mindspore

XuanCe supports multiple deep learning frameworks for reinforcement learning research and development. Among them, MindSpore serves as one of the supported backends for algorithm implementation.

All APIs introduced in this section are implemented based on MindSpore, ensuring high performance, modularity, and efficient computational graph execution within the MindSpore ecosystem. Users can easily extend or modify these APIs to build custom agents, policies, or learners by leveraging MindSpore’s rich set of neural network components (such as mindspore.nn, mindspore.ops, and mindspore.dataset).

• agents.

• communications.

• learners.

• policies.

• representations.

• runners.

• utils.