NevarokML: ANevarokMLTrainer

The ANevarokMLTrainer class represents a trainer actor in the NevarokML plugin. It is responsible for coordinating the training process and interacting with the environments and algorithms.

---

Overview

The ANevarokMLTrainer class serves as the central component for training machine learning models using the NevarokML plugin. It manages the interaction between the environments, algorithms, and the backend process. The trainer facilitates communication with the environments to collect observations, receive actions, and handle training episodes.

Key features of the ANevarokMLTrainer class:

• Manages the training process, including environment initialization, stepping, and resetting.
• Interacts with the backend process for training and inference.
• Communicates with the environments to collect observations and send actions.
• Handles events such as initialization, step updates, reset notifications, completion, and errors.
• Provides hooks for customization and event handling in Blueprint scripts.

API Reference

For detailed information on the API of the ANevarokMLTrainer class, refer to the following documentation pages:

• Properties
• Methods
• Override Methods
• Event Methods

Conclusion

The ANevarokMLTrainer class plays a vital role in coordinating the training process in the NevarokML plugin. It manages the communication between environments, the backend process, and algorithms. By leveraging the customizable event methods and properties, you can tailor the behavior of the trainer to suit your specific training requirements.

---