问题描述
我收到此错误: '''
InvalidArgumentError: Not enough time for target transition sequence (required: 138,available: 76)0You can turn this error into a warning by using the flag ignore_longer_outputs_than_inputs
[[node model_10/ctc/CTCLoss (defined at <ipython-input-211-3c40d9e71078>:3) ]] [Op:__inference_train_function_18256]
'''
这是我正在使用的代码,我可以把这个参数放在哪里ignore_longer_outputs_than_inputs?
'''
def ctc_lambda_func(args):
y_pred,labels,input_length,label_length = args
return K.ctc_batch_cost(labels,y_pred,label_length)
def add_ctc_loss(input_to_softmax):
the_labels = Input(name='the_labels',shape=(None,),dtype='float32')
input_lengths = Input(name='input_length',shape=(1,dtype='int64')
label_lengths = Input(name='label_length',dtype='int64')
output_lengths = Lambda(input_to_softmax.output_length)(input_lengths)
# CTC loss is implemented in a lambda layer
loss_out = Lambda(ctc_lambda_func,output_shape=(1,name='ctc')(
[input_to_softmax.output,the_labels,output_lengths,label_lengths])
model = Model(
inputs=[input_to_softmax.input,input_lengths,label_lengths],outputs=loss_out)
return model
'''
解决方法
暂无找到可以解决该程序问题的有效方法,小编努力寻找整理中!
如果你已经找到好的解决方法,欢迎将解决方案带上本链接一起发送给小编。
小编邮箱:dio#foxmail.com (将#修改为@)