" % e)
    pprint(e.body)
    if e.status == 422:
        pprint(e.body["detail"])
    pprint(e.headers)
    pprint(e.status)
    pprint(e.reason)
    pprint(e.round_trip_time)

Async

async support is available by prepending a to any method.

import asyncio
from pprint import pprint
from humanloop import Humanloop, ApiException

humanloop = Humanloop(
    api_key="YOUR_API_KEY",
    openai_api_key="YOUR_OPENAI_API_KEY",
    ai21_api_key="YOUR_AI21_API_KEY",
    mock_api_key="YOUR_MOCK_API_KEY",
    anthropic_api_key="YOUR_ANTHROPIC_API_KEY",
)


async def main():
    try:
        complete_response = await humanloop.acomplete(
            project="sdk-example",
            inputs={
                "text": "Llamas that are well-socialized and trained to halter and lead after weaning and are very friendly and pleasant to be around. They are extremely curious and most will approach people easily. However, llamas that are bottle-fed or over-socialized and over-handled as youth will become extremely difficult to handle when mature, when they will begin to treat humans as they treat each other, which is characterized by bouts of spitting, kicking and neck wrestling.[33]",
            },
            model_config={
                "model": "gpt-3.5-turbo",
                "max_tokens": -1,
                "temperature": 0.7,
                "prompt_template": "Summarize this for a second-grade student:\n\nText:\n{{text}}\n\nSummary:\n",
            },
            stream=False,
        )
        pprint(complete_response.body)
        pprint(complete_response.body["project_id"])
        pprint(complete_response.body["data"][0])
        pprint(complete_response.body["provider_responses"])
        pprint(complete_response.headers)
        pprint(complete_response.status)
        pprint(complete_response.round_trip_time)
    except ApiException as e:
        print("Exception when calling .complete: %s\n" % e)
        pprint(e.body)
        if e.status == 422:
            pprint(e.body["detail"])
        pprint(e.headers)
        pprint(e.status)
        pprint(e.reason)
        pprint(e.round_trip_time)


asyncio.run(main())

Streaming

Streaming support is available by suffixing a chat or complete method with _stream.

import asyncio
from humanloop import Humanloop

humanloop = Humanloop(
    api_key="YOUR_API_KEY",
    openai_api_key="YOUR_OPENAI_API_KEY",
    ai21_api_key="YOUR_AI21_API_KEY",
    mock_api_key="YOUR_MOCK_API_KEY",
    anthropic_api_key="YOUR_ANTHROPIC_API_KEY",
)


async def main():
    response = await humanloop.chat_stream(
        project="sdk-example",
        messages=[
            {
                "role": "user",
                "content": "Explain asynchronous programming.",
            }
        ],
        model_config={
            "model": "gpt-3.5-turbo",
            "max_tokens": -1,
            "temperature": 0.7,
            "chat_template": [
                {
                    "role": "system",
                    "content": "You are a helpful assistant who replies in the style of {{persona}}.",
                },
            ],
        },
        inputs={
            "persona": "the pirate Blackbeard",
        },
    )
    async for token in response.content:
        print(token)


asyncio.run(main())

Documentation for API Endpoints

All URIs are relative to https://api.humanloop.com/v4

Tag	Method	HTTP request	Description
Chats	create	POST /chat	Get a chat response by providing details of the model configuration in the request.
Chats	create_deployed	POST /chat-deployed	Get a chat response using the project's active deployment. The active deployment can be a specific model configuration or an experiment.
Chats	create_experiment	POST /chat-experiment	Get a chat response for a specific experiment.
Chats	create_model_config	POST /chat-model-config	Get chat response for a specific model configuration.
Completions	create	POST /completion	Create a completion by providing details of the model configuration in the request.
Completions	create_deployed	POST /completion-deployed	Create a completion using the project's active deployment. The active deployment can be a specific model configuration or an experiment.
Completions	create_experiment	POST /completion-experiment	Create a completion for a specific experiment.
Completions	create_model_config	POST /completion-model-config	Create a completion for a specific model configuration.
Evaluations	create	POST /projects/{project_id}/evaluations	Create an evaluation.
Evaluations	get	GET /evaluations/{id}	Get evaluation by ID.
Evaluations	get_all_for_project	GET /projects/{project_id}/evaluations	Get all the evaluations associated with your project.
Evaluations	get_testcases	GET /evaluations/{id}/testcases	Get testcases by evaluation ID.
Evaluators	create	POST /evaluation-functions	Create an evaluator within your organization.
Evaluators	delete	DELETE /evaluation-functions/{id}	Delete an evaluator within your organization.
Evaluators	get_all	GET /evaluation-functions	Get all evaluators within your organization.
Evaluators	update	PATCH /evaluation-functions/{id}	Update an evaluator within your organization.
Experiments	create	POST /projects/{project_id}/experiments	Create an experiment for your project. You can optionally specify IDs of your project's model configs to include in the experiment, along with a set of labels to consider as positive feedback and whether the experiment should be set as active.
Experiments	delete	DELETE /experiments/{experiment_id}	Delete the experiment with the specified ID.
Experiments	get_all	GET /projects/{project_id}/experiments	Get an array of experiments associated to your project.
Experiments	sample	GET /experiments/{experiment_id}/model-config	Samples a model config from the experiment's active model configs.
Experiments	update	PATCH /experiments/{experiment_id}	Update your experiment, including registering and de-registering model configs.
Feedback	feedback	POST /feedback	Submit an array of feedback for existing `data_ids`
Logs	log	POST /logs	Log a datapoint or array of datapoints to your Humanloop project.
Logs	update	PATCH /logs/{id}	Update a logged datapoint in your Humanloop project.
Logs	update_by_ref	PATCH /logs	Update a logged datapoint by its reference ID. The `reference_id` query parameter must be provided, and refers to the `reference_id` of a previously-logged datapoint.
Model Configs	get	GET /model-configs/{id}	Get a specific model config by ID.
Model Configs	register	POST /model-configs	Register a model config to a project and optionally add it to an experiment. If the project provided does not exist, a new project will be created automatically. If an experiment name is provided, the specified experiment must already exist. Otherwise, an error will be raised. If the model config is the first to be associated to the project, it will be set as the active model config.
Projects	create	POST /projects	Create a new project.
Projects	deactivate_config	DELETE /projects/{id}/active-config	Remove the project's active config, if set. This has no effect if the project does not have an active model config set.
Projects	deactivate_experiment	DELETE /projects/{id}/active-experiment	Remove the project's active experiment, if set. This has no effect if the project does not have an active experiment set.
Projects	delete_deployed_config	DELETE /projects/{project_id}/deployed-config/{environment_id}	Remove the model config deployed to environment. This has no effect if the project does not have an active model config set.
Projects	deploy_config	PATCH /projects/{project_id}/deploy-config	Deploy a model config to an environment. If the environment already has a model config deployed, it will be replaced.
Projects	export	POST /projects/{id}/export	Export all logged datapoints associated to your project. Results are paginated and sorts the datapoints based on `created_at` in descending order.
Projects	get	GET /projects/{id}	Get a specific project.
Projects	get_active_config	GET /projects/{id}/active-config	Retrieves a config to use to execute your model. A config will be selected based on the project's active config/experiment settings.
Projects	get_all	GET /projects	Get a paginated list of projects.
Projects	get_configs	GET /projects/{id}/configs	Get an array of configs associated to your project.
Projects	get_deployed_configs	GET /projects/{id}/deployed-configs	Get an array of environments with the deployed configs associated to your project.
Projects	update	PATCH /projects/{id}	Update a specific project. Set the project's active model config/experiment by passing either `active_experiment_id` or `active_model_config_id`. These will be set to the Default environment unless a list of environments are also passed in specifically detailing which environments to assign the active config or experiment. Set the feedback labels to be treated as positive user feedback used in calculating top-level project metrics by passing a list of labels in `positive_labels`.
Projects	update_feedback_types	PATCH /projects/{id}/feedback-types	Update feedback types. Allows creation of the default feedback types and setting status of feedback types/categorical values. This behaves like an upsert; any feedback categorical values that do not already exist in the project will be created.
Sessions	create	POST /sessions	Create a new session. Returns a session ID that can be used to log datapoints to the session.
Sessions	get	GET /sessions/{id}	Get a session by ID.
Sessions	get_all	GET /sessions	Get a page of sessions.
Testcases	delete	DELETE /testcases	Delete a list of testsets by their IDs.
Testcases	get	GET /testcases/{id}	Get a testcase by ID.
Testcases	update	PATCH /testcases/{id}	Edit the input, messages and criteria fields of a testcase. The fields passed in the request are the ones edited. Passing `null` as a value for a field will delete that field. In order to signify not changing a field, it should be omitted from the request body.
Testsets	create	POST /projects/{project_id}/testsets	Create a new testset for a project.
Testsets	create_testcase	POST /testsets/{testset_id}/testcases	Create a new testcase for a testset.
Testsets	delete	DELETE /testsets/{id}	Delete a testset by ID.
Testsets	get	GET /testsets/{id}	Get a single testset by ID.
Testsets	get_all_for_project	GET /projects/{project_id}/testsets	Get all testsets for a project.
Testsets	get_testcases	GET /testsets/{testset_id}/testcases	Get testcases for a testset.
Testsets	update	PATCH /testsets/{id}	Update a testset by ID.

Author

This Python package is automatically generated by Konfig