问题描述：
编写了一个多元线性回归问题，但是在以下环境下收敛速度非常缓慢：

• 版本、环境信息：

   1）PaddlePaddle版本：2.0及2.2.1
   2）CPU及相关环境：BML CodeLab默认CPU及标准预设环境
   3）GPU：未开启
   4）系统环境：同上2）
结果通过150000次的运算，loss依然在6000以上。
(最终收敛结果：epoch_id is 149000,avg_loss is [6082.4243])

于是我采用了以下环境进行重新训练，收敛速度呈现了上亿倍的差异。
   1）PaddlePaddle版本：1.8
   2）CPU及相关环境：BML CodeLab默认CPU及标准预设环境
   3）GPU：未开启
   4）系统环境：同上2）
通过150000次的运算，loss已经收敛到1.9*10^-10
(最终收敛结果：epoch_id is 149000,avg_loss is [1.9255095e-10])

• 完整代码

read data from hdfs #1.导入各类库
import paddle
import paddle.fluid as fluid
from paddle.fluid.dygraph.nn import Linear
import numpy as np
import random

#2 ，准备数据b=3a+5c-100
a=list(range(1000))
c=list(range(1000,0,-1))
random.shuffle(a)
random.shuffle(c)
a=np.array(a).reshape([-1,1])
c=np.array(c).reshape([-1,1])
b=a3+c5-100
b=np.reshape(b,[-1,1])
a_c_input=np.concatenate((a,c),axis=1)
a_c_input=np.array(a_c_input).astype("float32")
b=np.array(b).astype("float32")

#3 ,准备模型
class MNIST(fluid.dygraph.Layer):
definit(self):
super(MNIST,self).init()
self.fc=Linear(input_dim=2,output_dim=1,act=None)
def forward(self,inputs):
x=self.fc(inputs)
return x

#4 ，开始训练
with fluid.dygraph.guard():
model=MNIST()
model.train()
iterid=[]
losses=[]
image=fluid.dygraph.to_variable(a_c_input)
label=fluid.dygraph.to_variable(b)
optimizer = fluid.optimizer.AdamOptimizer(learning_rate=0.001, parameter_list=model.parameters())
#optimizer = fluid.optimizer.SGDOptimizer(learning_rate=0.001, parameter_list=model.parameters())
#optimizer = fluid.optimizer.AdamOptimizer(learning_rate=0.01, regularization=fluid.regularizer.L2Decay(regularization_coeff=0.1),
#parameter_list=model.parameters())
EPOCH_NUM=150000
for epoch_id in range(EPOCH_NUM):

predict=model(image)
    loss=fluid.layers.square_error_cost(predict,label)
    avg_loss=fluid.layers.mean(loss)
    iterid.append(epoch_id)
    losses.append(avg_loss.numpy())
    if epoch_id % 1000 ==0:
        print("epoch_id is {},avg_loss is {}".format(epoch_id,avg_loss.numpy()))
    avg_loss.backward()
    optimizer.minimize(avg_loss)
    model.clear_gradients
fluid.save_dygraph(model.state_dict(),"one_yuan")

import matplotlib.pyplot as plt
%matplotlib inline
plt.plot(iterid,losses)
plt.grid()
plt.show()

————————————————————————华丽分割线————————————————————————————

Thank you for contributing to PaddlePaddle.
Before submitting the issue, you could search issue in the github.Probably there was a similar issue submitted or resolved before.
If there is no solution,please make sure that this is a issue of models including the following details:

System information

-PaddlePaddle version （eg.1.1）or CommitID
-CPU: including CPUMKL/OpenBlas/MKLDNN version
-GPU: including CUDA/CUDNN version
-OS Platform (eg.Mac OS 10.14)
-Python version
-Name of Models&Dataset/details of operator
Note: You can get most of the information by running summary_env.py .

To Reproduce

Steps to reproduce the behavior

Describe your current behavior
Code to reproduce the issue
Other info / logs