Skip to content
Open
Changes from all commits
Commits
File filter

Filter by extension

Filter by extension

Conversations
Failed to load comments.
Loading
Jump to
Jump to file
Failed to load files.
Loading
Diff view
Diff view
7 changes: 5 additions & 2 deletions docs/NODE_REFERENCE.md
Original file line number Diff line number Diff line change
Expand Up @@ -36,9 +36,8 @@ func _physics_process(delta):
return
```

`needs_reset` is also set to `true` manually (along with `done`) when finishing the episode due to e.g. game over
conditions.
Calling `ai_controller.reset()` will set it to `false` and reset the `AIController`'s internal step counter to 0.
Note that `needs_reset` can be set to `true` manually (along with `done`) when finishing the episode due to e.g. game over conditions.

#### How to disable the time-out reset:

Expand Down Expand Up @@ -115,6 +114,10 @@ Speeds the physics up in the environments to enable faster training.

The path to a trained .onnx model file to use for inference (only needed for the `Onnx Inference` control mode).

#### Send Terminal Obs Info

When enabled, then the terminal observation is sent to the Python server on termination or truncation. By default, truncation is triggered automatically when a predefined number of steps have been exceeded, [see `Reset After`](#reset-after) of the `AIController` nodes. However, enabling this allows the Python server to correctly calculate the bootstrap estimations on truncated episodes by utilizing the terminal observations. Alternatively, a truncation can be triggered manually by setting `truncated` to true and then handling the env reset (e.g. set `needs_reset` to `true` and then handle `needs_reset`).

## Sensors

### RayCastSensor
Expand Down
Loading