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
Error Message "AttributeError: module '***' has no attribute '***'" Is Displayed Training Job Logs Symptom Error message "AttributeError: module '***' has no attribute '***'" is displayed in the logs of a training job, for example, "AttributeError: module 'torch' has no attribute
[Notice] Huawei Cloud ModelArts Has Discontinued the Old Version of Training Management The end of life (EOL) date for the old version of Huawei Cloud ModelArts training management is June 30, 2023, 00:00 GMT+08:00. Affected Regions All Huawei Cloud regions Impacts After the version
The numba Library Has Been Installed in a Notebook Instance and Error "import numba ModuleNotFoundError: No module named 'numba'" Is Reported Symptom After you install the numba library in a notebook instance by running the !pip install numba command, the library is running properly
Error Message "AttributeError: 'NoneType' object has no attribute 'dtype'" Displayed in Logs Symptom Code can run properly in the notebook Keras image. When tensorflow.keras is used for training, error message "AttributeError: 'NoneType' object has no attribute 'dtype'" is displayed
Size of the Log File Has Reached the Limit Symptom An error occurs during the running of a ModelArts training job, indicating that the size of the log file has reached the limit. modelarts-pope: log length overflow(max:1073741824; already: 107341771; new:90), process will continue
How Do I Reduce the Size of an Image Created Locally or on ECS? Choose a smaller image that fits your needs. For instance, if you are creating a PyTorch 2.1 + CUDA 12.2 image and cannot find an exact match from Huawei Cloud, avoid selecting an image with unrelated components (like
Debugging an SWR Image on an ECS ma-cli allows you to debug an SWR image on an ECS to determine whether to use the image in a ModelArts development environment. Table 1 Parameters Parameter Type Mandatory Description -swr / --swr-path String Yes SWR path to the image to be debugged
Creating a Custom Image on ECS and Using It Application Scenarios and Process You can write a Dockerfile based on a preset base image or third-party image to customize your image on ECS. Then, register the image to create a new development environment based on your needs. This section
Creating a Custom Image on an ECS and Using It in Notebook Application Scenarios and Process Step 1 Preparing a Docker Server and Configuring an Environment Step 2 Creating a Custom Image Step 3 Registering a New Image Step 5 Creating and Starting a Development Environment Parent
Creating a Custom Image on ECS If you want to use an AI engine that is not supported by ModelArts, create a custom image for the engine, import the image to ModelArts, and use the image to create models. This section describes how to use a custom image to create a model and deploy