我在OCR上使用CRNN(CNN + RNN + CTC Loss)的模型。我使用的是Tensorflow Keras
这是我的代码[来自CTC Loss]。
labels = Input(name='the_labels', shape=[max_label_len], dtype='float32')
input_length = Input(name='input_length', shape=[1], dtype='int64')
label_length = Input(name='label_length', shape=[1], dtype='int64')
def ctc_lambda_func(args):
y_pred, labels, input_length, label_length = args
return K.ctc_batch_cost(labels, y_pred, input_length, label_length)
loss_out = Lambda(ctc_lambda_func, output_shape=(1,), name='ctc')([outputs, labels, input_length, label_length])
#model to be used at training time
model = Model(inputs=[inputs, labels, input_length, label_length], outputs=loss_out)
model.compile(loss={'ctc': lambda y_true, y_pred: y_pred}, optimizer = 'adam')
filepath="best_model.hdf5"
checkpoint = ModelCheckpoint(filepath=filepath, monitor='val_loss', verbose=1, save_best_only=True, mode='auto')
callbacks_list = [checkpoint]
training_img = np.array(training_img)
train_input_length = np.array(train_input_length)
train_label_length = np.array(train_label_length)
valid_img = np.array(valid_img)
valid_input_length = np.array(valid_input_length)
valid_label_length = np.array(valid_label_length)
训练时这里出错。
batch_size = 256
epochs = 10
model.fit(x=[training_img, train_padded_txt, train_input_length, train_label_length], y=np.zeros(len(training_img)),
batch_size=batch_size, epochs = epochs,
validation_data = ([valid_img, valid_padded_txt, valid_input_length, valid_label_length], [np.zeros(len(valid_img))]),
verbose = 1, callbacks = callbacks_list)
ERROR RESULT:
Train on 448 samples, validate on 49 samples
Epoch 1/10
---------------------------------------------------------------------------
InvalidArgumentError Traceback (most recent call last)
<ipython-input-15-1322212af569> in <module>()
4 batch_size=batch_size, epochs = epochs,
5 validation_data = ([valid_img, valid_padded_txt, valid_input_length, valid_label_length], [np.zeros(len(valid_img))]),
----> 6 verbose = 1, callbacks = callbacks_list)
7 frames
/usr/local/lib/python3.6/dist-packages/tensorflow/python/eager/execute.py in quick_execute(op_name, num_outputs, inputs, attrs, ctx, name)
58 ctx.ensure_initialized()
59 tensors = pywrap_tfe.TFE_Py_Execute(ctx._handle, device_name, op_name,
---> 60 inputs, attrs, num_outputs)
61 except core._NotOkStatusException as e:
62 if name is not None:
InvalidArgumentError: sequence_length(0) <= 18
[[node ctc/CTCLoss (defined at /usr/local/lib/python3.6/dist-packages/keras/backend/tensorflow_backend.py:3009) ]] [Op:__inference_keras_scratch_graph_12073]
Function call stack:
keras_scratch_graph
My CRNN architecture is inspired by VGG-16, I'm using 13 conv layers and 3 bi-directional LSTM Layer.I am using CTC Loss and then I got error.My data is 1000 text-image contains 4-8 words (700 training&valid, 300 testing). I am using CTC Loss and then I got error.My data is 1000 text-image contains 4-8 words (700 for training&valid, 300 for testing)
如果你想查看我的代码:这是我的代码使用谷歌colab。https:/colab.research.google.comdrive1nMRNUsLDNrpgeTxPFQ4mhobnFdpbmwUx。
我修正了这个错误。就是因为这个
之前: 启用前
# split the 700 data into validation and training dataset as 10% and 90% respectively
if i%10 == 0:
valid_orig_txt.append(txt)
valid_label_length.append(len(txt))
valid_input_length.append(31)
valid_img.append(img)
valid_txt.append(encode_to_labels(txt))
else:
orig_txt.append(txt)
train_label_length.append(len(txt))
train_input_length.append(31)
training_img.append(img)
training_txt.append(encode_to_labels(txt))
之后:
# split the 700 data into validation and training dataset as 10% and 90% respectively
if i%10 == 0:
valid_orig_txt.append(txt)
valid_label_length.append(len(txt))
valid_input_length.append(18)
valid_img.append(img)
valid_txt.append(encode_to_labels(txt))
else:
orig_txt.append(txt)
train_label_length.append(len(txt))
train_input_length.append(18)
training_img.append(img)
training_txt.append(encode_to_labels(txt))