Created by: barrierye

1. 文档中为cube_prepare_quant .sh，实际为cube_quant_prepare.sh
2. quant 例子不能跑通（非quant版本可以跑通） 启动cube：

# sh cube_quant_prepare.sh&
[1] 20547
generate compressed sparse param sequence file
Version size: 4
[libprotobuf ERROR /data/barriery/Serving/build-server/third_party/protobuf/src/extern_protobuf/src/google/protobuf/message_lite.cc:119] Can't parse message of type "paddle.framework.proto.VarType.TensorDesc" because it is missing required fields: (cannot determine missing fields for lite message)
Cannot parse tensor desc
E0619 19:07:26.123454 20559 crovl_builder_increment.cpp:176] _data_buf_len == 0
E0619 19:07:26.123708 20559 crovl_builder_increment.cpp:205] _index_buf_len == 0
./
mv: 无法获取"./cube/data/0_0/test_dict_part0/*" 的文件状态(stat): 没有那个文件或目录

启动server进行预测，server端报错如下：

--------------------------------------------
C++ Call Stacks (More useful to developers):
--------------------------------------------

------------------------------------------
Python Call Stacks (More useful to users):
------------------------------------------
  File "/usr/lib64/python2.7/site-packages/paddle/fluid/framework.py", line 2525, in append_op
    attrs=kwargs.get("attrs", None))
  File "/usr/lib64/python2.7/site-packages/paddle/fluid/layer_helper.py", line 43, in append_op
    return self.main_program.current_block().append_op(*args, **kwargs)
  File "/usr/lib64/python2.7/site-packages/paddle/fluid/layers/nn.py", line 344, in fc
    "y_num_col_dims": 1})
  File "/Serving/python/examples/criteo_ctr_with_cube/network_conf.py", line 46, in mlp
    scale=1 / math.sqrt(mlp_input.shape[1]))))
  File "/Serving/python/examples/criteo_ctr_with_cube/network_conf.py", line 71, in dnn_model
    predict = mlp(mlp_in)
  File "local_train.py", line 45, in train
    args.sparse_feature_dim)
  File "local_train.py", line 100, in <module>
    train()

----------------------
Error Message Summary:
----------------------
InvalidArgumentError: After flatten the input tensor X and Y to 2-D dimensions matrix X1 and Y1, the matrix X1's width must be equal with matrix Y1's height. But received X's shape = [1, 845], X1's shape = [1, 845], X1's width = 845; Y's shape = [273, 400], Y1's shape = [273, 400], Y1's height = 273.
  [Hint: Expected x_mat_dims[1] == y_mat_dims[0], but received x_mat_dims[1]:845 != y_mat_dims[0]:273.] at (/paddle/paddle/fluid/operators/mul_op.cc:90)
  [operator < mul > error]