官方代码：
import paddle.fluid as fluid
import numpy as np
place = fluid.core.CPUPlace()

构建网络

x = fluid.layers.data(name="x", shape=[1], dtype="int32", lod_level=1)
res = fluid.layers.hash(name="res",input=x, hash_size=1000, num_hash=4)

创建CPU执行器

exe = fluid.Executor(place)
exe.run(fluid.default_startup_program())
in1 = np.array([[1,2],[3,4]]).astype("int32")
print(in1)
x_i = fluid.core.LoDTensor()
x_i.set(in1,place)
x_i.set_recursive_sequence_lengths(0,2)
res = exe.run(fluid.default_main_program(), feed={'x':x_i}, fetch_list=[res], return_numpy=False)
print(np.array(res[0]))
报错：
InvalidArgumentError: The Tensor in the hash Op's Input Variable X(creative_id) is not initialized.
[Hint: Expected t->IsInitialized() == true, but received t->IsInitialized():0 != true:1.] at (/paddle/paddle/fluid/framework/operator.cc:1264)
[operator < hash > error]