humanloop 0.7.0b10

🌐 Endpoint

/chat post

`humanloop.chat_deployed`

Get a chat response using the project's active deployment. The active deployment can be a specific model configuration or an experiment.

🛠️ Usage

create_deployed_response = humanloop.chat_deployed(
    messages=[
        {
            "role": "string_example",
        }
    ],
    project="string_example",
    project_id="string_example",
    session_id="string_example",
    session_reference_id="string_example",
    parent_id="string_example",
    parent_reference_id="string_example",
    inputs={},
    source="string_example",
    metadata={},
    save=True,
    source_datapoint_id="string_example",
    provider_api_keys={},
    num_samples=1,
    stream=False,
    user="string_example",
    seed=1,
    return_inputs=True,
    tool_choice="string_example",
    tool_call="string_example",
    response_format={
        "type": "json_object",
    },
    environment="string_example",
)

⚙️ Parameters

messages: List[`ChatMessageWithToolCall`]

The messages passed to the to provider chat endpoint.

project: `str`

Unique project name. If no project exists with this name, a new project will be created.

project_id: `str`

Unique ID of a project to associate to the log. Either this or project must be provided.

session_id: `str`

ID of the session to associate the datapoint.

session_reference_id: `str`

parent_id: `str`

ID associated to the parent datapoint in a session.

parent_reference_id: `str`

inputs: `Dict[str, Union[bool, date, datetime, dict, float, int, list, str, None]]`

The inputs passed to the prompt template.

source: `str`

Identifies where the model was called from.

metadata: `Dict[str, Union[bool, date, datetime, dict, float, int, list, str, None]]`

Any additional metadata to record.

save: `bool`

Whether the request/response payloads will be stored on Humanloop.

source_datapoint_id: `str`

ID of the source datapoint if this is a log derived from a datapoint in a dataset.

provider_api_keys: `ProviderApiKeys`

API keys required by each provider to make API calls. The API keys provided here are not stored by Humanloop. If not specified here, Humanloop will fall back to the key saved to your organization.

num_samples: `int`

The number of generations.

stream: `bool`

If true, tokens will be sent as data-only server-sent events. If num_samples > 1, samples are streamed back independently.

user: `str`

End-user ID passed through to provider call.

seed: `int`

Deprecated field: the seed is instead set as part of the request.config object.

return_inputs: `bool`

Whether to return the inputs in the response. If false, the response will contain an empty dictionary under inputs. This is useful for reducing the size of the response. Defaults to true.

tool_choice: Union[`str`, `str`, `ToolChoice`]

tool_call: Union[`str`, `Dict[str, str]`]

response_format: `ResponseFormat`

The format of the response. Only type json_object is currently supported for chat.

environment: `str`

The environment name used to create a chat response. If not specified, the default environment will be used.

⚙️ Request Body

ChatDeployedRequest

🔄 Return

🌐 Endpoint

/chat-deployed post

`humanloop.chat_experiment`

Get a chat response for a specific experiment.

🛠️ Usage

create_experiment_response = humanloop.chat_experiment(
    messages=[
        {
            "role": "string_example",
        }
    ],
    experiment_id="string_example",
    project="string_example",
    project_id="string_example",
    session_id="string_example",
    session_reference_id="string_example",
    parent_id="string_example",
    parent_reference_id="string_example",
    inputs={},
    source="string_example",
    metadata={},
    save=True,
    source_datapoint_id="string_example",
    provider_api_keys={},
    num_samples=1,
    stream=False,
    user="string_example",
    seed=1,
    return_inputs=True,
    tool_choice="string_example",
    tool_call="string_example",
    response_format={
        "type": "json_object",
    },
)

⚙️ Parameters

messages: List[`ChatMessageWithToolCall`]

The messages passed to the to provider chat endpoint.

experiment_id: `str`

If an experiment ID is provided a model configuration will be sampled from the experiments active model configurations.

project: `str`

Unique project name. If no project exists with this name, a new project will be created.

project_id: `str`

Unique ID of a project to associate to the log. Either this or project must be provided.

session_id: `str`

ID of the session to associate the datapoint.

session_reference_id: `str`

parent_id: `str`

ID associated to the parent datapoint in a session.

parent_reference_id: `str`

inputs: `Dict[str, Union[bool, date, datetime, dict, float, int, list, str, None]]`

The inputs passed to the prompt template.

source: `str`

Identifies where the model was called from.

metadata: `Dict[str, Union[bool, date, datetime, dict, float, int, list, str, None]]`

Any additional metadata to record.

save: `bool`

Whether the request/response payloads will be stored on Humanloop.

source_datapoint_id: `str`

ID of the source datapoint if this is a log derived from a datapoint in a dataset.

provider_api_keys: `ProviderApiKeys`

API keys required by each provider to make API calls. The API keys provided here are not stored by Humanloop. If not specified here, Humanloop will fall back to the key saved to your organization.

num_samples: `int`

The number of chat responses, where each chat response will use a model configuration sampled from the experiment.

stream: `bool`

If true, tokens will be sent as data-only server-sent events. If num_samples > 1, samples are streamed back independently.

user: `str`

End-user ID passed through to provider call.

seed: `int`

Deprecated field: the seed is instead set as part of the request.config object.

return_inputs: `bool`

Whether to return the inputs in the response. If false, the response will contain an empty dictionary under inputs. This is useful for reducing the size of the response. Defaults to true.

tool_choice: Union[`str`, `str`, `ToolChoice`]

tool_call: Union[`str`, `Dict[str, str]`]

response_format: `ResponseFormat`

The format of the response. Only type json_object is currently supported for chat.

⚙️ Request Body

ChatExperimentRequest

🔄 Return

🌐 Endpoint

/chat-experiment post

`humanloop.chat_model_config`

Get chat response for a specific model configuration.

🛠️ Usage

create_model_config_response = humanloop.chat_model_config(
    messages=[
        {
            "role": "string_example",
        }
    ],
    model_config_id="string_example",
    project="string_example",
    project_id="string_example",
    session_id="string_example",
    session_reference_id="string_example",
    parent_id="string_example",
    parent_reference_id="string_example",
    inputs={},
    source="string_example",
    metadata={},
    save=True,
    source_datapoint_id="string_example",
    provider_api_keys={},
    num_samples=1,
    stream=False,
    user="string_example",
    seed=1,
    return_inputs=True,
    tool_choice="string_example",
    tool_call="string_example",
    response_format={
        "type": "json_object",
    },
)

⚙️ Parameters

messages: List[`ChatMessageWithToolCall`]

The messages passed to the to provider chat endpoint.

model_config_id: `str`

Identifies the model configuration used to create a chat response.

project: `str`

Unique project name. If no project exists with this name, a new project will be created.

project_id: `str`

Unique ID of a project to associate to the log. Either this or project must be provided.

session_id: `str`

ID of the session to associate the datapoint.

session_reference_id: `str`

parent_id: `str`

ID associated to the parent datapoint in a session.

parent_reference_id: `str`

inputs: `Dict[str, Union[bool, date, datetime, dict, float, int, list, str, None]]`

The inputs passed to the prompt template.

source: `str`

Identifies where the model was called from.

metadata: `Dict[str, Union[bool, date, datetime, dict, float, int, list, str, None]]`

Any additional metadata to record.

save: `bool`

Whether the request/response payloads will be stored on Humanloop.

source_datapoint_id: `str`

ID of the source datapoint if this is a log derived from a datapoint in a dataset.

provider_api_keys: `ProviderApiKeys`

API keys required by each provider to make API calls. The API keys provided here are not stored by Humanloop. If not specified here, Humanloop will fall back to the key saved to your organization.

num_samples: `int`

The number of generations.

stream: `bool`

If true, tokens will be sent as data-only server-sent events. If num_samples > 1, samples are streamed back independently.

user: `str`

End-user ID passed through to provider call.

seed: `int`

Deprecated field: the seed is instead set as part of the request.config object.

return_inputs: `bool`

Whether to return the inputs in the response. If false, the response will contain an empty dictionary under inputs. This is useful for reducing the size of the response. Defaults to true.

tool_choice: Union[`str`, `str`, `ToolChoice`]

tool_call: Union[`str`, `Dict[str, str]`]

response_format: `ResponseFormat`

The format of the response. Only type json_object is currently supported for chat.

⚙️ Request Body

ChatModelConfigRequest

🔄 Return

🌐 Endpoint

/chat-model-config post

`humanloop.complete`

Create a completion by providing details of the model configuration in the request.

🛠️ Usage

create_response = humanloop.complete(
    model_config={
        "model": "model_example",
        "max_tokens": -1,
        "temperature": 1,
        "top_p": 1,
        "presence_penalty": 0,
        "frequency_penalty": 0,
        "prompt_template": "{{question}}",
    },
    project="string_example",
    project_id="string_example",
    session_id="string_example",
    session_reference_id="string_example",
    parent_id="string_example",
    parent_reference_id="string_example",
    inputs={},
    source="string_example",
    metadata={},
    save=True,
    source_datapoint_id="string_example",
    provider_api_keys={},
    num_samples=1,
    stream=False,
    user="string_example",
    seed=1,
    return_inputs=True,
    logprobs=1,
    suffix="string_example",
)

⚙️ Parameters

model_config: `ModelConfigCompletionRequest`

The model configuration used to generate.

project: `str`

Unique project name. If no project exists with this name, a new project will be created.

project_id: `str`

Unique ID of a project to associate to the log. Either this or project must be provided.

session_id: `str`

ID of the session to associate the datapoint.

session_reference_id: `str`

parent_id: `str`

ID associated to the parent datapoint in a session.

parent_reference_id: `str`

inputs: `Dict[str, Union[bool, date, datetime, dict, float, int, list, str, None]]`

The inputs passed to the prompt template.

source: `str`

Identifies where the model was called from.

metadata: `Dict[str, Union[bool, date, datetime, dict, float, int, list, str, None]]`

Any additional metadata to record.

save: `bool`

Whether the request/response payloads will be stored on Humanloop.

source_datapoint_id: `str`

ID of the source datapoint if this is a log derived from a datapoint in a dataset.

provider_api_keys: `ProviderApiKeys`

API keys required by each provider to make API calls. The API keys provided here are not stored by Humanloop. If not specified here, Humanloop will fall back to the key saved to your organization.

num_samples: `int`

The number of generations.

stream: `bool`

If true, tokens will be sent as data-only server-sent events. If num_samples > 1, samples are streamed back independently.

user: `str`

End-user ID passed through to provider call.

seed: `int`

Deprecated field: the seed is instead set as part of the request.config object.

return_inputs: `bool`

Whether to return the inputs in the response. If false, the response will contain an empty dictionary under inputs. This is useful for reducing the size of the response. Defaults to true.

logprobs: `int`

Include the log probabilities of the top n tokens in the provider_response

suffix: `str`

The suffix that comes after a completion of inserted text. Useful for completions that act like inserts.

⚙️ Request Body

CompletionRequest

🔄 Return

🌐 Endpoint

/completion post

CompletionDeployedRequest

`humanloop.complete_deployed`

Create a completion using the project's active deployment. The active deployment can be a specific model configuration or an experiment.

🛠️ Usage

create_deployed_response = humanloop.complete_deployed(
    project="string_example",
    project_id="string_example",
    session_id="string_example",
    session_reference_id="string_example",
    parent_id="string_example",
    parent_reference_id="string_example",
    inputs={},
    source="string_example",
    metadata={},
    save=True,
    source_datapoint_id="string_example",
    provider_api_keys={},
    num_samples=1,
    stream=False,
    user="string_example",
    seed=1,
    return_inputs=True,
    logprobs=1,
    suffix="string_example",
    environment="string_example",
)

⚙️ Parameters

project: `str`

Unique project name. If no project exists with this name, a new project will be created.

project_id: `str`

Unique ID of a project to associate to the log. Either this or project must be provided.

session_id: `str`

ID of the session to associate the datapoint.

session_reference_id: `str`

parent_id: `str`

ID associated to the parent datapoint in a session.

parent_reference_id: `str`

inputs: `Dict[str, Union[bool, date, datetime, dict, float, int, list, str, None]]`

The inputs passed to the prompt template.

source: `str`

Identifies where the model was called from.

metadata: `Dict[str, Union[bool, date, datetime, dict, float, int, list, str, None]]`

Any additional metadata to record.

save: `bool`

Whether the request/response payloads will be stored on Humanloop.

source_datapoint_id: `str`

ID of the source datapoint if this is a log derived from a datapoint in a dataset.

provider_api_keys: `ProviderApiKeys`

API keys required by each provider to make API calls. The API keys provided here are not stored by Humanloop. If not specified here, Humanloop will fall back to the key saved to your organization.

num_samples: `int`

The number of generations.

stream: `bool`

If true, tokens will be sent as data-only server-sent events. If num_samples > 1, samples are streamed back independently.

user: `str`

End-user ID passed through to provider call.

seed: `int`

Deprecated field: the seed is instead set as part of the request.config object.

return_inputs: `bool`

Whether to return the inputs in the response. If false, the response will contain an empty dictionary under inputs. This is useful for reducing the size of the response. Defaults to true.

logprobs: `int`

Include the log probabilities of the top n tokens in the provider_response

suffix: `str`

The suffix that comes after a completion of inserted text. Useful for completions that act like inserts.

environment: `str`

The environment name used to create a chat response. If not specified, the default environment will be used.

⚙️ Request Body

🔄 Return

🌐 Endpoint

/completion-deployed post

CompletionExperimentRequest

`humanloop.complete_experiment`

Create a completion for a specific experiment.

🛠️ Usage

create_experiment_response = humanloop.complete_experiment(
    experiment_id="string_example",
    project="string_example",
    project_id="string_example",
    session_id="string_example",
    session_reference_id="string_example",
    parent_id="string_example",
    parent_reference_id="string_example",
    inputs={},
    source="string_example",
    metadata={},
    save=True,
    source_datapoint_id="string_example",
    provider_api_keys={},
    num_samples=1,
    stream=False,
    user="string_example",
    seed=1,
    return_inputs=True,
    logprobs=1,
    suffix="string_example",
)

⚙️ Parameters

experiment_id: `str`

If an experiment ID is provided a model configuration will be sampled from the experiments active model configurations.

project: `str`

Unique project name. If no project exists with this name, a new project will be created.

project_id: `str`

Unique ID of a project to associate to the log. Either this or project must be provided.

session_id: `str`

ID of the session to associate the datapoint.

session_reference_id: `str`

parent_id: `str`

ID associated to the parent datapoint in a session.

parent_reference_id: `str`

inputs: `Dict[str, Union[bool, date, datetime, dict, float, int, list, str, None]]`

The inputs passed to the prompt template.

source: `str`

Identifies where the model was called from.

metadata: `Dict[str, Union[bool, date, datetime, dict, float, int, list, str, None]]`

Any additional metadata to record.

save: `bool`

Whether the request/response payloads will be stored on Humanloop.

source_datapoint_id: `str`

ID of the source datapoint if this is a log derived from a datapoint in a dataset.

provider_api_keys: `ProviderApiKeys`

API keys required by each provider to make API calls. The API keys provided here are not stored by Humanloop. If not specified here, Humanloop will fall back to the key saved to your organization.

num_samples: `int`

The number of chat responses, where each chat response will use a model configuration sampled from the experiment.

stream: `bool`

If true, tokens will be sent as data-only server-sent events. If num_samples > 1, samples are streamed back independently.

user: `str`

End-user ID passed through to provider call.

seed: `int`

Deprecated field: the seed is instead set as part of the request.config object.

return_inputs: `bool`

Whether to return the inputs in the response. If false, the response will contain an empty dictionary under inputs. This is useful for reducing the size of the response. Defaults to true.

logprobs: `int`

Include the log probabilities of the top n tokens in the provider_response

suffix: `str`

The suffix that comes after a completion of inserted text. Useful for completions that act like inserts.

⚙️ Request Body

🔄 Return

🌐 Endpoint

/completion-experiment post

CompletionModelConfigRequest

`humanloop.complete_model_configuration`

Create a completion for a specific model configuration.

🛠️ Usage

create_model_config_response = humanloop.complete_model_configuration(
    model_config_id="string_example",
    project="string_example",
    project_id="string_example",
    session_id="string_example",
    session_reference_id="string_example",
    parent_id="string_example",
    parent_reference_id="string_example",
    inputs={},
    source="string_example",
    metadata={},
    save=True,
    source_datapoint_id="string_example",
    provider_api_keys={},
    num_samples=1,
    stream=False,
    user="string_example",
    seed=1,
    return_inputs=True,
    logprobs=1,
    suffix="string_example",
)

⚙️ Parameters

model_config_id: `str`

Identifies the model configuration used to create a chat response.

project: `str`

Unique project name. If no project exists with this name, a new project will be created.

project_id: `str`

Unique ID of a project to associate to the log. Either this or project must be provided.

session_id: `str`

ID of the session to associate the datapoint.

session_reference_id: `str`

parent_id: `str`

ID associated to the parent datapoint in a session.

parent_reference_id: `str`

inputs: `Dict[str, Union[bool, date, datetime, dict, float, int, list, str, None]]`

The inputs passed to the prompt template.

source: `str`

Identifies where the model was called from.

metadata: `Dict[str, Union[bool, date, datetime, dict, float, int, list, str, None]]`

Any additional metadata to record.

save: `bool`

Whether the request/response payloads will be stored on Humanloop.

source_datapoint_id: `str`

ID of the source datapoint if this is a log derived from a datapoint in a dataset.

provider_api_keys: `ProviderApiKeys`

API keys required by each provider to make API calls. The API keys provided here are not stored by Humanloop. If not specified here, Humanloop will fall back to the key saved to your organization.

num_samples: `int`

The number of generations.

stream: `bool`

If true, tokens will be sent as data-only server-sent events. If num_samples > 1, samples are streamed back independently.

user: `str`

End-user ID passed through to provider call.

seed: `int`

Deprecated field: the seed is instead set as part of the request.config object.

return_inputs: `bool`

Whether to return the inputs in the response. If false, the response will contain an empty dictionary under inputs. This is useful for reducing the size of the response. Defaults to true.

logprobs: `int`

Include the log probabilities of the top n tokens in the provider_response

suffix: `str`

The suffix that comes after a completion of inserted text. Useful for completions that act like inserts.

⚙️ Request Body

🔄 Return

🌐 Endpoint

/completion-model-config post

DatapointsDeleteRequest

`humanloop.datapoints.delete`

Delete a list of datapoints by their IDs.

🛠️ Usage

humanloop.datapoints.delete(
    body=["datapoints_delete_request_example"],
)

⚙️ Request Body

🌐 Endpoint

/datapoints delete

`humanloop.datapoints.get`

Get a datapoint by ID.

🛠️ Usage

get_response = humanloop.datapoints.get(
    id="id_example",
)

⚙️ Parameters

id: `str`

String ID of datapoint. Starts with evtc_.

🔄 Return

DatapointResponse

🌐 Endpoint

/datapoints/{id} get

`humanloop.datapoints.update`

Edit the input, messages and criteria fields of a datapoint. 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.

🛠️ Usage

update_response = humanloop.datapoints.update(
    id="id_example",
    inputs={
        "key": "string_example",
    },
    messages=[
        {
            "role": "string_example",
        }
    ],
    target={
        "key": "string_example",
    },
)

⚙️ Parameters

id: `str`

String ID of datapoint. Starts with evtc_.

inputs: `UpdateDatapointRequestInputs`

messages: List[`ChatMessageWithToolCall`]

The chat messages for this datapoint.

target: `UpdateDatapointRequestTarget`

⚙️ Request Body

UpdateDatapointRequest

🔄 Return

DatapointResponse

🌐 Endpoint

/datapoints/{id} patch

`humanloop.datasets.create`

Create a new dataset for a project.

🛠️ Usage

create_response = humanloop.datasets.create(
    description="string_example",
    name="string_example",
    project_id="project_id_example",
)

⚙️ Parameters

description: `str`

The description of the dataset.

name: `str`

The name of the dataset.

project_id: `str`

⚙️ Request Body

CreateDatasetRequest

🔄 Return

🌐 Endpoint

/projects/{project_id}/datasets post

DatasetsCreateDatapointResponse

`humanloop.datasets.create_datapoint`

Create a new testcase for a testset.

🛠️ Usage

create_datapoint_response = humanloop.datasets.create_datapoint(
    body={
        "log_ids": ["log_ids_example"],
    },
    dataset_id="dataset_id_example",
    log_ids=["string_example"],
    inputs={
        "key": "string_example",
    },
    messages=[
        {
            "role": "string_example",
        }
    ],
    target={
        "key": "string_example",
    },
)

⚙️ Parameters

dataset_id: `str`

String ID of dataset. Starts with evts_.

requestBody: `DatasetsCreateDatapointRequest`

🔄 Return

🌐 Endpoint

/datasets/{dataset_id}/datapoints post

`humanloop.datasets.delete`

Delete a dataset by ID.

🛠️ Usage

delete_response = humanloop.datasets.delete(
    id="id_example",
)

⚙️ Parameters

id: `str`

String ID of dataset. Starts with evts_.

🔄 Return

🌐 Endpoint

/datasets/{id} delete

`humanloop.datasets.get`

Get a single dataset by ID.

🛠️ Usage

get_response = humanloop.datasets.get(
    id="id_example",
)

⚙️ Parameters

id: `str`

String ID of dataset. Starts with evts_.

🔄 Return

🌐 Endpoint

/datasets/{id} get

DatasetsListAllForProjectResponse

`humanloop.datasets.list_all_for_project`

Get all datasets for a project.

🛠️ Usage

list_all_for_project_response = humanloop.datasets.list_all_for_project(
    project_id="project_id_example",
)

⚙️ Parameters

project_id: `str`

🔄 Return

🌐 Endpoint

/projects/{project_id}/datasets get

PaginatedDataDatapointResponse

`humanloop.datasets.list_datapoints`

Get datapoints for a dataset.

🛠️ Usage

list_datapoints_response = humanloop.datasets.list_datapoints(
    dataset_id="dataset_id_example",
    page=0,
    size=50,
)

⚙️ Parameters

dataset_id: `str`

String ID of dataset. Starts with evts_.

page: `int`

size: `int`

🔄 Return

🌐 Endpoint

/datasets/{dataset_id}/datapoints get

`humanloop.datasets.update`

Update a testset by ID.

🛠️ Usage

update_response = humanloop.datasets.update(
    id="id_example",
    description="string_example",
    name="string_example",
)

⚙️ Parameters

id: `str`

String ID of testset. Starts with evts_.

description: `str`

The description of the dataset.

name: `str`

The name of the dataset.

⚙️ Request Body

UpdateDatasetRequest

🔄 Return

🌐 Endpoint

/datasets/{id} patch

`humanloop.evaluations.add_evaluators`

Add evaluators to an existing evaluation run.

🛠️ Usage

add_evaluators_response = humanloop.evaluations.add_evaluators(
    evaluator_ids=["string_example"],
    id="id_example",
)

⚙️ Parameters

evaluator_ids: `AddEvaluatorsRequestEvaluatorIds`

id: `str`

String ID of evaluation run. Starts with ev_.

⚙️ Request Body

AddEvaluatorsRequest

🔄 Return

🌐 Endpoint

/evaluations/{id}/evaluators patch

CreateEvaluationRequest

`humanloop.evaluations.create`

Create an evaluation.

🛠️ Usage

create_response = humanloop.evaluations.create(
    config_id="string_example",
    evaluator_ids=["string_example"],
    dataset_id="string_example",
    project_id="project_id_example",
    provider_api_keys={},
    max_concurrency=5,
    hl_generated=True,
)

⚙️ Parameters

config_id: `str`

ID of the config to evaluate. Starts with config_.

evaluator_ids: `CreateEvaluationRequestEvaluatorIds`

dataset_id: `str`

ID of the dataset to use in this evaluation. Starts with evts_.

project_id: `str`

String ID of project. Starts with pr_.

provider_api_keys: `ProviderApiKeys`

API keys required by each provider to make API calls. The API keys provided here are not stored by Humanloop. If not specified here, Humanloop will fall back to the key saved to your organization. Ensure you provide an API key for the provider for the model config you are evaluating, or have one saved to your organization.

max_concurrency: `int`

The maximum number of concurrent generations to run. A higher value will result in faster completion of the evaluation but may place higher load on your provider rate-limits.

hl_generated: `bool`

Whether the log generations for this evaluation should be performed by Humanloop. If False, the log generations should be submitted by the user via the API.

⚙️ Request Body

🔄 Return

🌐 Endpoint

/projects/{project_id}/evaluations post

`humanloop.evaluations.get`

Get evaluation by ID.

🛠️ Usage

get_response = humanloop.evaluations.get(
    id="id_example",
    evaluator_aggregates=True,
)

⚙️ Parameters

id: `str`

String ID of evaluation run. Starts with ev_.

evaluator_aggregates: `bool`

Whether to include evaluator aggregates in the response.

🔄 Return

🌐 Endpoint

/evaluations/{id} get

PaginatedDataEvaluationResponse

`humanloop.evaluations.list`

Get the evaluations associated with a project. Sorting and filtering are supported through query params for categorical columns and the created_at timestamp. Sorting is supported for the dataset, config, status and evaluator-{evaluator_id} columns. Specify sorting with the sort query param, with values {column}.{ordering}. E.g. ?sort=dataset.asc&sort=status.desc will yield a multi-column sort. First by dataset then by status. Filtering is supported for the id, dataset, config and status columns. Specify filtering with the id_filter, dataset_filter, config_filter and status_filter query params. E.g. ?dataset_filter=my_dataset&dataset_filter=my_other_dataset&status_filter=running will only show rows where the dataset is "my_dataset" or "my_other_dataset", and where the status is "running". An additional date range filter is supported for the created_at column. Use the start_date and end_date query parameters to configure this.

🛠️ Usage

list_response = humanloop.evaluations.list(
    project_id="project_id_example",
    id=["string_example"],
    start_date="1970-01-01",
    end_date="1970-01-01",
    size=50,
    page=0,
)

⚙️ Parameters

project_id: `str`

String ID of project. Starts with pr_.

id: List[`str`]

A list of evaluation run ids to filter on. Starts with ev_.

start_date: `date`

Only return evaluations created after this date.

end_date: `date`

Only return evaluations created before this date.

size: `int`

page: `int`

🔄 Return

🌐 Endpoint

/evaluations get

EvaluationsGetForProjectResponse

`humanloop.evaluations.list_all_for_project`

Get all the evaluations associated with your project. Deprecated: This is a legacy unpaginated endpoint. Use /evaluations instead, with appropriate sorting, filtering and pagination options.

🛠️ Usage

list_all_for_project_response = humanloop.evaluations.list_all_for_project(
    project_id="project_id_example",
    evaluator_aggregates=True,
)

⚙️ Parameters

project_id: `str`

String ID of project. Starts with pr_.

evaluator_aggregates: `bool`

Whether to include evaluator aggregates in the response.

🔄 Return

🌐 Endpoint

/projects/{project_id}/evaluations get

PaginatedDataEvaluationDatapointSnapshotResponse

`humanloop.evaluations.list_datapoints`

Get testcases by evaluation ID.

🛠️ Usage

list_datapoints_response = humanloop.evaluations.list_datapoints(
    id="id_example",
    page=1,
    size=10,
)

⚙️ Parameters

id: `str`

String ID of evaluation. Starts with ev_.

page: `int`

Page to fetch. Starts from 1.

size: `int`

Number of evaluation results to retrieve.

🔄 Return

🌐 Endpoint

/evaluations/{id}/datapoints get

CreateEvaluationLogRequest

`humanloop.evaluations.log`

Log an external generation to an evaluation run for a datapoint. The run must have status 'running'.

🛠️ Usage

log_response = humanloop.evaluations.log(
    datapoint_id="string_example",
    log={
        "save": True,
    },
    evaluation_id="evaluation_id_example",
)

⚙️ Parameters

datapoint_id: `str`

The datapoint for which a log was generated. Must be one of the datapoints in the dataset being evaluated.

log: `LogRequest`

The log generated for the datapoint.

evaluation_id: `str`

ID of the evaluation run. Starts with evrun_.

⚙️ Request Body

🔄 Return

CreateLogResponse

🌐 Endpoint

/evaluations/{evaluation_id}/log post

CreateEvaluationResultLogRequest

`humanloop.evaluations.result`

Log an evaluation result to an evaluation run. The run must have status 'running'. One of result or error must be provided.

🛠️ Usage

result_response = humanloop.evaluations.result(
    log_id="string_example",
    evaluator_id="string_example",
    evaluation_id="evaluation_id_example",
    result=True,
    error="string_example",
)

⚙️ Parameters

log_id: `str`

The log that was evaluated. Must have as its source_datapoint_id one of the datapoints in the dataset being evaluated.

evaluator_id: `str`

ID of the evaluator that evaluated the log. Starts with evfn_. Must be one of the evaluator IDs associated with the evaluation run being logged to.

evaluation_id: `str`

ID of the evaluation run. Starts with evrun_.

result: Union[`bool`, `int`, `Union[int, float]`]

The result value of the evaluation.

error: `str`

An error that occurred during evaluation.

⚙️ Request Body

🔄 Return

EvaluationResultResponse

🌐 Endpoint

/evaluations/{evaluation_id}/result post

UpdateEvaluationStatusRequest

`humanloop.evaluations.update_status`

Update the status of an evaluation run. Can only be used to update the status of an evaluation run that uses external or human evaluators. The evaluation must currently have status 'running' if swithcing to completed, or it must have status 'completed' if switching back to 'running'.

🛠️ Usage

update_status_response = humanloop.evaluations.update_status(
    status="string_example",
    id="id_example",
)

⚙️ Parameters

status: `EvaluationStatus`

The new status of the evaluation.

id: `str`

String ID of evaluation run. Starts with ev_.

⚙️ Request Body

🔄 Return

🌐 Endpoint

/evaluations/{id}/status patch

`humanloop.evaluators.create`

Create an evaluator within your organization.

🛠️ Usage

create_response = humanloop.evaluators.create(
    description="string_example",
    name="a",
    arguments_type="string_example",
    return_type="string_example",
    type="string_example",
    code="string_example",
    model_config={
        "model": "model_example",
        "max_tokens": -1,
        "temperature": 1,
        "top_p": 1,
        "presence_penalty": 0,
        "frequency_penalty": 0,
        "prompt_template": "{{question}}",
    },
)

⚙️ Parameters

description: `str`

The description of the evaluator.

name: `str`

The name of the evaluator.

arguments_type: `EvaluatorArgumentsType`

Whether this evaluator is target-free or target-required.

return_type: `EvaluatorReturnTypeEnum`

The type of the return value of the evaluator.

type: `EvaluatorType`

The type of the evaluator.

code: `str`

The code for the evaluator. This code will be executed in a sandboxed environment.

model_config: `ModelConfigCompletionRequest`

The model configuration used to generate.

⚙️ Request Body

CreateEvaluatorRequest

🔄 Return

EvaluatorResponse

🌐 Endpoint

/evaluators post

`humanloop.evaluators.delete`

Delete an evaluator within your organization.

🛠️ Usage

humanloop.evaluators.delete(
    id="id_example",
)

⚙️ Parameters

id: `str`

🌐 Endpoint

/evaluators/{id} delete

`humanloop.evaluators.get`

Get an evaluator within your organization.

🛠️ Usage

get_response = humanloop.evaluators.get(
    id="id_example",
)

⚙️ Parameters

id: `str`

🔄 Return

EvaluatorResponse

🌐 Endpoint

/evaluators/{id} get

`humanloop.evaluators.list`

Get all evaluators within your organization.

🛠️ Usage

list_response = humanloop.evaluators.list()

🔄 Return

EvaluatorsListResponse

🌐 Endpoint

/evaluators get

`humanloop.evaluators.update`

Update an evaluator within your organization.

🛠️ Usage

update_response = humanloop.evaluators.update(
    id="id_example",
    description="string_example",
    name="string_example",
    arguments_type="string_example",
    return_type="string_example",
    code="string_example",
    model_config={
        "model": "model_example",
        "max_tokens": -1,
        "temperature": 1,
        "top_p": 1,
        "presence_penalty": 0,
        "frequency_penalty": 0,
        "prompt_template": "{{question}}",
    },
)

⚙️ Parameters

id: `str`

description: `str`

The description of the evaluator.

name: `str`

The name of the evaluator.

arguments_type: `EvaluatorArgumentsType`

Whether this evaluator is target-free or target-required.

return_type: `EvaluatorReturnTypeEnum`

The type of the return value of the evaluator.

code: `str`

The code for the evaluator. This code will be executed in a sandboxed environment.

model_config: `ModelConfigCompletionRequest`

The model configuration used to generate.

⚙️ Request Body

UpdateEvaluatorRequest

🔄 Return

EvaluatorResponse

🌐 Endpoint

/evaluators/{id} patch

CreateExperimentRequest

`humanloop.experiments.create`

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.

🛠️ Usage

create_response = humanloop.experiments.create(
    name="string_example",
    positive_labels=[
        {
            "type": "type_example",
            "value": "value_example",
        }
    ],
    project_id="project_id_example",
    config_ids=["string_example"],
    set_active=False,
)

⚙️ Parameters

name: `str`

Name of experiment.

positive_labels: List[`PositiveLabel`]

Feedback labels to treat as positive user feedback. Used to monitor the performance of model configs in the experiment.

project_id: `str`

String ID of project. Starts with pr_.

config_ids: `CreateExperimentRequestConfigIds`

set_active: `bool`

Whether to set the created project as the project's active experiment.

⚙️ Request Body

🔄 Return

ExperimentResponse

🌐 Endpoint

/projects/{project_id}/experiments post

`humanloop.experiments.delete`

Delete the experiment with the specified ID.

🛠️ Usage

humanloop.experiments.delete(
    experiment_id="experiment_id_example",
)

⚙️ Parameters

experiment_id: `str`

String ID of experiment. Starts with exp_.

🌐 Endpoint

/experiments/{experiment_id} delete

ExperimentsListResponse

`humanloop.experiments.list`

Get an array of experiments associated to your project.

🛠️ Usage

list_response = humanloop.experiments.list(
    project_id="project_id_example",
)

⚙️ Parameters

project_id: `str`

String ID of project. Starts with pr_.

🔄 Return

🌐 Endpoint

/projects/{project_id}/experiments get

`humanloop.experiments.sample`

Samples a model config from the experiment's active model configs.

🛠️ Usage

sample_response = humanloop.experiments.sample(
    experiment_id="experiment_id_example",
)

⚙️ Parameters

experiment_id: `str`

String ID of experiment. Starts with exp_.

🔄 Return

GetModelConfigResponse

🌐 Endpoint

/experiments/{experiment_id}/model-config get

UpdateExperimentRequest

`humanloop.experiments.update`

Update your experiment, including registering and de-registering model configs.

🛠️ Usage

update_response = humanloop.experiments.update(
    experiment_id="experiment_id_example",
    name="string_example",
    positive_labels=[
        {
            "type": "type_example",
            "value": "value_example",
        }
    ],
    config_ids_to_register=["string_example"],
    config_ids_to_deregister=["string_example"],
)

⚙️ Parameters

experiment_id: `str`

String ID of experiment. Starts with exp_.

name: `str`

Name of experiment.

positive_labels: List[`PositiveLabel`]

Feedback labels to treat as positive user feedback. Used to monitor the performance of model configs in the experiment.

config_ids_to_register: `UpdateExperimentRequestConfigIdsToRegister`

config_ids_to_deregister: `UpdateExperimentRequestConfigIdsToDeregister`

⚙️ Request Body

🔄 Return

ExperimentResponse

🌐 Endpoint

/experiments/{experiment_id} patch

`humanloop.feedback`

Submit an array of feedback for existing data_ids

🛠️ Usage

feedback_response = humanloop.feedback(
    body=[
        {
            "type": "string_example",
        }
    ],
    type="string_example",
    value="string_example",
    data_id="string_example",
    user="string_example",
    created_at="1970-01-01T00:00:00.00Z",
    unset=True,
)

⚙️ Parameters

type: Union[`FeedbackType`, `str`]

The type of feedback. The default feedback types available are 'rating', 'action', 'issue', 'correction', and 'comment'.

value: `str`

The feedback value to be set. This field should be left blank when unsetting 'rating', 'correction' or 'comment', but is required otherwise.

data_id: `str`

ID to associate the feedback to a previously logged datapoint.

user: `str`

A unique identifier to who provided the feedback.

created_at: `datetime`

User defined timestamp for when the feedback was created.

unset: `bool`

If true, the value for this feedback is unset.

⚙️ Request Body

FeedbackSubmitRequest

🔄 Return

FeedbackSubmitResponse

🌐 Endpoint

/feedback post

`humanloop.finetunes.create`

Trigger the fine-tuning process for a specific base model and data snapshot.

🛠️ Usage

create_response = humanloop.finetunes.create(
    name="string_example",
    dataset_id="string_example",
    config={
        "base_model": "base_model_example",
    },
    project_id="project_id_example",
    metadata={},
    provider_api_keys={},
)

⚙️ Parameters

name: `str`

User defined friendly name for a finetuning run

dataset_id: `str`

ID of dataset used for finetuning

config: `FinetuneConfig`

Configuration and hyper-parameters for the fine-tuning process

project_id: `str`

metadata: `Dict[str, Union[bool, date, datetime, dict, float, int, list, str, None]]`

Any additional metadata that you would like to log for reference.

provider_api_keys: `ProviderApiKeys`

API keys required by each provider to make API calls. The API keys provided here are not stored by Humanloop. If not specified here, Humanloop will fall back to the key saved to your organization.

⚙️ Request Body

FinetuneRequest

🔄 Return

FinetuneResponse

🌐 Endpoint

/projects/{project_id}/finetunes post

FinetunesListAllForProjectResponse

`humanloop.finetunes.list_all_for_project`

Get a list of all fine-tuned models associated to a given project.

🛠️ Usage

list_all_for_project_response = humanloop.finetunes.list_all_for_project(
    project_id="project_id_example",
)

⚙️ Parameters

project_id: `str`

🔄 Return

🌐 Endpoint

/projects/{project_id}/finetunes get

`humanloop.finetunes.summary`

Checks data for errors and generates finetune data summary. Does not actually trigger the finetuning process or persist the data.

🛠️ Usage

summary_response = humanloop.finetunes.summary(
    name="string_example",
    dataset_id="string_example",
    config={
        "base_model": "base_model_example",
    },
    project_id="project_id_example",
    metadata={},
    provider_api_keys={},
)

⚙️ Parameters

name: `str`

User defined friendly name for a finetuning run

dataset_id: `str`

ID of dataset used for finetuning

config: `FinetuneConfig`

Configuration and hyper-parameters for the fine-tuning process

project_id: `str`

metadata: `Dict[str, Union[bool, date, datetime, dict, float, int, list, str, None]]`

Any additional metadata that you would like to log for reference.

provider_api_keys: `ProviderApiKeys`

API keys required by each provider to make API calls. The API keys provided here are not stored by Humanloop. If not specified here, Humanloop will fall back to the key saved to your organization.

⚙️ Request Body

FinetuneRequest

🔄 Return

FinetuneDataSummary

🌐 Endpoint

/projects/{project_id}/finetunes/summary post

`humanloop.finetunes.update`

Update user-specified attributes of the specified finetuned models.

🛠️ Usage

update_response = humanloop.finetunes.update(
    id="id_example",
    project_id="project_id_example",
    name="string_example",
)

⚙️ Parameters

id: `str`

project_id: `str`

name: `str`

⚙️ Request Body

UpdateFinetuneRequest

🔄 Return

FinetuneResponse

🌐 Endpoint

/finetunes/{id} patch

`humanloop.logs.delete`

Delete

🛠️ Usage

humanloop.logs.delete(
    id=["string_example"],
)

⚙️ Parameters

id: List[`str`]

🌐 Endpoint

/logs delete

`humanloop.logs.get`

Retrieve a log by log id.

🛠️ Usage

get_response = humanloop.logs.get(
    id="id_example",
)

⚙️ Parameters

id: `str`

String ID of log to return. Starts with data_.

🔄 Return

LogResponse

🌐 Endpoint

/logs/{id} get

PaginatedDataLogResponse

`humanloop.logs.list`

Retrieve paginated logs from the server. Sorting and filtering are supported through query params. Sorting is supported for the source, model, timestamp, and feedback-{output_name} columns. Specify sorting with the sort query param, with values {column}.{ordering}. E.g. ?sort=source.asc&sort=model.desc will yield a multi-column sort. First by source then by model. Filtering is supported for the source, model, feedback-{output_name}, evaluator-{evaluator_external_id} columns. Specify filtering with the source_filter, model_filter, feedback-{output.name}_filter and evaluator-{evaluator_external_id}_filter query params. E.g. ?source_filter=AI&source_filter=user_1234&feedback-explicit_filter=good will only show rows where the source is "AI" or "user_1234", and where the latest feedback for the "explicit" output group is "good". An additional date range filter is supported for the Timestamp column (i.e. Log.created_at). These are supported through the start_date and end_date query parameters. Searching is supported for the model inputs and output. Specify a search term with the search query param. E.g. ?search=hello%20there will cause a case-insensitive search across model inputs and output.

🛠️ Usage

list_response = humanloop.logs.list(
    project_id="project_id_example",
    search="string_example",
    metadata_search="string_example",
    start_date="1970-01-01",
    end_date="1970-01-01",
    size=50,
    page=0,
)

⚙️ Parameters

project_id: `str`

search: `str`

metadata_search: `str`

start_date: `date`

end_date: `date`

size: `int`

page: `int`

🔄 Return

🌐 Endpoint

/logs get

`humanloop.log`

Log a datapoint or array of datapoints to your Humanloop project.

🛠️ Usage

log_response = humanloop.log(
    body=[
        {
            "save": True,
        }
    ],
    project="string_example",
    project_id="string_example",
    session_id="string_example",
    session_reference_id="string_example",
    parent_id="string_example",
    parent_reference_id="string_example",
    inputs={},
    source="string_example",
    metadata={},
    save=True,
    source_datapoint_id="string_example",
    reference_id="string_example",
    trial_id="string_example",
    messages=[
        {
            "role": "string_example",
        }
    ],
    output="string_example",
    config_id="string_example",
    config={
        "model": "model_example",
        "max_tokens": -1,
        "temperature": 1,
        "top_p": 1,
        "presence_penalty": 0,
        "frequency_penalty": 0,
        "type": "ModelConfigRequest",
    },
    environment="string_example",
    feedback={
        "type": "string_example",
        "value": 3.14,
    },
    created_at="1970-01-01T00:00:00.00Z",
    error="string_example",
    duration=3.14,
    output_message={
        "role": "string_example",
    },
    prompt_tokens=1,
    output_tokens=1,
    provider_request={},
    provider_response={},
)

⚙️ Parameters

project: `str`

Unique project name. If no project exists with this name, a new project will be created.

project_id: `str`

Unique ID of a project to associate to the log. Either this or project must be provided.

session_id: `str`

ID of the session to associate the datapoint.

session_reference_id: `str`

parent_id: `str`

ID associated to the parent datapoint in a session.

parent_reference_id: `str`

inputs: `Dict[str, Union[bool, date, datetime, dict, float, int, list, str, None]]`

The inputs passed to the prompt template.

source: `str`

Identifies where the model was called from.

metadata: `Dict[str, Union[bool, date, datetime, dict, float, int, list, str, None]]`

Any additional metadata to record.

save: `bool`

Whether the request/response payloads will be stored on Humanloop.

source_datapoint_id: `str`

ID of the source datapoint if this is a log derived from a datapoint in a dataset.

reference_id: `str`

A unique string to reference the datapoint. Allows you to log nested datapoints with your internal system IDs by passing the same reference ID as parent_id in a subsequent log request.

trial_id: `str`

Unique ID of an experiment trial to associate to the log.

messages: List[`ChatMessageWithToolCall`]

The messages passed to the to provider chat endpoint.

output: `str`

Generated output from your model for the provided inputs. Can be None if logging an error, or if logging a parent datapoint with the intention to populate it later

config_id: `str`

Unique ID of a config to associate to the log.

config: Union[`ModelConfigRequest`, `ToolConfigRequest`]

The model config used for this generation. Required unless config_id or trial_id is provided.

environment: `str`

The environment name used to create the log.

feedback: Union[`Feedback`, List[`Feedback`]]

Optional parameter to provide feedback with your logged datapoint.

created_at: `datetime`

User defined timestamp for when the log was created.

error: `str`

Error message if the log is an error.

duration: `Union[int, float]`

Duration of the logged event in seconds.

output_message: `ChatMessageWithToolCall`

The message returned by the provider.

prompt_tokens: `int`

Number of tokens in the prompt used to generate the output.

output_tokens: `int`

Number of tokens in the output generated by the model.

provider_request: `Dict[str, Union[bool, date, datetime, dict, float, int, list, str, None]]`

Raw request sent to provider.

provider_response: `Dict[str, Union[bool, date, datetime, dict, float, int, list, str, None]]`

Raw response received the provider.

⚙️ Request Body

LogDatapointRequest

🔄 Return

LogsLogResponse

🌐 Endpoint

/logs post

`humanloop.logs.update`

Update a logged datapoint in your Humanloop project.

🛠️ Usage

update_response = humanloop.logs.update(
    id="id_example",
    output="string_example",
    error="string_example",
    duration=3.14,
)

⚙️ Parameters

id: `str`

String ID of logged datapoint to return. Starts with data_.

output: `str`

Generated output from your model for the provided inputs.

error: `str`

Error message if the log is an error.

duration: `Union[int, float]`

Duration of the logged event in seconds.

⚙️ Request Body

UpdateLogRequest

🔄 Return

LogResponse

🌐 Endpoint

/logs/{id} patch

`humanloop.logs.update_by_ref`

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.

🛠️ Usage

update_by_ref_response = humanloop.logs.update_by_ref(
    reference_id="reference_id_example",
    output="string_example",
    error="string_example",
    duration=3.14,
)

⚙️ Parameters

reference_id: `str`

A unique string to reference the datapoint. Identifies the logged datapoint created with the same reference_id.

output: `str`

Generated output from your model for the provided inputs.

error: `str`

Error message if the log is an error.

duration: `Union[int, float]`

Duration of the logged event in seconds.

⚙️ Request Body

UpdateLogRequest

🔄 Return

LogResponse

🌐 Endpoint

/logs patch

BodyModelConfigsDeserialize

`humanloop.model_configs.deserialize`

Deserialize a model config from a .prompt file format.

🛠️ Usage

deserialize_response = humanloop.model_configs.deserialize(
    config="string_example",
)

⚙️ Parameters

config: `str`

⚙️ Request Body

🔄 Return

ModelConfigResponse

🌐 Endpoint

/model-configs/deserialize post

`humanloop.model_configs.export`

Export a model config to a .prompt file by ID.

🛠️ Usage

export_response = humanloop.model_configs.export(
    id="id_example",
)

⚙️ Parameters

id: `str`

String ID of the model config. Starts with config_.

🌐 Endpoint

/model-configs/{id}/export post

`humanloop.model_configs.get`

Get a specific model config by ID.

🛠️ Usage

get_response = humanloop.model_configs.get(
    id="id_example",
)

⚙️ Parameters

id: `str`

String ID of the model config. Starts with config_.

🔄 Return

ModelConfigResponse

🌐 Endpoint

/model-configs/{id} get

ProjectModelConfigRequest

`humanloop.model_configs.register`

Register a model config to a project and optionally add it to an experiment. If the project name 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.

🛠️ Usage

register_response = humanloop.model_configs.register(
    model="string_example",
    description="string_example",
    name="string_example",
    provider="string_example",
    max_tokens=-1,
    temperature=1,
    top_p=1,
    stop="string_example",
    presence_penalty=0,
    frequency_penalty=0,
    other={},
    seed=1,
    response_format={
        "type": "json_object",
    },
    project="string_example",
    project_id="string_example",
    experiment="string_example",
    prompt_template="string_example",
    chat_template=[
        {
            "role": "string_example",
        }
    ],
    endpoint="string_example",
    tools=[
        {
            "id": "id_example",
            "source": "organization",
        }
    ],
)

⚙️ Parameters

model: `str`

The model instance used. E.g. text-davinci-002.

description: `str`

A description of the model config.

name: `str`

A friendly display name for the model config. If not provided, a name will be generated.

provider: `ModelProviders`

The company providing the underlying model service.

max_tokens: `int`

The maximum number of tokens to generate. Provide max_tokens=-1 to dynamically calculate the maximum number of tokens to generate given the length of the prompt

temperature: `Union[int, float]`

What sampling temperature to use when making a generation. Higher values means the model will be more creative.

top_p: `Union[int, float]`

An alternative to sampling with temperature, called nucleus sampling, where the model considers the results of the tokens with top_p probability mass.

stop: Union[`str`, `List[str]`]

The string (or list of strings) after which the model will stop generating. The returned text will not contain the stop sequence.

presence_penalty: `Union[int, float]`

Number between -2.0 and 2.0. Positive values penalize new tokens based on whether they appear in the generation so far.

frequency_penalty: `Union[int, float]`

Number between -2.0 and 2.0. Positive values penalize new tokens based on how frequently they appear in the generation so far.

other: `Dict[str, Union[bool, date, datetime, dict, float, int, list, str, None]]`

Other parameter values to be passed to the provider call.

seed: `int`

If specified, model will make a best effort to sample deterministically, but it is not guaranteed.

response_format: `ResponseFormat`

The format of the response. Only type json_object is currently supported for chat.

project: `str`

Unique project name. If it does not exist, a new project will be created.

project_id: `str`

Unique project ID

experiment: `str`

If specified, the model config will be added to this experiment. Experiments are used for A/B testing and optimizing hyperparameters.

prompt_template: `str`

Prompt template that will take your specified inputs to form your final request to the provider model. NB: Input variables within the prompt template should be specified with syntax: {{INPUT_NAME}}.

chat_template: List[`ChatMessageWithToolCall`]

Messages prepended to the list of messages sent to the provider. These messages that will take your specified inputs to form your final request to the provider model. NB: Input variables within the prompt template should be specified with syntax: {{INPUT_NAME}}.

endpoint: `ModelEndpoints`

Which of the providers model endpoints to use. For example Complete or Edit.

tools: `ProjectModelConfigRequestTools`

⚙️ Request Body

🔄 Return

ProjectConfigResponse

🌐 Endpoint

/model-configs post

ModelConfigsSerializeRequest

`humanloop.model_configs.serialize`

Serialize a model config to a .prompt file format.

🛠️ Usage

serialize_response = humanloop.model_configs.serialize(
    body={
        "model": "model_example",
        "max_tokens": -1,
        "temperature": 1,
        "top_p": 1,
        "presence_penalty": 0,
        "frequency_penalty": 0,
    },
    description="string_example",
    name="string_example",
    provider="string_example",
    model="string_example",
    max_tokens=-1,
    temperature=1,
    top_p=1,
    stop="string_example",
    presence_penalty=0,
    frequency_penalty=0,
    other={},
    seed=1,
    response_format={
        "type": "json_object",
    },
    endpoint="string_example",
    chat_template=[
        {
            "role": "string_example",
        }
    ],
    tools=[
        {
            "id": "id_example",
            "source": "organization",
        }
    ],
    prompt_template="{{question}}",
)

⚙️ Parameters

description: `str`

A description of the model config.

name: `str`

A friendly display name for the model config. If not provided, a name will be generated.

provider: `ModelProviders`

The company providing the underlying model service.

model: `str`

The model instance used. E.g. text-davinci-002.

max_tokens: `int`

The maximum number of tokens to generate. Provide max_tokens=-1 to dynamically calculate the maximum number of tokens to generate given the length of the prompt

temperature: `Union[int, float]`

What sampling temperature to use when making a generation. Higher values means the model will be more creative.

top_p: `Union[int, float]`

An alternative to sampling with temperature, called nucleus sampling, where the model considers the results of the tokens with top_p probability mass.

stop: Union[`str`, `List[str]`]

The string (or list of strings) after which the model will stop generating. The returned text will not contain the stop sequence.

presence_penalty: `Union[int, float]`

Number between -2.0 and 2.0. Positive values penalize new tokens based on whether they appear in the generation so far.

frequency_penalty: `Union[int, float]`

Number between -2.0 and 2.0. Positive values penalize new tokens based on how frequently they appear in the generation so far.

other: `Dict[str, Union[bool, date, datetime, dict, float, int, list, str, None]]`

Other parameter values to be passed to the provider call.

seed: `int`

If specified, model will make a best effort to sample deterministically, but it is not guaranteed.

response_format: `ResponseFormat`

The format of the response. Only type json_object is currently supported for chat.

endpoint: `ModelEndpoints`

The provider model endpoint used.

chat_template: List[`ChatMessageWithToolCall`]

Messages prepended to the list of messages sent to the provider. These messages that will take your specified inputs to form your final request to the provider model. Input variables within the template should be specified with syntax: {{INPUT_NAME}}.

tools: `ModelConfigChatRequestTools`

prompt_template: `str`

Prompt template that will take your specified inputs to form your final request to the model. Input variables within the prompt template should be specified with syntax: {{INPUT_NAME}}.

⚙️ Request Body

🌐 Endpoint

/model-configs/serialize post

`humanloop.projects.create`

Create a new project.

🛠️ Usage

create_response = humanloop.projects.create(
    name="string_example",
    feedback_types=[
        {
            "type": "type_example",
        }
    ],
    directory_id="string_example",
)

⚙️ Parameters

name: `str`

Unique project name.

feedback_types: List[`FeedbackTypeRequest`]

Feedback types to be created.

directory_id: `str`

ID of directory to assign project to. Starts with dir_. If not provided, the project will be created in the root directory.

⚙️ Request Body

CreateProjectRequest

🔄 Return

🌐 Endpoint

/projects post

`humanloop.projects.create_feedback_type`

Create Feedback Type

🛠️ Usage

create_feedback_type_response = humanloop.projects.create_feedback_type(
    type="string_example",
    id="id_example",
    values=[
        {
            "value": "value_example",
            "sentiment": "positive",
        }
    ],
    _class="string_example",
)

⚙️ Parameters

type: `str`

The type of feedback to update.

id: `str`

String ID of project. Starts with pr_.

values: List[`FeedbackLabelRequest`]

The feedback values to be available. This field should only be populated when updating a 'select' or 'multi_select' feedback class.

_class: `FeedbackClass`

The data type associated to this feedback type; whether it is a 'text'/'select'/'multi_select'. This is optional when updating the default feedback types (i.e. when type is 'rating', 'action' or 'issue').

⚙️ Request Body

FeedbackTypeRequest

🔄 Return

FeedbackTypeModel

🌐 Endpoint

/projects/{id}/feedback-types post

`humanloop.projects.deactivate_config`

Remove the project's active config, if set. This has no effect if the project does not have an active model config set.

🛠️ Usage

deactivate_config_response = humanloop.projects.deactivate_config(
    id="id_example",
    environment="string_example",
)

⚙️ Parameters

id: `str`

String ID of project. Starts with pr_.

environment: `str`

Name for the environment. E.g. 'production'. If not provided, will delete the active config for the default environment.

🔄 Return

🌐 Endpoint

/projects/{id}/active-config delete

`humanloop.projects.deactivate_experiment`

Remove the project's active experiment, if set. This has no effect if the project does not have an active experiment set.

🛠️ Usage

deactivate_experiment_response = humanloop.projects.deactivate_experiment(
    id="id_example",
    environment="string_example",
)

⚙️ Parameters

id: `str`

String ID of project. Starts with pr_.

environment: `str`

Name for the environment. E.g. 'producton'. If not provided, will return the experiment for the default environment.

🔄 Return

🌐 Endpoint

/projects/{id}/active-experiment delete

`humanloop.projects.delete`

Delete a specific file.

🛠️ Usage

humanloop.projects.delete(
    id="id_example",
)

⚙️ Parameters

id: `str`

String ID of project. Starts with pr_.

🌐 Endpoint

/projects/{id} delete

`humanloop.projects.delete_deployed_config`

Remove the verion deployed to environment. This has no effect if the project does not have an active version set.

🛠️ Usage

delete_deployed_config_response = humanloop.projects.delete_deployed_config(
    project_id="project_id_example",
    environment_id="environment_id_example",
)

⚙️ Parameters

project_id: `str`

environment_id: `str`

🌐 Endpoint

/projects/{project_id}/deployed-config/{environment_id} delete

EnvironmentProjectConfigRequest

`humanloop.projects.deploy_config`

Deploy a model config to an environment. If the environment already has a model config deployed, it will be replaced.

🛠️ Usage

deploy_config_response = humanloop.projects.deploy_config(
    project_id="project_id_example",
    config_id="string_example",
    experiment_id="string_example",
    environments=[
        {
            "id": "id_example",
        }
    ],
)

⚙️ Parameters

project_id: `str`

config_id: `str`

Model config unique identifier generated by Humanloop.

experiment_id: `str`

String ID of experiment. Starts with exp_.

environments: List[`EnvironmentRequest`]

List of environments to associate with the model config.

⚙️ Request Body

🔄 Return

ProjectsDeployConfigToEnvironmentsResponse

🌐 Endpoint

/projects/{project_id}/deploy-config patch

PaginatedDataLogResponse

`humanloop.projects.export`

Export all logged datapoints associated to your project. Results are paginated and sorts the datapoints based on created_at in descending order.

🛠️ Usage

export_response = humanloop.projects.export(
    id="id_example",
    page=0,
    size=10,
)

⚙️ Parameters

id: `str`

String ID of project. Starts with pr_.

page: `int`

Page offset for pagination.

size: `int`

Page size for pagination. Number of logs to export.

🔄 Return

🌐 Endpoint

/projects/{id}/export post

`humanloop.projects.get`

Get a specific project.

🛠️ Usage

get_response = humanloop.projects.get(
    id="id_example",
)

⚙️ Parameters

id: `str`

String ID of project. Starts with pr_.

🔄 Return

🌐 Endpoint

/projects/{id} get

`humanloop.projects.get_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.

🛠️ Usage

get_active_config_response = humanloop.projects.get_active_config(
    id="id_example",
    environment="string_example",
)

⚙️ Parameters

id: `str`

String ID of project. Starts with pr_.

environment: `str`

Name for the environment. E.g. 'producton'. If not provided, will return the active config for the default environment.

🔄 Return

GetModelConfigResponse

🌐 Endpoint

/projects/{id}/active-config get

PaginatedDataProjectResponse

`humanloop.projects.list`

Get a paginated list of files.

🛠️ Usage

list_response = humanloop.projects.list(
    page=0,
    size=10,
    filter="string_example",
    user_filter="string_example",
    sort_by="string_example",
    order="string_example",
)

⚙️ Parameters

page: `int`

Page offset for pagination.

size: `int`

Page size for pagination. Number of projects to fetch.

filter: `str`

Case-insensitive filter for project name.

user_filter: `str`

Case-insensitive filter for users in the project. This filter matches against both email address and name of users.

sort_by: `ProjectSortBy`

Field to sort projects by

order: `SortOrder`

Direction to sort by.

🔄 Return

🌐 Endpoint

/projects get

ProjectsGetConfigsResponse

`humanloop.projects.list_configs`

Get an array of versions associated to your file.

🛠️ Usage

list_configs_response = humanloop.projects.list_configs(
    id="id_example",
    evaluation_aggregates=True,
)

⚙️ Parameters

id: `str`

String ID of project. Starts with pr_.

evaluation_aggregates: `bool`

🔄 Return

🌐 Endpoint

/projects/{id}/configs get

ProjectsGetDeployedConfigsResponse

`humanloop.projects.list_deployed_configs`

Get an array of environments with the deployed configs associated to your project.

🛠️ Usage

list_deployed_configs_response = humanloop.projects.list_deployed_configs(
    id="id_example",
)

⚙️ Parameters

id: `str`

String ID of project. Starts with pr_.

🔄 Return

🌐 Endpoint

/projects/{id}/deployed-configs get

`humanloop.projects.update`

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.

🛠️ Usage

update_response = humanloop.projects.update(
    id="id_example",
    name="string_example",
    active_experiment_id="string_example",
    active_config_id="string_example",
    positive_labels=[
        {
            "type": "type_example",
            "value": "value_example",
        }
    ],
    directory_id="string_example",
)

⚙️ Parameters

id: `str`

String ID of project. Starts with pr_.

name: `str`

The new unique project name. Caution, if you are using the project name as the unique identifier in your API calls, changing the name will break the calls.

active_experiment_id: `str`

ID for an experiment to set as the project's active deployment. Starts with 'exp_'. At most one of 'active_experiment_id' and 'active_model_config_id' can be set.

active_config_id: `str`

ID for a config to set as the project's active deployment. Starts with 'config_'. At most one of 'active_experiment_id' and 'active_config_id' can be set.

positive_labels: List[`PositiveLabel`]

The full list of labels to treat as positive user feedback.

directory_id: `str`

ID of directory to assign project to. Starts with dir_.

⚙️ Request Body

UpdateProjectRequest

🔄 Return

🌐 Endpoint

/projects/{id} patch

`humanloop.projects.update_feedback_types`

Update feedback types. Allows enabling the available 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.

🛠️ Usage

update_feedback_types_response = humanloop.projects.update_feedback_types(
    body=[
        {
            "type": "type_example",
        }
    ],
    id="id_example",
)

⚙️ Parameters

id: `str`

String ID of project. Starts with pr_.

requestBody: `ProjectsUpdateFeedbackTypesRequest`

🔄 Return

FeedbackTypes

🌐 Endpoint

/projects/{id}/feedback-types patch

`humanloop.sessions.create`

Create a new session. Returns a session ID that can be used to log datapoints to the session.

🛠️ Usage

create_response = humanloop.sessions.create()

🔄 Return

CreateSessionResponse

🌐 Endpoint

/sessions post

`humanloop.sessions.get`

Get a session by ID.

🛠️ Usage

get_response = humanloop.sessions.get(
    id="id_example",
)

⚙️ Parameters

id: `str`

String ID of session to return. Starts with sesh_.

🔄 Return

SessionResponse

🌐 Endpoint

/sessions/{id} get

PaginatedDataSessionResponse

`humanloop.sessions.list`

Get a page of sessions.

🛠️ Usage

list_response = humanloop.sessions.list(
    project_id="project_id_example",
    page=1,
    size=10,
)

⚙️ Parameters

project_id: `str`

String ID of project to return sessions for. Sessions that contain any datapoints associated to this project will be returned. Starts with pr_.

page: `int`

Page to fetch. Starts from 1.

size: `int`

Number of sessions to retrieve.

🔄 Return

🌐 Endpoint

/sessions get