Navigation
BlueprintAPI > BlueprintAPI/LearningAgents
Specifies a new enum action. This represents an action which is an exclusive choice from entries of an Enum, sampled from a Categorical distribution.
Target is Learning Agents Actions
Inputs
| Type | Name | Description |
|---|---|---|
| object | Schema | The Action Schema |
| object | Enum | The Enum type. |
| byte | Prior Probabilities | The prior probabilities of each enum element. Can be left empty to use a uniform distribution over elements. Should sum to one. |
| name | Tag | The tag of this new action. Used during action object validation and debugging. |
Outputs
| Type | Name | Description |
|---|---|---|
| struct | Return Value | The newly created action schema element. |