def test_compute_class_weights(self):
# regression test for issue #181
np.random.seed(0)
y = np.random.choice(a=[0, 1, 2], size=1000, p=[0.3, 0.6, 0.1])
class_counts = Counter(y)
weights = compute_class_weights('log', class_counts=class_counts)
self.assertEqual(weights[1], 1.0)
def _dataset_with_targets(self, Xs, Y, train, context=None):
if not callable(Xs) and not callable(Y):
if self.config.use_auxiliary_info:
dataset = lambda: zip(
Xs, Y, context
) # Do not need to check if context is callable - it is turned in along with Xs, and thus must have the same form
else:
dataset = lambda: zip(Xs, Y)
elif callable(Xs) and callable(Y):
if self.config.use_auxiliary_info:
dataset = lambda: zip(Xs(), Y(), context()
) # encode one sample at a time.
else:
dataset = lambda: zip(Xs(), Y())
else:
raise ValueError(
"Either neither or both of Xs and Y should be callable, not a mixture"
)
dataset_encoded = lambda: itertools.chain.from_iterable(
map(lambda xy: self.text_to_tokens_mask(*xy), dataset()))
if not callable(Y) and train:
dataset_encoded_list = list(dataset_encoded())
dataset_encoded_list = self._filter_empty_examples(
dataset_encoded_list)
class_counts = self._compute_class_counts(dataset_encoded_list)
self.config.dataset_size = len(dataset_encoded_list)
if self.config.class_weights is not None:
self.config.class_weights = compute_class_weights(
class_weights=self.config.class_weights,
class_counts=class_counts)
shape_def = self.feed_shape_type_def()
return Dataset.from_generator(
lambda: self.wrap_tqdm(dataset_encoded(), train), *shape_def)
def _dataset_with_targets(self, Xs, Y, train):
if not callable(Xs) and not callable(Y):
dataset = lambda: zip(Xs, Y)
elif callable(Xs) and callable(Y):
dataset = lambda: zip(Xs(), Y()) # encode one sample at a time.
else:
raise ValueError(
"Either neither or both of Xs and Y should be callable, not a mixture"
)
dataset_encoded = lambda: itertools.chain.from_iterable(
map(lambda xy: self.text_to_tokens_mask(*xy), dataset()))
shape_def = self.feed_shape_type_def()
if not callable(Y) and train:
dataset_encoded_list = list(dataset_encoded())
class_counts = self._compute_class_counts(dataset_encoded_list)
self.config.dataset_size = len(dataset_encoded_list)
if self.config.class_weights is not None:
self.config.class_weights = compute_class_weights(
class_weights=self.config.class_weights,
class_counts=class_counts)
return Dataset.from_generator(
lambda: self.wrap_tqdm(dataset_encoded(), train), *shape_def)
def _compute_class_weights(self, class_weights, class_counts):
class_weights = compute_class_weights(
class_weights=class_weights,
class_counts=class_counts,
n_total=self.config.dataset_size,
multilabel=True
)
return class_weights
def _post_data_initialization(self, Y):
self.label_encoder = self._target_encoder()
if not callable(Y):
Y_fit = Y
self.label_encoder.fit(Y)
else:
Y_fit = list(itertools.islice(Y(), 10000))
self.label_encoder.fit(Y_fit)
self.config.pad_idx = self.pad_idx
target_dim = self.label_encoder.target_dim
self.lm_loss_coef = self.config.lm_loss_coef if target_dim is not None else 1.0
self.target_dim = target_dim
if Y_fit is not None:
self.config.class_weights = compute_class_weights(
class_weights=self.config.class_weights, Y=Y_fit)
def _compute_class_weights(self, class_weights, class_counts):
return compute_class_weights(class_weights=class_weights,
class_counts=class_counts)
