【新⼿⼊门】课程3-Paddle⼊门-波⼠顿房价预测
经典的线性回归模型主要⽤来预测⼀些存在着线性关系的数据集。回归模型可以理解为:存在⼀个点集,⽤⼀条曲线去拟合它分布的过程。如果拟合曲线是⼀条直线,则称为线性回归。如果是⼀条⼆次曲线,则被称为⼆次回归。线性回归是回归模型中最简单的⼀种。 本教程使⽤PaddlePaddle建⽴起⼀个房价预测模型。
在线性回归中:
linux如何恢复被删数据(1)假设函数是指,⽤数学的⽅法描述⾃变量和因变量之间的关系,它们之间可以是⼀个线性函数或⾮线性函数。 在本次线性回顾模型中,我们的假设函数为 Y’= wX+b ,其中,Y’表⽰模型的预测结果(预测房价),⽤来和真实的Y区分。模型要学习的参数即:w,b。
(2)损失函数是指,⽤数学的⽅法衡量假设函数预测结果与真实值之间的误差。这个差距越⼩预测越准确,⽽算法的任务就是使这个差距越来越⼩。 建⽴模型后,我们需要给模型⼀个优化⽬标,使得学到的参数能够让预测值Y’尽可能地接近真实值Y。这个实值通常⽤来反映模型误差的⼤⼩。不同问题场景下采⽤不同的损失函数。 对于线性模型来讲,最常⽤的损失函数就是均⽅误差(Mean Squared Error,MSE)。
(3)优化算法:神经⽹络的训练就是调整权重(参数)使得损失函数值尽可能得⼩,在训练过程中,将损失函数值逐渐收敛,得到⼀组使得神经⽹络拟合真实模型的权重(参数)。所以,优化算法的最终⽬标是到损失函数的最⼩值。⽽这个寻过程就是不断地微调变量w和b的值,⼀步⼀步地试出这个最⼩值。 常见的优化算法有随机梯度下降法(SGD)、Adam算法等等
⾸先导⼊必要的包,分别是:
paddle.fluid--->PaddlePaddle深度学习框架
numpy---------->python基本库,⽤于科学计算
os------------------>python的模块,可使⽤该模块对操作系统进⾏操作
matplotlib----->python绘图库,可⽅便绘制折线图、散点图等图形
In[1]
import paddle.fluid as fluid
import paddle
import numpy as np
import os
import matplotlib.pyplot as plt
Step1:准备数据。
(1)uci-housing数据集介绍
数据集共506⾏,每⾏14列。前13列⽤来描述房屋的各种信息,最后⼀列为该类房屋价格中位数。
PaddlePaddle提供了读取uci_housing训练集和测试集的接⼝,分别为paddle.dataset.ain()和
paddle.dataset.st()。
(2)train_reader和test_reader
paddle.batch()表⽰每BATCH_SIZE组成⼀个batch
In[2]
BUF_SIZE=500
BATCH_SIZE=20
#⽤于训练的数据提供器,每次从缓存中随机读取批次⼤⼩的数据
train_reader = paddle.batch(
buf_size=BUF_SIZE),
batch_size=BATCH_SIZE)
#⽤于测试的数据提供器,每次从缓存中随机读取批次⼤⼩的数据
test_reader = paddle.batch(
buf_size=BUF_SIZE),
batch_size=BATCH_SIZE)
[==================================================]housing/housing.data not found, downloading paddlemodels.bj.bcebos/uci_hou /opt/conda/envs/python35-paddle120-env/lib/python3.5/site-packages/paddle/dataset/uci_housing.py:49: UserWarning:
This call to matplotlib.use() has no effect because the backend has already
been chosen; matplotlib.use() must be called *before* pylab, matplotlib.pyplot,
or matplotlib.backends is imported for the first time.
The backend was *originally* set to 'module://ipykernel.pylab.backend_inline' by the following code:
File "/opt/conda/envs/python35-paddle120-env/lib/python3.5/runpy.py", line 193, in _run_module_as_main
"__main__", mod_spec)
File "/opt/conda/envs/python35-paddle120-env/lib/python3.5/runpy.py", line 85, in _run_code
exec(code, run_globals)
File "/opt/conda/envs/python35-paddle120-env/lib/python3.5/site-packages/ipykernel_launcher.py", line 16, in <module>
app.launch_new_instance()
File "/opt/conda/envs/python35-paddle120-env/lib/python3.5/site-packages/traitlets/config/application.py", line 658, in launch_instance
app.start()
File "/opt/conda/envs/python35-paddle120-env/lib/python3.5/site-packages/ipykernel/kernelapp.py", line 505, in start
self.io_loop.start()
File "/opt/conda/envs/python35-paddle120-env/lib/python3.5/site-packages/tornado/platform/asyncio.py", line 132, in start
self.asyncio_loop.run_forever()
File "/opt/conda/envs/python35-paddle120-env/lib/python3.5/asyncio/base_events.py", line 421, in run_forever
self._run_once()
File "/opt/conda/envs/python35-paddle120-env/lib/python3.5/asyncio/base_events.py", line 1425, in _run_once
handle._run()
File "/opt/conda/envs/python35-paddle120-env/lib/python3.5/asyncio/events.py", line 127, in _run
self._callback(*self._args)
File "/opt/conda/envs/python35-paddle120-env/lib/python3.5/site-packages/tornado/ioloop.py", line 758, in _run_callback
ret = callback()软件测试员岗位要求
File "/opt/conda/envs/python35-paddle120-env/lib/python3.5/site-packages/tornado/stack_context.py", line 300, in null_wrapper
return fn(*args, **kwargs)
File "/opt/conda/envs/python35-paddle120-env/lib/python3.5/site-packages/tornado/gen.py", line 1233, in inner
self.run()
File "/opt/conda/envs/python35-paddle120-env/lib/python3.5/site-packages/tornado/gen.py", line 1147, in run
yielded = send(value)
File "/opt/conda/envs/python35-paddle120-env/lib/python3.5/site-packages/ipykernel/kernelbase.py", line 357, in process_one
yield gen.maybe_future(dispatch(*args))
File "/opt/conda/envs/python35-paddle120-env/lib/python3.5/site-packages/tornado/gen.py", line 326, in wrapper
yielded = next(result)
File "/opt/conda/envs/python35-paddle120-env/lib/python3.5/site-packages/ipykernel/kernelbase.py", line 267, in dispatch_shell
yield gen.maybe_future(handler(stream, idents, msg))
File "/opt/conda/envs/python35-paddle120-env/lib/python3.5/site-packages/tornado/gen.py", line 326, in wrapper
yielded = next(result)
File "/opt/conda/envs/python35-paddle120-env/lib/python3.5/site-packages/ipykernel/kernelbase.py", line 534, in execute_request
user_expressions, allow_stdin,
File "/opt/conda/envs/python35-paddle120-env/lib/python3.5/site-packages/tornado/gen.py", line 326, in wrapper
python入门教程appyielded = next(result)
File "/opt/conda/envs/python35-paddle120-env/lib/python3.5/site-packages/ipykernel/ipkernel.py", line 294, in do_execute
res = shell.run_cell(code, store_history=store_history, silent=silent)
File "/opt/conda/envs/python35-paddle120-env/lib/python3.5/site-packages/ipykernel/zmqshell.py", line 536, in run_cell
File "/opt/conda/envs/python35-paddle120-env/lib/python3.5/site-packages/ipykernel/zmqshell.py", line 536, in run_cell
return super(ZMQInteractiveShell, self).run_cell(*args, **kwargs)
File "/opt/conda/envs/python35-paddle120-env/lib/python3.5/site-packages/IPython/core/interactiveshell.py", line 2821, in run_cell
igger('post_run_cell', result)
File "/opt/conda/envs/python35-paddle120-env/lib/python3.5/site-packages/IPython/core/events.py", line 88, in trigger
func(*args, **kwargs)
File "/opt/conda/envs/python35-paddle120-env/lib/python3.5/site-packages/ipykernel/pylab/backend_inline.py", line 164, in configure_once
activate_matplotlib(backend)
File "/opt/conda/envs/python35-paddle120-env/lib/python3.5/site-packages/IPython/core/pylabtools.py", line 314, in activate_matplotlib
matplotlib.pyplot.switch_backend(backend)
File "/opt/conda/envs/python35-paddle120-env/lib/python3.5/site-packages/matplotlib/pyplot.py", line 231, in switch_backend
matplotlib.use(newbackend, warn=False, force=True)
crayon中文翻译File "/opt/conda/envs/python35-paddle120-env/lib/python3.5/site-packages/matplotlib/__init__.py", line 1422, in use
dules['matplotlib.backends'])
File "/opt/conda/envs/python35-paddle120-env/lib/python3.5/importlib/__init__.py", line 166, in reloadc++ 链表排序
_bootstrap._exec(spec, module)
File "/opt/conda/envs/python35-paddle120-env/lib/python3.5/site-packages/matplotlib/backends/__init__.py", line 16, in <module>
line for line in traceback.format_stack()
matplotlib.use('Agg')
(3)打印看下数据是什么样的?PaddlePaddle接⼝提供的数据已经经过归⼀化等处理
(array([-0.02964322, -0.11363636, 0.39417967, -0.06916996, 0.14260276, -0.10109875, 0.30715859, -0.13176829, -0.24127857, 0.05489093, 0.29196451, -0.2368098 , 0.12850267]), array([15.6])),
In[3]
#⽤于打印,查看uci_housing数据
train_data=paddle.dataset.ain();
sampledata=next(train_data())
print(sampledata)
(array([-0.0405441 , 0.06636364, -0.32356227, -0.06916996, -0.03435197,
0.05563625, -0.03475696, 0.02682186, -0.37171335, -0.21419304,
-0.33569506, 0.10143217, -0.21172912]), array([24.]))
Step2:⽹络配置
(1)⽹络搭建:对于线性回归来讲,它就是⼀个从输⼊到输出的简单的全连接层。
对于波⼠顿房价数据集,假设属性和房价之间的关系可以被属性间的线性组合描述。
In[4]
#定义张量变量x,表⽰13维的特征值
x = fluid.layers.data(name='x', shape=[13], dtype='float32')
#定义张量y,表⽰⽬标值
y = fluid.layers.data(name='y', shape=[1], dtype='float32')
#定义⼀个简单的线性⽹络,连接输⼊和输出的全连接层
#input:输⼊tensor;
#size:该层输出单元的数⽬
#act:激活函数
y_predict=fluid.layers.fc(input=x,size=1,act=None)
(2)定义损失函数
此处使⽤均⽅差损失函数。
square_error_cost(input,lable):接受输⼊预测值和⽬标值,并返回⽅差估计,即为(y-y_predict)的平⽅
In[5]
cost = fluid.layers.square_error_cost(input=y_predict, label=y) #求⼀个batch的损失值
avg_cost = an(cost) #对损失值求平均值
(3)定义优化函数
此处使⽤的是随机梯度下降。
In[6]
optimizer = fluid.optimizer.SGDOptimizer(learning_rate=0.001)
opts = optimizer.minimize(avg_cost)
In[7]
test_program = fluid.default_main_program().clone(for_test=True)
在上述模型配置完毕后,得到两个fluid.Program:fluid.default_startup_program() 与fluid.default_main_program() 配置完毕了。
参数初始化操作会被写⼊fluid.default_startup_program()
fluid.default_main_program()⽤于获取默认或全局main program(主程序)。该主程序⽤于训练和测试模型。fluid.layers 中的所有layer 函数可以向 default_main_program 中添加算⼦和变量。default_main_program 是fluid的许多编程接⼝(API)的Program参数的缺省值。例如,当⽤户program没有传⼊的时候, Executor.run() 会默认执⾏ default_main_program 。
Step3.模型训练 and Step4.模型评估
(1)创建Executor
⾸先定义运算场所 fluid.CPUPlace()和 fluid.CUDAPlace(0)分别表⽰运算场所为CPU和GPU
Executor:接收传⼊的program,通过run()⽅法运⾏program。
In[8]
use_cuda = False #use_cuda为False,表⽰运算场所为CPU;use_cuda为True,表⽰运算场所为GPUfseek功能
place = fluid.CUDAPlace(0) if use_cuda else fluid.CPUPlace()
exe = fluid.Executor(place) #创建⼀个Executor实例exe
exe.run(fluid.default_startup_program()) #Executor的run()⽅法执⾏startup_program(),进⾏参数初始化
[]
(2)定义输⼊数据维度
DataFeeder负责将数据提供器(train_reader,test_reader)返回的数据转成⼀种特殊的数据结构,使其可以输⼊到Executor中。
feed_list设置向模型输⼊的向变量表或者变量表名
In[9]
# 定义输⼊数据维度
feeder = fluid.DataFeeder(place=place, feed_list=[x, y])#feed_list:向模型输⼊的变量表或变量表名
(3)定义绘制训练过程的损失值变化趋势的⽅法draw_train_process
In[10]
iter=0;
iters=[]
train_costs=[]
def draw_train_process(iters,train_costs):
title="training cost"
plt.title(title, fontsize=24)
plt.xlabel("iter", fontsize=14)
plt.ylabel("cost", fontsize=14)
plt.plot(iters, train_costs,color='red',label='training cost')
plt.show()
(4)训练并保存模型
Executor接收传⼊的program,并根据feed map(输⼊映射表)和fetch_list(结果获取表) 向program中添加feed operators(数据输⼊算⼦)和fetch operators(结果获取算⼦)。 feed map为该program提供输⼊数据。fetch_list提供program训练结束后⽤户预期的变量。
注:enumerate() 函数⽤于将⼀个可遍历的数据对象(如列表、元组或字符串)组合为⼀个索引序列,同时列出数据和数据下标,
In[11]
版权声明:本站内容均来自互联网,仅供演示用,请勿用于商业和其他非法用途。如果侵犯了您的权益请与我们联系QQ:729038198,我们将在24小时内删除。
发表评论