TypeError: while_loop() got an unexpected keyword argument 'maximum_iterations'

绝地灬酷狼 2021-12-19 23:35 229阅读 0赞

转载自：[https://blog.csdn.net/zyh2004883/article/details/84337872][https_blog.csdn.net_zyh2004883_article_details_84337872]

错误:

TypeError: while\_loop() got an unexpected keyword argument 'maximum\_iterations'

参照[https://blog.csdn.net/qq\_27009517/article/details/83275846][https_blog.csdn.net_qq_27009517_article_details_83275846]

重新安装keras

因为我的CUDA版本为CUDA8.0，所以tensorflow版本有限制，只能用1.4.0以下的版本

所以tensorflow-gpu=1.4.0

之前keras版本太高了，所以才会报以上错误

更新keras版本后，错误解决

pip install keras==2.1.2

# -*- coding: utf-8 -*- 
    # @Time : 2019/7/1 8:17 
    # @Author : ZXL 
    # @Site :  
    # @File : many_input_API.py 
    # @Software: PyCharm
    
    from keras.models import Model
    from keras import layers
    from keras import Input
    
    text_vocabulary_size = 10000
    question_vocabulary_size = 10000
    answer_vocabulary_size = 500
    
    text_input = Input(shape=(None,), dtype='int32', name='text')
    embedded_text = layers.Embedding(64, text_vocabulary_size)(text_input)
    encoded_text = layers.LSTM(32)(embedded_text)
    
    question_input = Input(shape=(None,),dtype='int32',name='question')
    embedded_question = layers.Embedding(32, question_vocabulary_size)(question_input)
    encoded_question = layers.LSTM(16)(embedded_question)
    
    concatenated = layers.concatenate([encoded_text, encoded_question],axis=-1)#两个输入处理结果联系起来
    
    answer = layers.Dense(answer_vocabulary_size,activation='softmax')(concatenated)
    model = Model([text_input, question_input], answer)#模型声明
    model.compile(optimizer='rmsprop',loss='categorical_crossentropy',metrics=['acc'])#compile阶段
    
    model.summary()

![在这里插入图片描述][20190701083952169.png]

[https_blog.csdn.net_zyh2004883_article_details_84337872]: https://blog.csdn.net/zyh2004883/article/details/84337872
[https_blog.csdn.net_qq_27009517_article_details_83275846]: https://blog.csdn.net/qq_27009517/article/details/83275846
[20190701083952169.png]: /images/20211219/fc402b0ed6ed4a1f89dc3535b56afd3d.png