tuneapi.endpoints package

Submodules

tuneapi.endpoints.assistants module

class tuneapi.endpoints.assistants.AssistantsAPI(tune_org_id: str = None, tune_api_key: str = None, base_url: str = 'https://studio.tune.app/')

Bases: object

create(model_name: str, instructions: str, function_model: GenerationConfig, response_model: GenerationConfig = None, user_filters: Dict[str, str] = None, tools: List[Tool] = [])
delete(assistant_id: str)
get(assistant_id: str)
list_assistants(limit: int = 10, order: str = 'desc')
modify(assistant_id: str, *, instructions: str = '', assistant_name: str = '', function_model: GenerationConfig = None, response_model: GenerationConfig = None, user_filters: Dict[str, str] = None, tools: List[Tool] = [])
class tuneapi.endpoints.assistants.GenerationConfig(model_id: str, temperature: float)

Bases: object

model_id: str
temperature: float

tuneapi.endpoints.common module

tuneapi.endpoints.common.get_sub(base_url, tune_org_id: str, tune_api_key: str) → Subway

tuneapi.endpoints.finetune module

Use the Threads API for training models on the Tune AI platform.

class tuneapi.endpoints.finetune.FTDataset(path: str, type: str)

Bases: object

path: str
type: str
class tuneapi.endpoints.finetune.FinetuningAPI(tune_org_id: str = None, tune_api_key: str = None, base_url: str = 'https://studio.tune.app/')

Bases: object

finetune(name: str, datasets: List[FTDataset], base_model: str = 'tune:5fmycsn2', num_epochs: int = 3, lr: float = 0.0001, training_config: dict | None = {}) → TuneModel

Read more about supported models here.

get_job(job_id: str)
upload_dataset(threads: ThreadsList | str, name: str, ds_folder: str = 'tuneds', override: bool = False) → FTDataset

Upload a list of threads to the Tune AI platform to be used for finetuning a model.

Parameters:

  • dataset (-) – A list of threads to be uploaded.

  • filepath (-) – The filepath to save the dataset to for uploading.

upload_dataset_file(filepath: str, name: str)

tuneapi.endpoints.threads module

Use the Threads API for managing threads and messages on the Tune AI platform.

class tuneapi.endpoints.threads.ThreadsAPI(tune_org_id: str = None, tune_api_key: str = None, base_url: str = 'https://studio.tune.app/')

Bases: object

fill_thread_messages(thread: Thread)
get_thread(thread_id: str, messages: bool = False) → Thread
list_threads(limit: int = 20, order: str | None = 'desc', after: str | None = None, before: str | None = None, dataset_id: str | None = None) → List[Thread]
put_thread(thread: Thread) → Thread

Module contents