Question

使用ctc_batch_cost时计算出的精度是什么，模型输出丢失了，我不明白精度是如何计算的

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)

input_x = Input(tensor=image,name='image')

labels = Input(name='ctc_label',tensor=ctc_label,dtype='int64')
input_length = Input(name='input_length', tensor=input_length, dtype='int64')
label_length = Input(name='label_length', tensor=label_length, dtype='int64')

loss_out = Lambda(ctc_lambda_func, output_shape=(1,), name='ctc')([y_pred, labels, input_length, label_length]) 

model = Model(inputs=[input_x, labels, input_length, label_length], outputs=loss_out)
adam = Adam()



model.compile(loss={'ctc': lambda y_true, y_pred: y_pred},target_tensors=[output], optimizer=adam,metrics=['accuracy'])

Answer 1

使用ctc_batch_cost时如何计算准确性

使用ctc_accuracy时如何计算精度

1 个答案: