Obtaining the Model Object List

Sample Code

In ModelArts notebook, you do not need to enter authentication parameters for session authentication. For details about session authentication of other development environments, see Session Authentication.

• Scenario 1: Obtain all model objects of a user.

  from modelarts.session import Session
  from modelarts.model import Model
  
  session = Session()
  model_object_list = Model.get_model_object_list(session)
  

• Scenario 2: Obtain a model object based on the search criteria.

  from modelarts.session import Session
  from modelarts.model import Model
  
  session = Session()
  model_object_list = Model.get_model_object_list(session, model_status="published", model_name="digit", order="desc")
  

Parameters

• After you obtain a model list, the size of the list is equal to the number of models that have been deployed by the current user. Each element in the list is a model object. The object attributes are the same as those in Querying the Details About a Model. For example, in model_list = [model_instance1, model_instance2, model_instance3 ...], each model_instance in the list is a model API that can be called.

• The model list can be obtained based on the query parameters. Table 1 describes the query parameters.

• When you obtain a model list, details about the models are returned. See Table 2 and Table 3.

• A maximum of 150 model objects can be obtained.

Table 1 Query parameters

Parameter	Mandatory	Type	Description
model_name	No	String	Model name. Fuzzy match is supported.
model_version	No	String	Model version
model_status	No	String	Model status. The value can be publishing, published, or failed. You can obtain models based on their statuses.
description	No	String	Description. Fuzzy match is supported.
offset	No	Integer	Index of the page to be obtained. Default value: 0
limit	No	Integer	Maximum number of records returned on each page. Default value: 280
sort_by	No	String	Sorting mode. The value can be create_at, model_version, or model_size. Default value: create_at
order	No	String	Sorting order. The value can be asc or desc, indicating the ascending or descending order. Default value: desc
workspace_id	No	String	Workspace ID. Default value: 0

Table 2 get_model_list parameters

Parameter	Type	Description
total_count	Integer	Total number of models that meet the search criteria when no paging is implemented
count	Integer	Number of models
models	model array	Model metadata

Table 3 model parameters

Parameter	Type	Description
model_id	String	Model ID
model_name	String	Model name
model_version	String	Model version
model_type	String	Model type. The value can be TensorFlow, MXNet, Spark_MLlib, Scikit_Learn, XGBoost, MindSpore, Image, or PyTorch.
model_size	Long	Model size, in bytes
tenant	String	Tenant to whom a model belongs
project	String	Project to which a model belongs
owner	String	User to which a model belongs
create_at	Long	Time when a model is created, in milliseconds calculated from 1970.1.1 0:0:0 UTC
description	String	Model description
source_type	String	Model source type. This parameter is valid only when the model is deployed by an ExeML project. The value is auto.

last updated: 2024-12-11 13:55 UTC - commit: afb4a7dfa15e256dc933e946d9e874db041cb35c