def test_can_get_inconditional_proba_array(
initial_information: MarkovComponents, ) -> None:
ini, trans_matrix = initial_information
mc = MarkovChain(ini, trans_matrix)
pi2 = mc.calc_inconditional_probability(n=2)
expected_vector = [0.372, 0.31, 0.318]
assert pi2 == expected_vector, "Wrong probability vector"
def test_can_calculate_transition_matrix_n_steps(
initial_information: MarkovComponents, n: int,
expectation: TransitionMatrix) -> None:
ini, trans_matrix = initial_information
mc = MarkovChain(ini, trans_matrix)
p_matrix = mc.get_n_steps_probability_matrix(n)
assert p_matrix == expectation, "The expected result was not found."
def test_can_create_markovchain_from_dict() -> None:
data_dict: MarkovRepr = {
"initial_prob": [0.1, 0.2],
"transition_matrix": [
[0.1, 0.9],
[0.8, 0.2],
],
}
MarkovChain.from_dict(data_dict)
def test_p0_matrix_is_identity(initial_information: MarkovComponents) -> None:
ini, trans_matrix = initial_information
mc = MarkovChain(ini, trans_matrix)
p0_matrix = mc.get_n_steps_probability_matrix(n=0)
expected_result = [
[1.0, 0.0, 0.0],
[0.0, 1.0, 0.0],
[0.0, 0.0, 1.0],
]
assert (p0_matrix == expected_result
), "non-identity matrix was returned for p^0 operation"
def test_has_proper_representation(
initial_information: MarkovComponents) -> None:
initial, trans_matrix = initial_information
expected_repr = {"initial": initial, "transition_matrix": trans_matrix}
mc = MarkovChain(initial, trans_matrix)
assert str(expected_repr) == repr(
mc), "An incorrect representation was found"
def test_can_calculate_prob_with_initial_state(
initial_information: MarkovComponents, ) -> None:
ini, trans_matrix = initial_information
mc = MarkovChain(ini, trans_matrix)
assert mc.get_probability_from_sequence([1, 0], initial_state=0) == 0.18
def test_can_calculate_prob_from_sequence(
initial_information: MarkovComponents, ) -> None:
ini, trans_matrix = initial_information
mc = MarkovChain(ini, trans_matrix)
assert mc.get_probability_from_sequence([0, 1, 0, 2, 0]) == 0.00378
def test_can_create_markovchain_from_str() -> None:
input_str = "0.1 0.2\n0.1 0.9\n0.8 0.2"
MarkovChain.from_str(input_str)
def test_raises_exception_when_passing_invalid_matrix(
invalid_trans_matrix: TransitionMatrix, ) -> None:
with pytest.raises(InvalidTransitionMatrix):
MarkovChain(initial_prob=[0.1, 0.2],
transition_matrix=invalid_trans_matrix)