Function Recurrent

Compute a recurrent neural net. Roughly, `Recurrent()` computes the following:: state = state0 for t in inputs' sequence length: state, extras = cell_fn(theta, state, inputs[t, :]) accumulate_state[t, :] = state return accumulate_state, state `theta`, `state`

(theta,
              state0,
              inputs,
              cell_fn,
              cell_grad=None,
              cell_type=None,
              stop_fn=None,
              extras=None,
              check_stateful_ops=False,
              accumulator_layer=None,
              allow_implicit_capture=False,
              allowed_tensor_captures=None,
              backward_cleanup=None,
              return_acc_extras=False)