Skip to content

Asset#

lit.sdk.data.asset #

This module provides a set of classes and functions for managing and interacting with data assets in the LIT platform.

Asset #

The data file prepared and optimized for training a neural network. The result of the data build process.

Examples:

>>> asset = Asset("contoso", "schema_1707673805399")
>>> len(asset)
11648
>>> asset.datasets
['100_1_min', 'per_min', 'rr_atr_2_2_1800']

datasets = train_data.get('datasets') instance-attribute #

A list of feature inputs and labels names contained in the asset.

files property #

Retrieves individual data files from the asset.

Returns:

Type Description
list[PaAssetFileth]

A list asset file objects.

Examples:

>>> asset = Asset("contoso", "schema_1707673805399")
>>> asset.files
[Path('/data/contoso/assets/schema_1707673805399/schema_1707673805399_test.h5'), Path('/data/contoso/assets/schema_1707673805399/schema_1707673805399_train_0.h5')]

name = name instance-attribute #

The name of the schema.

path = Path(lit.config.get_path(self.team, 'assets')) / self.name instance-attribute #

The path to asset artifacts on disk.

team = team instance-attribute #

The name of the team.

__init__(team, name) #

Initializes a new instance of Asset.

Parameters:

Name Type Description Default
team str

The team which owns the Asset.

 required
name str

The name of the Asset.

 required

__len__() #

Total number of unique data samples in the test/train dataset.

promote() #

Promotes newly built assets from the queue to production.

get_asset_names(team) #

Gets all of the training datasets for a given team.

Parameters:

Name Type Description Default
team str

The name of the team.

 required

Returns:

Type Description
list[str]

A list of asset objects.

Examples:

>>> get_assets("contoso")
['schema_1707673805399']

get_assets(team) #

Gets all of the training datasets for a given team.

Parameters:

Name Type Description Default
team str

The name of the team.

 required

Returns:

Type Description
list[Asset]

A list of asset objects.

Examples:

>>> get_assets("contoso")
[Asset(team='contoso', name='schema_1707673805399')]

list_open_assets(team) #

Lists all assets currently in use by experiments.

Parameters:

Name Type Description Default
team str

The name of the team.

 required

Returns:

Type Description
list[Path]

A list of paths.

Examples:

>>> list_open_assets("contoso")
[Path('/data/contoso/assets/schema_1707673805399/schema_1707673805399_test.h5'), Path('/data/contoso/assets/schema_1707673805399/schema_1707673805399_train_0.h5')]