def test_ctc_add_blanks():
BATCHES = 3
N_LABELS = 3
y = T.lmatrix('phonemes')
y_mask = T.matrix('phonemes_mask')
blanked_y, blanked_y_mask = ctc_cost._add_blanks(
y=y,
blank_symbol=1,
y_mask=y_mask)
Y = np.zeros((N_LABELS, BATCHES), dtype='int64')
Y_mask = np.asarray(np.ones_like(Y), dtype=floatX)
Y_mask[-1, 0] = 0
Blanked_y_mask = blanked_y_mask.eval({y_mask: Y_mask})
Blanked_y = blanked_y.eval({y: Y})
assert (Blanked_y == np.array([[1, 1, 1],
[0, 0, 0],
[1, 1, 1],
[0, 0, 0],
[1, 1, 1],
[0, 0, 0],
[1, 1, 1]], dtype='int32')).all()
assert (Blanked_y_mask == np.array([[1., 1., 1.],
[1., 1., 1.],
[1., 1., 1.],
[1., 1., 1.],
[1., 1., 1.],
[0., 1., 1.],
[0., 1., 1.]], dtype=floatX)).all()
def test_ctc_log_path_probabs():
LENGTH = 10
BATCHES = 3
CLASSES = 2
N_LABELS = 3
y_hat = T.tensor3('features')
input_mask = T.matrix('features_mask')
y_hat_mask = input_mask
y = T.lmatrix('phonemes')
y_mask = T.matrix('phonemes_mask')
blanked_y, blanked_y_mask = ctc_cost._add_blanks(
y=y,
blank_symbol=1,
y_mask=y_mask)
p = ctc_cost._log_path_probabs(blanked_y, y_hat, blanked_y_mask, y_hat_mask, 1)
Y_hat = np.zeros((LENGTH, BATCHES, CLASSES + 1), dtype=floatX)
Y_hat[:, :, 0] = .7
Y_hat[:, :, 1] = .2
Y_hat[:, :, 2] = .1
Y = np.zeros((N_LABELS, BATCHES), dtype='int64')
Y_hat_mask = np.ones((LENGTH, BATCHES), dtype=floatX)
Y_hat_mask[-2:, 0] = 0
Y_mask = np.asarray(np.ones_like(Y), dtype=floatX)
forward_probs = p.eval({y_hat: Y_hat, y: Y,
y_hat_mask: Y_hat_mask, y_mask: Y_mask})
assert forward_probs[-2, 0, 0] == -np.inf
Y_mask[-1] = 0
forward_probs_y_mask = p.eval({y_hat: Y_hat, y: Y,
y_hat_mask: Y_hat_mask, y_mask: Y_mask})
assert forward_probs_y_mask[-1, 1, -2] == -np.inf
assert not np.isnan(forward_probs).any()
def test_ctc_log_forward_backward():
LENGTH = 8
BATCHES = 4
CLASSES = 2
N_LABELS = 3
y_hat = T.tensor3('features')
input_mask = T.matrix('features_mask')
y_hat_mask = input_mask
y = T.lmatrix('phonemes')
y_mask = T.matrix('phonemes_mask')
blanked_y, blanked_y_mask = ctc_cost._add_blanks(y=y,
blank_symbol=1,
y_mask=y_mask)
f, b = ctc_cost._log_forward_backward(blanked_y, y_hat, blanked_y_mask,
y_hat_mask, CLASSES)
Y_hat = np.zeros((LENGTH, BATCHES, CLASSES + 1), dtype=floatX)
Y_hat[:, :, 0] = .7
Y_hat[:, :, 1] = .2
Y_hat[:, :, 2] = .1
Y_hat[3, :, 0] = .3
Y_hat[3, :, 1] = .4
Y_hat[3, :, 2] = .3
Y = np.zeros((N_LABELS, BATCHES), dtype='int64')
Y_hat_mask = np.ones((LENGTH, BATCHES), dtype=floatX)
Y_hat_mask[-2:] = 0
Y_mask = np.asarray(np.ones_like(Y), dtype=floatX)
Y_mask[-2:, 0] = 0
y_prob = ctc_cost._class_batch_to_labeling_batch(blanked_y, y_hat,
y_hat_mask)
forward_probs = f.eval({
y_hat: Y_hat,
y: Y,
y_hat_mask: Y_hat_mask,
y_mask: Y_mask
})
backward_probs = b.eval({
y_hat: Y_hat,
y: Y,
y_hat_mask: Y_hat_mask,
y_mask: Y_mask
})
y_probs = y_prob.eval({y_hat: Y_hat, y: Y, y_hat_mask: Y_hat_mask})
assert not ((forward_probs + backward_probs)[:, 0, :] == -np.inf).all()
marg = forward_probs + backward_probs - np.log(y_probs)
forward_probs = np.exp(forward_probs)
backward_probs = np.exp(backward_probs)
L = (forward_probs * backward_probs[::-1][:, :, ::-1] / y_probs).sum(2)
assert not np.isnan(forward_probs).any()
def test_ctc_log_forward_backward():
LENGTH = 8
BATCHES = 4
CLASSES = 2
N_LABELS = 3
y_hat = T.tensor3('features')
input_mask = T.matrix('features_mask')
y_hat_mask = input_mask
y = T.lmatrix('phonemes')
y_mask = T.matrix('phonemes_mask')
blanked_y, blanked_y_mask = ctc_cost._add_blanks(
y=y,
blank_symbol=1,
y_mask=y_mask)
f, b = ctc_cost._log_forward_backward(blanked_y, y_hat,
blanked_y_mask, y_hat_mask, CLASSES)
Y_hat = np.zeros((LENGTH, BATCHES, CLASSES + 1), dtype=floatX)
Y_hat[:, :, 0] = .7
Y_hat[:, :, 1] = .2
Y_hat[:, :, 2] = .1
Y_hat[3, :, 0] = .3
Y_hat[3, :, 1] = .4
Y_hat[3, :, 2] = .3
Y = np.zeros((N_LABELS, BATCHES), dtype='int64')
Y_hat_mask = np.ones((LENGTH, BATCHES), dtype=floatX)
Y_hat_mask[-2:] = 0
Y_mask = np.asarray(np.ones_like(Y), dtype=floatX)
Y_mask[-2:, 0] = 0
y_prob = ctc_cost._class_batch_to_labeling_batch(blanked_y,
y_hat,
y_hat_mask)
forward_probs = f.eval({y_hat: Y_hat, y: Y,
y_hat_mask: Y_hat_mask, y_mask: Y_mask})
backward_probs = b.eval({y_hat: Y_hat, y: Y,
y_hat_mask: Y_hat_mask, y_mask: Y_mask})
y_probs = y_prob.eval({y_hat: Y_hat, y: Y, y_hat_mask: Y_hat_mask})
assert not ((forward_probs + backward_probs)[:, 0, :] == -np.inf).all()
marg = forward_probs + backward_probs - np.log(y_probs)
forward_probs = np.exp(forward_probs)
backward_probs = np.exp(backward_probs)
L = (forward_probs * backward_probs[::-1][:, :, ::-1] / y_probs).sum(2)
assert not np.isnan(forward_probs).any()
def test_ctc_add_blanks():
BATCHES = 3
N_LABELS = 3
y = T.lmatrix('phonemes')
y_mask = T.matrix('phonemes_mask')
blanked_y, blanked_y_mask = ctc_cost._add_blanks(y=y,
blank_symbol=1,
y_mask=y_mask)
Y = np.zeros((N_LABELS, BATCHES), dtype='int64')
Y_mask = np.asarray(np.ones_like(Y), dtype=floatX)
Y_mask[-1, 0] = 0
Blanked_y_mask = blanked_y_mask.eval({y_mask: Y_mask})
Blanked_y = blanked_y.eval({y: Y})
assert (Blanked_y == np.array([[1, 1, 1], [0, 0, 0], [1, 1, 1], [0, 0, 0],
[1, 1, 1], [0, 0, 0], [1, 1, 1]],
dtype='int32')).all()
assert (Blanked_y_mask == np.array(
[[1., 1., 1.], [1., 1., 1.], [1., 1., 1.], [1., 1., 1.], [1., 1., 1.],
[0., 1., 1.], [0., 1., 1.]],
dtype=floatX)).all()
def test_ctc_log_path_probabs():
LENGTH = 10
BATCHES = 3
CLASSES = 2
N_LABELS = 3
y_hat = T.tensor3('features')
input_mask = T.matrix('features_mask')
y_hat_mask = input_mask
y = T.lmatrix('phonemes')
y_mask = T.matrix('phonemes_mask')
blanked_y, blanked_y_mask = ctc_cost._add_blanks(y=y,
blank_symbol=1,
y_mask=y_mask)
p = ctc_cost._log_path_probabs(blanked_y, y_hat, blanked_y_mask,
y_hat_mask, 1)
Y_hat = np.zeros((LENGTH, BATCHES, CLASSES + 1), dtype=floatX)
Y_hat[:, :, 0] = .7
Y_hat[:, :, 1] = .2
Y_hat[:, :, 2] = .1
Y = np.zeros((N_LABELS, BATCHES), dtype='int64')
Y_hat_mask = np.ones((LENGTH, BATCHES), dtype=floatX)
Y_hat_mask[-2:, 0] = 0
Y_mask = np.asarray(np.ones_like(Y), dtype=floatX)
forward_probs = p.eval({
y_hat: Y_hat,
y: Y,
y_hat_mask: Y_hat_mask,
y_mask: Y_mask
})
assert forward_probs[-2, 0, 0] == -np.inf
Y_mask[-1] = 0
forward_probs_y_mask = p.eval({
y_hat: Y_hat,
y: Y,
y_hat_mask: Y_hat_mask,
y_mask: Y_mask
})
assert forward_probs_y_mask[-1, 1, -2] == -np.inf
assert not np.isnan(forward_probs).any()
