如何解决张量流版本之间的准确性不一致
我是keras / tensorflow的新手。
在另一个版本的keras中,张量流在准确性方面不一致。
我不知道为什么。
提前谢谢!
import tensorflow as tf
tf.__version__
'1.15.2'
from tensorflow import keras
keras.__version__
'2.2.4-tf'
from keras.models import Model,Sequential
from keras.layers import InputLayer,Dense,BatchNormalization,Activation,Dropout
from keras.callbacks import EarlyStopping
from keras import regularizers
classify = [
InputLayer(input_shape=(X_train.shape[1],)),BatchNormalization(),Dense(128),Activation('relu'),Dense(64,activity_regularizer=regularizers.l1(1e-5)),Dense(1),Activation('sigmoid')
]
model = Sequential(classify)
model.compile(loss='binary_crossentropy',optimizer='adam',metrics=['accuracy'])
model.fit(X_train,y_train,epochs=5,batch_size=128,shuffle="batch")
print(model.metrics_names,model.evaluate(X_test,y_test))
['loss','acc'] [0.02403441002866048,0.994511238891793]
import tensorflow as tf
tf.__version__
“ 2.3.0”
from tensorflow import keras
keras.__version__
“ 2.4.0”
model.compile(loss='binary_crossentropy',y_test))
['损失','准确性'] [0.6886715888977051,0.5517511963844299]
解决方法
尝试:
for i,var in enumerate(model.trainable_weights):
print(model.trainable_weights[i].name)
ref:https://github.com/tensorflow/tensorflow/issues/40638
,tf.executing_eagerly()
tf.compat.v1.disable_eager_execution()
tf.executing_eagerly()
这个问题由Eager在张量流2中引起
版权声明:本文内容由互联网用户自发贡献,该文观点与技术仅代表作者本人。本站仅提供信息存储空间服务,不拥有所有权,不承担相关法律责任。如发现本站有涉嫌侵权/违法违规的内容, 请发送邮件至 dio@foxmail.com 举报,一经查实,本站将立刻删除。