检测到您已登录华为云国际站账号，为了您更好的体验，建议您访问国际站服务网站 https://www.huaweicloud.com/intl/zh-cn
不再显示此消息
检测到您已登录华为云国际站账号，为了您更好的体验，建议您访问国际站服务网站 https://www.huaweicloud.com/intl/zh-cn
不再显示此消息
Data from OBS is used. dataset: Data from a specified dataset is used. data_url String OBS bucket path Table 4 model_metric_list parameters Parameter Type Description metric JSON Array Validation metrics of a class of a training job total_metric JSON Array All validation metrics
Options: obs: Data from OBS is used. dataset: Data from a specified dataset is used. data_url String OBS bucket path Table 5 volumes parameters Parameter Type Description nfs Object Storage volume of the shared file system type.
Symptom: After the labeled data is uploaded to OBS and synchronized, the data is displayed as unlabeled. Possible causes: Automatic encryption is enabled in the OBS bucket. Solution: Create an OBS bucket and upload data again, or disable bucket encryption and upload data again.
To avoid unsaved code caused by the end of processes, you are advised to periodically save the code to an OBS bucket or the ./work directory of the container. Parent topic: Notebook Instances
Solutions: Write the configuration file and inference code, and save them to the OBS directory where the model to be deployed resides. For details, see Introduction to Model Package Specifications. Figure 1 Error Parent topic: Using PyCharm Toolkit
Call the API for querying the logs of a specified task in a training job (OBS link) to obtain the OBS path of the training job logs.
Use the OBS path to copy datasets. OBS path (recommended) Call the copy_parallel API of MoXing to copy the corresponding OBS path.
To quickly obtain the latest data in the OBS bucket, click Synchronize Data Source on the Unlabeled tab page of the dataset details page to add the data uploaded using OBS to the dataset.
If a model is deployed by a training job or OBS model file, this parameter is left blank. model_type String Model type.
For a model imported from OBS, if the response you received contains an MR error code, for example, MR.0105, view logs on the Logs tab of the real-time service details page to identify the cause.
Training Job Querying the Details About a Training Job Modifying the Description of a Training Job Deleting a Training Job Terminating a Training Job Querying the Logs of a Specified Task in a Given Training Job (Preview) Querying the Logs of a Specified Task in a Training Job (OBS
Code Directory: directory where the boot script file is stored in OBS, for example, obs://test-modelarts/pytorch/demo-code/.
To quickly obtain the latest data in the OBS bucket, click Synchronize Data Source on the Unlabeled tab page of the dataset details page to add the data uploaded using OBS to the dataset.
The options are as follows: OBS_SOURCE: OBS path. (Default value) LOCAL_SOURCE: local path. source_location Yes String Path (parent directory) of the model file If source_location_type is set to OBS_SOURCE, the model file path is an OBS path in the format of /obs_bucketname/...
Options: obs: OBS bucket (default value) dws: GaussDB(DWS) dli: DLI rds: RDS mrs: MRS inference: Inference service import_path Yes String OBS path or manifest path to be imported. When importing a manifest file, ensure that the path is accurate to the manifest file.
If type is set to obs, this parameter is mandatory. The value must be a valid OBS bucket path and end with a slash (/). The value must be a specific directory in an OBS bucket rather than the root directory of an OBS bucket.
Before performing operations on your resources (such as OBS buckets) in a backend job, you are required to explicitly authorize ModelArts through an IAM agency.
You can choose OBS or datasets as the input.
The value can be obs or dataset. data_url No String OBS bucket path. This parameter cannot be used with dataset_id or dataset_version. Table 4 update_job_configs response parameters Parameter Type Description error_msg String Error message when the API call fails.
In the displayed dialog box, select or deselect Delete the source files from OBS as required. After confirmation, click Yes to delete the images. If a tick is displayed in the upper left corner of an image, the image is selected.
