尝试使用未初始化的值scale5c_branch2c / scale5c_branch2c_beta

时间:2018-10-11 07:33:46

标签: keras initialization

sess = K.get_session()
resnetBuilder = ResnetBuilder()
fc,att,模型= resnetBuilder.ResNet101('resnet101.h5',img.shape)
数组= sess.run(att)
=============================
错误:

ResNet101中的文件“ /Users/workspace/misc/resnet_keras.py”,第456行
    x = identity_block(x,3,[512,512,2048],stage = 5,block ='c')
  在identity_block中,文件“ /Users/heshuguo/workspace/misc/resnet_keras.py”第229行
    x =比例尺(axis = bn_axis,name = scale_name_base +'2c')(x)
  在调用
中的文件“ /usr/local/lib/python2.7/site-packages/keras/engine/topology.py”,第576行     self.build(input_shapes [0])
  在版本
中,文件“ /Users/workspace/misc/resnet_keras.py”,第333行     self.beta = K.variable(self.beta_init(shape),name ='%s_beta'%self.name)
  文件“ /usr/local/lib/python2.7/site-packages/keras/backend/tensorflow_backend.py”,行380,在变量中     v = tf.Variable(value,dtype = tf.as_dtype(dtype),name = name)
   init 中的文件“ /usr/local/lib/python2.7/site-packages/tensorflow/python/ops/variables.py”,第259行     约束=约束)
  _init_from_args中的文件“ /usr/local/lib/python2.7/site-packages/tensorflow/python/ops/variables.py”,第422行
    self._snapshot = array_ops.identity(self._variable,name =“ read”)
  标识为“ /usr/local/lib/python2.7/site-packages/tensorflow/python/ops/array_ops.py”的第80行,
    返回gen_array_ops.identity(input,name = name)
  文件“ /usr/local/lib/python2.7/site-packages/tensorflow/python/ops/gen_array_ops.py”,行3264,标识为     “身份”,输入=输入,名称=名称)
  _apply_op_helper中的文件“ /usr/local/lib/python2.7/site-packages/tensorflow/python/framework/op_def_library.py”,行787     op_def = op_def)
  文件“ /usr/local/lib/python2.7/site-packages/tensorflow/python/util/deprecation.py”,行454,在new_func中     return func(* args,** kwargs)
  在create_op中,文件“ /usr/local/lib/python2.7/site-packages/tensorflow/python/framework/ops.py”,行3155     op_def = op_def)
   init
中的文件“ /usr/local/lib/python2.7/site-packages/tensorflow/python/framework/ops.py”,第1717行     self._traceback = tf_stack.extract_stack()

FailedPreconditionError(请参阅上面的回溯):尝试使用未初始化的值scale5c_branch2c / scale5c_branch2c_beta
     [[节点:scale5c_branch2c / scale5c_branch2c_beta / read = IdentityT = DT_FLOAT,_device =“ / job:localhost /副本:0 / task:0 / device:CPU:0”]]
========================================
如何解决该问题?

0 个答案:

没有答案