Created by: alexqdh

程序还没有开始运行，加上optimizer以后报错

optimizer = fluid.optimizer.Adam(learning_rate=self.lr)
train_op = optimizer.minimize(cost)

报错信息:

Input(Out@GRAD) shouldn't be null. at [/paddle/paddle/fluid/operators/reshape_op.cc:314]
PaddlePaddle Call Stacks:
0       0x7fd071284455p void paddle::platform::EnforceNotMet::IniTraceback (most recent call last):
  File "main.py", line 24, in <module>
    autodl_exe = autodl.AutoDL()
  File "/paddle/parl/paddle-demo/autodl/autodl.py", line 40, in __init__
    self.controller = AutoDLController(self.algorithm, self.parse_args)#self.parse_args.num_nodes,
  File "/paddle/parl/paddle-demo/autodl/autodl_controller.py", line 34, in __init__
    super(AutoDLController, self).__init__(algorithm)
  File "/usr/local/lib/python2.7/dist-packages/parl-1.0-py2.7.egg/parl/framework/agent_base.py", line 46, in __init__
    self.build_program()
  File "/paddle/parl/paddle-demo/autodl/autodl_controller.py", line 74, in build_program
    self.update_op = self.alg.define_learn(obs=inputs, reward=reward, action=None)
  File "/paddle/parl/paddle-demo/autodl/reinforce_policy_gradient.py", line 71, in define_learn
    train_op = optimizer.minimize(cost)
  File "/usr/local/lib/python2.7/dist-packages/paddle/fluid/optimizer.py", line 404, in minimize
    parameter_list, no_grad_set)
  File "/usr/local/lib/python2.7/dist-packages/paddle/fluid/optimizer.py", line 316, in backward
    return append_backward(loss, parameter_list, no_grad_set, callbacks)
  File "/usr/local/lib/python2.7/dist-packages/paddle/fluid/backward.py", line 518, in append_backward
    _append_backward_vars_(root_block, fwd_op_num, grad_to_var, grad_info_map)
  File "/usr/local/lib/python2.7/dist-packages/paddle/fluid/backward.py", line 354, in _append_backward_vars_
    op_desc.infer_shape(block.desc)
EnforceNotMet: Input(Out@GRAD) shouldn't be null. at [/paddle/paddle/fluid/operators/reshape_op.cc:314]