Common Concepts of ModelArts
ExeML
ExeML is the process of automating model design, parameter tuning, and model training, model compression, and model deployment with the labeled data. The process is code-free and does not require developers to have experience in model development. A model can be built in three steps: labeling data, training a model, and deploying the model.
Device-Edge-Cloud
Device-Edge-Cloud indicates devices, intelligent edge nodes, and the public cloud.
Inference
Inference is a process of deriving a new judgment from a known judgment according to a certain strategy. In AI, machines simulate human intelligence and perform inference based on neural networks.
Real-Time Inference
Real-time inference is a web service that synchronously provides inference results for each inference request.
Batch Inference
Batch inference processes batch data for inference.
Ascend Chip
Ascend chips are Huawei-developed AI chips that offer high computing performance while consuming low power.
Resource pool
ModelArts provides large-scale compute clusters for model development, training, and deployment. Both public resource pool and dedicated resource pool are available for you to select.
ModelArts Standard provides public resource pools by default. ModelArts Standard dedicated resource pools are created separately and used exclusively.
Both ModelArts Lite Server and ModelArts Lite Cluster use dedicated resource pools.
MoXing
A lightweight distributed framework developed by the ModelArts team and built on deep learning engines such as TensorFlow, PyTorch, MXNet, and MindSpore. It improves performance of these engines and makes them easier to use. MoXing contains many components. MoXing Framework is a basic common component that can be used to access OBS. It is decoupled from AI engines and can be used in all ModelArts-supported AI engines such as TensorFlow, PyTorch, MXNet, and MindSpore.
MoXing Framework provides common data file operations in OBS, such as reading, writing, listing, creating folders for, querying, moving, copying, and deleting data files.
You can call MoXing APIs in ModelArts notebook instances without downloading or installing the SDKs. Therefore, MoXing is more convenient than ModelArts and OBS SDKs.
Feedback
Was this page helpful?
Provide feedbackThank you very much for your feedback. We will continue working to improve the documentation.See the reply and handling status in My Cloud VOC.
For any further questions, feel free to contact us through the chatbot.
Chatbot