a2rl.Simulator.get_valid_actions

Simulator.get_valid_actions(seq, max_size)

Return a dataframe of sampled action tokens, given the input context.

Parameters:

• seq (ndarray) – Input context sequence (1-dim) where context = (s, a, r, …, s) which ends with state dataframe tokens.

• max_size (int) – Number of samples to draw

Return type:

WiDataFrame

Returns:

Whatif dataframe where each row denotes a sample with action columns, and the actions are in the tokenized forms.