Model Schema

Model Schema are important to train an agent. A model schema defines the architecture of the agent.

The model schema model

The agent model contains all the information about the agent.

Properties

  • Name
    input_size
    Type
    int
    Description

    Number of input features.

  • Name
    hidden_size
    Type
    int
    Description

    Number of hidden units.

  • Name
    num_layers
    Type
    int
    Description

    Number of recurrent layers.

  • Name
    output_size
    Type
    int
    Description

    Number of output features.

Example Model Schema

{
  "input_size": 55,
  "hidden_size": 32,
  "num_layers": 2,
  "output_size": 39,
}