Functions

featurize_async

Featurize a DataFrame using asynchronous featurizer

Parameters

• df (DataFrame) - The DataFrame to featurize


• input_column (str) - The input column


• output_column (str) - The output column


• featurizer (Any) - The async featurizer function


• concurrency (int) - The number of concurrent jobs to run


• dedup () - Whether to deduplicate the results, defaults to True Default: true


• mini_batch_size () - The mini batch size, defaults to 0 Default: 0

Returns

The DataFrame with the output column

get_cache_base_path

Get the base path for the cache

Returns

Tuple[pyarrow._fs.FileSystem, str]: The filesystem and the path

get_df_schema

Get the Tecton schema of the DataFrame

Parameters

• df (DataFrame) - The DataFrame


• as_attributes (bool) - Whether to return the schema as attributes, defaults to False Default: false

Returns

List[Any]: The schema of the DataFrame

make_request_source

Make a request source

Parameters

• fields () -

Returns

RequestSource: The request source

run_async

Run the coroutine asynchronously in both Tecton and Jupyter notebooks

Parameters

• coro (Any) - The coroutine to run

Returns

Any: The result of the coroutine

run_async_jobs

Run the list of coroutines asynchronously in both Tecton and Jupyter notebooks

Parameters

• jobs (List[Any]) - The list of coroutines to run


• concurrency (int) - The number of concurrent jobs to run

Returns

List[Any]: The results of the coroutines

set_conf

Parameters

• conf (dict) -

set_secrets_env

Set the secrets in the environment variables

Parameters

• secrets_dict (Dict[str,Any]) - A dictionary of secrets

Returns

None

Example

from tecton import Secret

secrets = set_secrets_env({"env_var_name": Secret(scope="", key="")})

@batch_feature_view(
    sources=[your_source, secrets],
    ...
)
def your_feature_view(your_source, secrets):
    # no need to do anything with secrets, they are already set in the environment
    ...