User Guides
Python SDK Reference

Asset

Properties

  • dataset_version_id UUID of DatasetVersion where this Asset is

  • data_id UUID of Data of this Asset

  • object_name Object name of this Asset

  • filename Filename of this Asset

  • large If true, this Asset file is considered large

  • type Type of this Asset

  • width Width of this Asset if this is an Image.

  • height Height of this Asset if this is an Image.

  • duration Duration of this Asset if this is a Video.

Methods

reset_url

reset_url()

Description

Reset url of this object

add_tags

add_tags(
   tags: Union[Tag, List[Tag]]
)

Description

Add some tags to an object.
It can be used on Data/MultiData/Asset/MultiAsset/DatasetVersion/Dataset/Model/ModelVersion.

You can give a Tag or a list of Tag.

Examples

tag_bicycle = client.create_tag("bicycle", Target.DATA)
tag_car = client.create_tag("car", Target.DATA)
tag_truck = client.create_tag("truck", Target.DATA)

data.add_tags(tag_bicycle)
data.add_tags([tag_car, tag_truck])

reset_url

reset_url()

Description

Reset url property of this Asset by calling platform.

Returns

A url as str of this Asset.

to_data_schema

to_data_schema()

Description

get_tags

get_tags()

Description

Retrieve the tags of your asset.

Examples

tags = asset.get_tags()
assert tags[0].name == "bicycle"

Returns

List of tags as Tag

get_data

get_data()

Description

Retrieve data of this asset

data = asset.get_data()
assert data.id == asset.data_id
assert data.filename == asset.filename

Returns

A Data object

get_data_tags

get_data_tags()

Description

Retrieve data tag of this asset

tags = asset.get_data_tags()
assert tags[0].name == "bicycle"

Returns

List of tags as Tag

get_annotation

get_annotation(
   worker: Optional[Worker] = None
)

Description

Retrieve the annotation of this asset by the given worker (if none given, it is the current user)

Examples

some_annotation = one_asset.get_annotation(my_worker)
my_annotation = one_asset.get_annotation()
assert some_annotation == my_annotation

Returns

An object Annotation

create_annotation

create_annotation(
   duration: Union[float, int] = 0.0, worker: Optional[Worker] = None
)

Description

Create an annotation on this asset

Examples

some_annotation = one_asset.create_annotation(0.120, my_worker_1)

Returns

An object Annotation

list_annotations

list_annotations()

Description

List all annotation of an asset

Examples

annotations = one_asset.list_annotations()

Returns

A list of Annotation

delete_annotations

delete_annotations(
   workers: Optional[List[Worker]] = None
)

Description

Delete all annotations of an asset: it will erase every shape of every annotation.

You can give workers on which it will be effectively erased.

:warning: DANGER ZONE: Be careful here !

Examples

one_asset.delete_annotations()

delete

delete()

Description

Delete this asset from its dataset

:warning: DANGER ZONE: Be very careful here!

Remove this asset and its annotation from the dataset it belongs

Examples

one_asset.delete()

Properties

Methods

add_tags

add_tags(
   tags: Union[Tag, List[Tag]]
)

Description

Add some tags to an object.
It can be used on Data/MultiData/Asset/MultiAsset/DatasetVersion/Dataset/Model/ModelVersion.

You can give a Tag or a list of Tag.

Examples

tag_bicycle = client.create_tag("bicycle", Target.DATA)
tag_car = client.create_tag("car", Target.DATA)
tag_truck = client.create_tag("truck", Target.DATA)

data.add_tags(tag_bicycle)
data.add_tags([tag_car, tag_truck])

copy

copy()

Description

split

split(
   ratio: float
)

Description

delete

delete()

Description

Delete assets from their dataset

:warning: DANGER ZONE: Be very careful here!

Remove these assets and its annotation from the dataset it belongs

Examples

some_assets = dataset.list_assets()[:10]
some_assets.delete()

download

download(
   target_path: Union[str, Path] = './', force_replace: bool = False,
   max_workers: Optional[int] = None
)

Description

Download this multi asset in given target path

Examples

bunch_of_assets = client.get_dataset("foo_dataset").get_version("first").list_assets()
bunch_of_assets.download('./downloads/')

Arguments

  • target_path (str or Path, optional) : Target path where to download. Defaults to './'.

  • max_workers (int, optional) : Number of max workers used to download. Defaults to os.cpu_count() + 4.

  • force_replace : (bool, optional): Replace an existing file if exists. Defaults to False.

delete_annotations

delete_annotations(
   workers: Optional[List[Worker]] = None
)

Description

Delete all annotations of all these assets: it will erase every shape of every annotation of every asset.

You can give workers on which it will be effectively erased.

:warning: DANGER ZONE: Be careful here !

Examples

multiple_assets.delete_annotations(workers=)

as_list_of_data

as_list_of_data()

Description

Convert a MultiAsset into a List of Data. Assets can come from different Datalake.
This is slower than calling .as_multidata(), so if you know that all your data are coming
from the same datalake, you should call .as_multidata instead

Returns

a list of Data object

as_multidata

as_multidata(
   force_refresh: bool = True
)

Description

Convert a MultiAsset into a MultiData.
Assets must all be in the same datalake to be retrieved.
In case of a dataset with multiple datalake source, you can use .as_list_of_data() that will return a list of data but slower

Args

  • force_refresh (bool, optional) : when False, will not refresh data by calling platform.

Returns

a MultiData object