Properties

• origin_name Name of origin of this ModelVersion

• origin_id UUID of the origin Model of this ModelVersion

• name Name of this ModelVersion

• version Version number of this ModelVersion

• type Type of this ModelVersion

• framework Framework of this ModelVersion

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])

get_tags

get_tags()

Description

Retrieve the tags of your model version.

Examples

tags = my_model_version.get_tags()
assert tags[0].name == "my-model-version-1"

Returns

A list of Tag objects

update

update(
   labels: Optional[dict] = None, docker_image_name: Optional[str] = None,
   docker_flags: Optional[list[str]] = None, thumb_object_name: Optional[str] = None,
   notebook_link: Optional[str] = None, base_parameters: Optional[dict] = None,
   docker_env_variables: Optional[dict] = None, framework: Union[str, Framework,
   None] = None, type: Union[str, InferenceType, None] = None,
   name: Optional[str] = None, description: Optional[str] = None,
   docker_tag: Optional[str] = None
)

Description

Update this model version with some new infos.

Examples

model_v1.update(docker_image_name="docker.io/model1")

Arguments

• labels (dict, optional) : Labels of this model version. Defaults to None.

• docker_image_name (str, optional) : Docker image name of this model version. Defaults to None.

• docker_flags (list[str], optional) : Docker flags of this model version. Defaults to None.

• thumb_object_name (str, optional) : Thumbnail object name of this model version. Defaults to None.

• notebook_link (str, optional) : Notebook link of this model version. Defaults to None.

• base_parameters (dict, optional) : Base parameters of this model version. Defaults to None.

• docker_env_variables (dict, optional) : Docker env variables of this model version. Defaults to None.

• framework (Union[str, Framework, None], optional) : Framework of this model version (tensorflow, pytorch, etc.). Defaults to None.

• type (Union[str, InferenceType, None], optional) : Type of this model version (classification, object_detection, segmentation). Defaults to None.

• name (str, optional) : Name of this model version. Defaults to None.

• description (str, optional) : Description of this model version. Defaults to None.

• docker_tag (str, optional) : Docker tag of this model version. Defaults to None.

delete

delete()

Description

Delete model version.

Delete the model in Picsellia database

Examples

model_v1.delete()

get_context

get_context()

Description

Get ModelContext of this model

Examples

model_v1 = client.get_model(name="my-model").get_version(0)
context = model_v1.get_context()
context.get_infos()

Returns

ModelContext objects that you can use and manipulate

list_files

list_files()

Description

Get a list of ModelFile that were stored with this model

Examples

model_v1 = client.get_model(name="my-model").get_version(0)
files = model_v1.list_files()
files[0].download()

Returns

A list of ModelFile that you can use and manipulate

get_file

get_file(
   name: str
)

Description

Retrieve a ModelFile that were stored with this name into this model

Examples

model_v1 = client.get_model(name="my-model").get_version(0)
file = model_v1.get_file("model-latest")
file.download()

Arguments

• name (str) : Name of the file you want to retrieve

Returns

A ModelFile that you can use and manipulate

store

store(
   name: str, path: Union[str, Path], do_zip: bool = False, replace: bool = False
)

Description

Store a file into picsellia storage and attach it to this model.

Examples

model.store("model-latest", "./lg_test_file.pb")

Arguments

• name (str) : Name of file

• path (str or Path) : Path of file to store

• do_zip (bool, optional) : If true, zip directory to store it. Defaults to False.

• replace (bool, optional) : If true, if a file with given name exists, it will be replaced. Defaults to False.

Returns

A ModelFile object

update_thumbnail

update_thumbnail(
   path: Union[str, Path]
)

Description

Updates the model thumbnail.

Update the model thumbnail with given file.
File size shall be less than 5Mb.

Examples

model.update_thumb("test.png")

Arguments

• path (str or Path) : Path of the thumbnail you want to push

Raises

• FileNotFoundException : If there is no file in given path

• InvalidQueryError : If file is too large

• PicselliaError : If an unexpected error occurred while uploading file

deploy

deploy(
   name: Optional[str] = None, target_datalake: Optional[Datalake] = None,
   min_threshold: Optional[float] = None
)

Description

Create a Deployment for a model.

This method allows you to create a Deployment on Picsellia. You will then have
access to the monitoring dashboard and the model management part!

Examples

model_version = client.get_model(name="my-awesome-model").get_version(0)
deployment = model_version.deploy(name="my-awesome-deployment", min_threshold=0.5)

Arguments

• name (str) : Name of your deployment. Defaults to a random name.

• min_threshold (float) : Threshold of detection scores used by models when predicting. Defaults to 0.

• target_datalake Datalake : Datalake to use when data are pushed into Picsellia.
  Defaults to organization default Datalake.

Returns

A Deployment