d3rlpy.algos.TransformerActionSampler¶

class d3rlpy.algos.TransformerActionSampler(*args, **kwargs)[source]¶

Interface of TransformerActionSampler.

Methods

__call__(transformer_output)[source]¶

Returns sampled action from Transformer output.

Parameters:: transformer_output (ndarray[tuple[int, ...], dtype[Any]]) – Output of Transformer algorithms.
Returns:: Sampled action.
Return type:: Union[ndarray[tuple[int, …], dtype[Any]], int]