+def stationary_density(policy, start_i, start_j):
+ probas = policy.new_zeros(policy.size()[:-1])
+ pred_probas = probas.clone()
+ probas[start_i, start_j] = 1.0
+
+ while not pred_probas.equal(probas):
+ pred_probas.copy_(probas)
+ probas.zero_()
+ probas[1:, :] = pred_probas[:-1, :] * policy[0, :-1, :]
+ probas[:-1, :] = pred_probas[1:, :] * policy[1, 1:, :]
+ probas[:, 1:] = pred_probas[:, :-1] * policy[2, :, :-1]
+ probas[:, :-1] = pred_probas[:, 1:] * policy[3, :, 1:]
+ probas[start_i, start_j] = 1.0
+
+