def fit(self, dataset, batch_size=32, epochs=10, callbacks=[]):
self.network.compile(loss=self.loss(), optimizer=self.optimizer(), metrics=self.metrics())
train_sequence = DatasetSequence(dataset.x_train, dataset.y_train, batch_size, augment_fn=self.batch_augment_fn, format_fn=self.batch_format_fn)
test_sequence = DatasetSequence(dataset.x_test, dataset.y_test, batch_size, augment_fn=self.batch_augment_fn, format_fn=self.batch_format_fn)
self.network.fit_generator(
train_sequence,
epochs=epochs,
callbacks=callbacks,
validation_data=test_sequence,
use_multiprocessing=True,
workers=1,
shuffle=True
)
def evaluate(self, x, y, batch_size: int=16, verbose=True) -> float:
test_sequence = DatasetSequence(x, y, batch_size, format_fn=self.batch_format_fn)
# We can use the `ctc_decoded` layer that is part of our model here.
decoding_model = KerasModel(inputs=self.network.input, outputs=self.network.get_layer('ctc_decoded').output)
preds = decoding_model.predict_generator(test_sequence)
trues = np.argmax(y, -1)
pred_strings = [''.join(self.data.mapping.get(label, '') for label in pred).strip(' |_') for pred in preds]
true_strings = [''.join(self.data.mapping.get(label, '') for label in true).strip(' |_') for true in trues]
char_accuracies = [
1 - editdistance.eval(true_string, pred_string) / len(true_string)
for pred_string, true_string in zip(pred_strings, true_strings)
]
if verbose:
sorted_ind = np.argsort(char_accuracies)
print("\nLeast accurate predictions:")
for ind in sorted_ind[:5]:
print(f'True: {true_strings[ind]}')
print(f'Pred: {pred_strings[ind]}')
print("\nMost accurate predictions:")
for ind in sorted_ind[-5:]:
print(f'True: {true_strings[ind]}')
print(f'Pred: {pred_strings[ind]}')
print("\nRandom predictions:")
for ind in np.random.randint(0, len(char_accuracies), 5):
print(f'True: {true_strings[ind]}')
print(f'Pred: {pred_strings[ind]}')
mean_accuracy = np.mean(char_accuracies)
return mean_accuracy
def evaluate(self, x, y, verbose=True):
sequence = DatasetSequence(x, y)
preds_raw = self.network.predict_generator(sequence)
trues = np.argmax(y, -1)
preds = np.argmax(preds_raw, -1)
pred_strings = [
''.join(self.data.mapping.get(label, '')
for label in pred).strip(' |_') for pred in preds
]
true_strings = [
''.join(self.data.mapping.get(label, '')
for label in true).strip(' |_') for true in trues
]
char_accuracies = [
1 - editdistance.eval(true_string, pred_string) / len(true_string)
for pred_string, true_string in zip(pred_strings, true_strings)
]
if verbose:
sorted_ind = np.argsort(char_accuracies)
print("\nLeast accurate predictions:")
for ind in sorted_ind[:5]:
print(f'True: {true_strings[ind]}')
print(f'Pred: {pred_strings[ind]}')
print("\nMost accurate predictions:")
for ind in sorted_ind[-5:]:
print(f'True: {true_strings[ind]}')
print(f'Pred: {pred_strings[ind]}')
print("\nRandom predictions:")
for ind in np.random.randint(0, len(char_accuracies), 5):
print(f'True: {true_strings[ind]}')
print(f'Pred: {pred_strings[ind]}')
mean_accuracy = np.mean(char_accuracies)
return mean_accuracy
def evaluate(self, x, y):
sequence = DatasetSequence(x, y, batch_size=16) # Use a small batch size to use less memory
preds = self.network.predict_generator(sequence)
return np.mean(np.argmax(preds, -1) == np.argmax(y, -1))
