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
Prerequisites The training job has been successfully executed, and the model has been stored in the OBS directory where the training output is stored (the input parameter is train_url).
If the output path is an OBS directory, set the path to a value starting with obs://. If labels are changed for label data, perform operations in If Labels Are Changed before running mox.run.
Upload your training data to OBS if it does not need further processing. To create a training job, enter the OBS bucket path directly as the input parameter path. Import your unlabeled or unpreprocessed training dataset to ModelArts data management for processing.
Code Directory Select the code directory required for this training job, for example, obs://test-modelarts/ascend/code/ in this case. Boot Command Python boot command of the image, for example, bash ${MA_JOB_DIR}/code/run_torch_ddp_npu.sh in this case.
Algorithm Type: Custom algorithm Boot Mode: Custom image Image: swr.cn-north-4.myhuaweicloud.com/deep-learning/pytorch:2.1.0-cann7.0.0 Code Directory: directory where the boot script file is stored in OBS, for example, obs://test-modelarts/pytorch/demo-code/.
According to model building logs, "Not only a Dockerfile in your OBS path, please make sure, The dockerfile list" is displayed, indicating that the file directory is incorrect and that the file should be removed from the directory.
The value can be obs or dataset. obs and dataset cannot be used at the same time. data_url No String OBS bucket path. This parameter cannot be used together with dataset_id or dataset_version.
path of the training job output file log_url String OBS URL of the logs of a training job.
OBS: The parallel file system name is in the format of obs://< bucket-name>/[/topic/body/section/table/tgroup/tbody/row/entry/ul/li/p/b/br {""}) (br]/. Log in to the OBS console. In the parallel file system list, the file system name is the bucket name.
Data and its labeling information are still stored in the OBS directory. However, this affects version management. Exercise caution when performing this operation. Parent topic: Publishing Data
Name: name of the new dataset Storage Path: input path of the new dataset, that is, the OBS path where the data to be exported is stored Output Path: output path of the new dataset, that is, the output path after labeling is complete The output path cannot be the same as the storage
Figure 1 Obtaining images OBS Object Storage Service (OBS) is a cloud storage service optimized for storing massive amounts of data. It provides unlimited, secure, and highly reliable storage capabilities at a relatively low cost. ModelArts exchanges data with OBS.
When ModelArts is used for AI development, data is stored in OBS, EVS, or SFS. In this case, storage fees are generated based on the OBS, EVS, and SFS billing standards.
You can call SDKs on a notebook instance to perform operations such as OBS management, job management, model management, and service management by referring to the SDK reference.
How Do I Improve Training Efficiency While Reducing Interaction with OBS in ModelArts? How Do I Define Path Variables When Using MoXing to Copy Data in ModelArts? How Do I Create a Training Job That References a Third-Party Dependency Package in ModelArts?
By default, this parameter is left blank. train_url Yes String OBS path to the visualization file. The visualization file is provided for the visualization job to read and display, and is usually located in the training output path.
Code Directory: Select the OBS path where the sshd boot script file is stored.
NOTE: Creating an auto labeling job is free, but you will be billed for OBS storage based on usage. For details, see Product Pricing Details. To avoid wasting resources, clear your OBS bucket after labeling jobs and their subsequent tasks are complete.
succeeds. job_name String Name of a visualization job service_url String Endpoint of a visualization job is_success Boolean Whether the API call succeeds duration Long Running duration of a visualization job create_time Long Time when a visualization job is created train_url String OBS
s3://dgg-test-user/snt9-test-cases/mindspore/lenet/ Type 2: time="xxx" level="xxx" msg="xxx" file="xxx" Command=xxx Component=xxx Platform=xxx time="2021-07-26T19:24:11+08:00" level=info msg="start the periodic upload task, upload period = 5 seconds " file="upload.go:46" Command=obs
