Created by: Hsoms

# -*- coding: utf-8 -*-
"""
Created on Thu Dec 19 21:48:34 2019

线性回归问题

@author: 杨
"""
import paddle.fluid as fluid
import paddle
import numpy as np




#定义线性回归模型
x = fluid.layers.data(name='x', shape=[13], dtype='float32')  
hidden = fluid.layers.fc(input=x, size=100, act='relu')   #提取100个特征
net = fluid.layers.fc(input=hidden, size=1, act=None)

#定义损失函数
y=fluid.layers.data(name="y",shape=[1],dtype="float32")
cost=fluid.layers.square_error_cost(input=net,label=y)
avg_cost=fluid.layers.mean(cost)


#复制一个主程序，方便以后使用
test_program=fluid.default_main_program().clone(for_test=True)


#定义优化算法
optimizer=fluid.optimizer.SGDOptimizer(learning_rate=0.01)
opts=optimizer.minimize(avg_cost)


# 创建一个使用CPU的执行器
place = fluid.CPUPlace()
exe = fluid.Executor(place)
# 进行参数初始化
exe.run(fluid.default_startup_program())


# 定义训练和测试数据
x_data = np.array([[1.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0], 
                   [2.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0], 
                   [3.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0], 
                   [4.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0], 
                   [5.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0]]).astype('float32')
y_data = np.array([[3.0], [5.0], [7.0], [9.0], [11.0]]).astype('float32')
test_data = np.array([[6.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0]]).astype('float32')



# 开始训练100个pass
for pass_id in range(10):
    train_cost = exe.run(program=fluid.default_main_program(),
                         feed={'x': x_data, 'y': y_data},
                         fetch_list=[avg_cost])
    print("Pass:%d, Cost:%0.5f" % (pass_id, train_cost[0]))


# 开始训练和测试
for pass_id in range(10):
    # 开始训练并输出最后一个batch的损失值
    train_cost = 0
    for batch_id, data in enumerate(train_reader()):
        train_cost = exe.run(program=fluid.default_main_program(),
                             feed=feeder.feed(data),
                             fetch_list=[avg_cost])
    print("Pass:%d, Cost:%0.5f" % (pass_id, train_cost[0][0]))

    # 开始测试并输出最后一个batch的损失值
    test_cost = 0
    for batch_id, data in enumerate(test_reader()):
        test_cost = exe.run(program=fluid.default_main_program(),
                            feed=feeder.feed(data),
                            fetch_list=[avg_cost])
    print('Test:%d, Cost:%0.5f' % (pass_id, test_cost[0][0]))

报错：

代码报错 ---------------------- Error Message Summary: ---------------------- PaddleCheckError: Expected dtype != -1, but received dtype:-1 == -1:-1. Sum operator should have at least one tensor at [D:\1.6.1\paddle\paddle\fluid\operators\sum_op.cc:142] [operator < sum > error] #21852