Boot Mode

Preset image (custom script)
To use a custom algorithm, use an image built in ModelArts.

ModelArts offers almost all mainstream AI engines. These built-in engines are pre-loaded with some additional Python packages, such as NumPy. You can also use the requirements.txt file in the code directory to install dependency packages. For details about how to use a custom script to create a training job, see Developing a Custom Script.
Custom image
The built-in training engines can be used in most training scenarios. In certain scenarios, ModelArts allows you to create custom images to train models. Custom images can be used to train models in the cloud only after they are uploaded to the Software Repository for Container (SWR). Customizing an image requires a deep understanding of containers. Use this method only if the subscribed algorithms and custom scripts cannot meet your requirements.

ModelArts provides three algorithm sources for model training.

Algorithm Source	Complexity	Workload
Subscribed algorithm	Simple	Algorithm development is not required.
Custom script	Medium	Algorithm development is required.
Custom image	Difficult	Algorithm development is required and a custom image needs to be created for training.