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
The current version supports modelarts.vm.cpu.2u, modelarts.vm.gpu.pnt004 (must be requested), modelarts.vm.ai1.snt3 (must be requested), and custom (available only when the service is deployed in a dedicated resource pool).
Tokens Per Minute (TPM): The number of tokens (input + output) processed per minute. Requests Per Minute (RPM): The number of requests processed per minute. If the model service has an RPM of 300, it means that up to 10 requests can be processed per second (300/30 = 10).
ModelArts Studio (MaaS) MaaS console UI CN-Hong Kong ModelArts Standard ModelArts console UI All Huawei Cloud regions ModelArts Lite Server ModelArts console Create a Lite Server node through the UI or API.
Using PyTorch to Create a Training Job (New-Version Training) This section describes how to train a model by calling ModelArts APIs.
A model is deployed as a web service on an edge node through Intelligent EdgeFabric (IEF). that provides a real-time test UI and monitoring capabilities. The service keeps running. You need to create a node on IEF beforehand.]
MindSpeed-LLM Throughput (tokens/s/p): Global batch size x sequence length/(Total number of PUs x elapsed time per iteration) x 1000. The global batch size (GBS) and sequence length (SEQ_LEN) are set during training and printed in the logs.
You can set CPU (vCPUs), Memory (GB), Ascend (PU), and Compute Nodes as required. Custom specifications must match or stay below the node specifications of the dedicated resource pool. For CPU specifications, you can only customize the number of vCPUs and memory.
The notebook instances with remote SSH enabled have VS Code plug-ins (such as Python and Jupyter) and the VS Code server package pre-installed, which occupy about 1 GB persistent storage space. Key Pair Set a key pair after remote SSH is enabled.
Action Access Level Resource Type (*: required) Condition Key Alias Dependencies modelarts:service:getMonitor Read service * g:ResourceTag/<tag-key> - - URI GET /v1/{project_id}/services/{service_id}/monitor Table 1 Path Parameters Parameter Mandatory Type Description project_id Yes
Letters, digits, and hyphens (-) are allowed.
Letters, digits, and hyphens (-) are allowed.
Adjust the values of micro-batch-size (MBS, minimum number of samples processed in a batch) and global-batch-size (GBS, number of samples processed in an iteration).
The maximum size of the content to be injected (before encoding) is 32K. Range: N/A Default Value: N/A Response Parameters Status code: 200 Table 3 Response header parameters Parameter Type Description X-Request-Id String Link trace ID.
The maximum size of the content to be injected (before encoding) is 32K. Range: N/A Default Value: N/A Response Parameters Status code: 200 Table 3 Response header parameters Parameter Type Description X-Request-Id String Link trace ID.
ModelArts Studio (MaaS) ModelArts MaaS offers an end-to-end toolchain for foundation model production, along with compute resources and popular open-source models. It is designed for users who need to develop production-ready models using a MaaS platform.
Action Access Level Resource Type (*: required) Condition Key Alias Dependencies modelarts:service:getLogs Read service * g:ResourceTag/<tag-key> - - URI GET /v1/{project_id}/services/{service_id}/logs Table 1 Path Parameters Parameter Mandatory Type Description project_id Yes String
It is recommended that the Linux server have sufficient memory (more than 8 GB) and hard disk (more than 100 GB).
Lowercase letters, digits, hyphens (-), underscores (_), and periods (.) are allowed. namespace String Definition: Organization to which the image belongs. You can create and view an organization on the Organization Management page of the SWR console.
If the model service (server) initiates a disconnection, but the connection is being used by ModelArts (client), a communication error occurs and this error code is returned.
Node Pool Flavor Node Type Common node: A physical or virtual server that provides independent basic compute, storage, and network resources.
