Hunter Park

[Question] Multi Output Policy Support?

8

### Question Are multi output policies supported yet? I see that [dictionary observations](https://stable-baselines3.readthedocs.io/en/master/guide/examples.html#dict-observations) are supported per the docs, however I do not see anything out multi output policies... ### Additional...

See https://github.com/Dentosal/python-sc2/issues/69 for discussion on the topic.

The difference of IDs between sequentially created units, regardless of which player created them, is 262144 (for probes) plus/minus 2. This can be used to determine how many units your...

Returns a dict with the following keys ```['single_select', 'multi_select', 'build_queue', 'cargo', 'cargo_slots_available', 'screen', 'minimap','game_loop', 'score_cumulative', 'player', 'control_groups', 'available_actions']``` Information of each of those can be found here : https://github.com/deepmind/pysc2/blob/master/docs/environment.md

### Description When training using the RLModule API , `policy.export` is unable to export to ONNX, citing `ValueError: ONNX export not supported for RLModule API.` ### Use case As a...