def predict(self):
file_list = []
X, Y = get_image_data_ctc('./img_data/ctc_test/', file_list)
y_pred = self.base_model.predict(X)
shape = y_pred[:, :, :].shape # 2:
out = K.get_value(
K.ctc_decode(y_pred[:, :, :],
input_length=np.ones(shape[0]) *
shape[1])[0][0])[:, :seq_len] # 2:
print()
error_count = 0
for i in range(len(X)):
print(file_list[i])
str_src = str(os.path.split(
file_list[i])[-1]).split('.')[0].split('_')[-1]
print(out[i])
str_out = ''.join([str(char_set[x]) for x in out[i] if x != -1])
print(str_src, str_out)
if str_src != str_out:
error_count += 1
print('This is a error image---------------------------:',
error_count)
def train(self, batch_size=32, nb_epoch=15, data_augmentation=False):
X, Y = get_image_data_ctc(dir='./img_data/ctc/')
print('train----------', X.shape, Y.shape)
conv_shape = self.conv_shape
maxin = 45000
result = self.ctc_model.fit(
[
X[:maxin], Y[:maxin],
np.array(np.ones(len(X)) * int(conv_shape[1]))[:maxin],
np.array(np.ones(len(X)) * seq_len)[:maxin]
],
Y[:maxin],
batch_size=256,
epochs=50,
callbacks=[EarlyStopping(patience=10)
], # checkpointer, history,history, plotter,
validation_data=([
X[maxin:], Y[maxin:],
np.array(np.ones(len(X)) * int(conv_shape[1]))[maxin:],
np.array(np.ones(len(X)) * seq_len)[maxin:]
], Y[maxin:]),
)