PaddlePaddle一步一步預測波士頓房價
如何開始
針對PaddlePaddle新手入門的閱讀順序，本人建議先閱讀Fluid編程指南，然後再看快速入門，在針對上一篇搭建簡單網絡之後，本次爲大家解讀一下快速入門-線性迴歸，個人感覺目錄的名有點問題，線性迴歸應該改爲房價預測，才能與數字識別對應上，哈哈。
現實問題
房價是大家都關心的事情，但是房價由哪些因素決定，在已知一些因素之後房價大概是多少？我們需要使用PaddlePaddle解決線性迴歸的問題。
轉換爲PaddlePaddle的問題
加載房價數據，使用PaddlePaddle定義全連接網絡（線性迴歸模型），加載數據到網絡中進行學習，通過前向計算和反向傳播後（PaddlePaddle已封裝好），得到優化後參數，給定一些房子相關信息，加載優化後的參數，得到預測的房價。
使用PyCharm創建工程、顯示數據
創建InferBostonHousingPrice工程，使用之前已有的環境，新建train.py文件測試paddlepaddle是否可用。
import paddle.fluid
顯示波士頓郊區房價數據
train_reader = paddle.dataset.uci_housing.train()
test_reader = paddle.dataset.uci_housing.test()
print(paddle.dataset.uci_housing.feature_names)
print("-----------train---------------")
for i, data in enumerate(train_reader()):
print(i, data)
print("-----------test---------------")
for i, data in enumerate(test_reader()):
print(i, data)
一共506個數據，404個訓練數據，102個測試數據。
原版的說明文件大家可以參考一下：
-
Title: Boston Housing Data
-
Sources:
(a) Origin: This dataset was taken from the StatLib library which is
maintained at Carnegie Mellon University.
(b) Creator: Harrison, D. and Rubinfeld, D.L. ‘Hedonic prices and the
demand for clean air’, J. Environ. Economics & Management,
vol.5, 81-102, 1978.
© Date: July 7, 1993 -
Past Usage:
- Used in Belsley, Kuh & Welsch, ‘Regression diagnostics …’, Wiley,
1980. N.B. Various transformations are used in the table on
pages 244-261. - Quinlan,R. (1993). Combining Instance-Based and Model-Based Learning.
In Proceedings on the Tenth International Conference of Machine
Learning, 236-243, University of Massachusetts, Amherst. Morgan
Kaufmann.
- Used in Belsley, Kuh & Welsch, ‘Regression diagnostics …’, Wiley,
-
Relevant Information:
Concerns housing values in suburbs of Boston.
-
Number of Instances: 506
-
Number of Attributes: 13 continuous attributes (including “class”
attribute “MEDV”), 1 binary-valued attribute. -
Attribute Information:
- CRIM per capita crime rate by town
- ZN proportion of residential land zoned for lots over
25,000 sq.ft. - INDUS proportion of non-retail business acres per town
- CHAS Charles River dummy variable (= 1 if tract bounds
river; 0 otherwise) - NOX nitric oxides concentration (parts per 10 million)
- RM average number of rooms per dwelling
- AGE proportion of owner-occupied units built prior to 1940
- DIS weighted distances to five Boston employment centres
- RAD index of accessibility to radial highways
- TAX full-value property-tax rate per $10,000
- PTRATIO pupil-teacher ratio by town
- B 1000(Bk - 0.63)^2 where Bk is the proportion of blacks
by town - LSTAT % lower status of the population
- MEDV Median value of owner-occupied homes in $1000’s
-
Missing Attribute Values: None.
定義網絡
# 將數據轉換爲feed所需的數據類型
list = []
for data in train_reader():
list.append(data)
# 定義網絡
x = fluid.layers.data(name="x", shape=[13], dtype='float32')
y = fluid.layers.data(name="y", shape=[1], dtype='float32')
y_predict = fluid.layers.fc(input=x, size=1, act=None)
# 定義損失函數
cost = fluid.layers.square_error_cost(input=y_predict, label=y)
avg_cost = fluid.layers.mean(cost)
sgd_optimizer = fluid.optimizer.SGD(learning_rate=0.001)
sgd_optimizer.minimize(avg_cost)
加載數據並進行訓練
feeder = fluid.DataFeeder(place=cpu, feed_list=[x, y])
for i in range(1000):
avg_loss_value = exe.run(
None,
feed=feeder.feed(list),
fetch_list=[avg_cost])
參數保存到文件
model_path = "./infer_bhp_model"
fluid.io.save_params(executor=exe, dirname=model_path, main_program=None)
選擇預測所需的數據
在原數據集中任意選擇幾個樣本，作爲模擬房價的13個變量的值
# 定義數據
new_x = numpy.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]]).astype('float32')
#24
new_x1 = numpy.array([[-0.03894423, -0.11363636, -0.0944567 ,
-0.06916996, -0.07138901,0.08476061,
0.11663336, -0.09250268, -0.19780031,
-0.04625411,0.26004962, 0.09603603, -0.08921256]]).astype('float32')
#27.5
new_x2 = numpy.array([[-0.01460746, -0.11363636, 0.30950225,
-0.06916996, 0.10350811,-0.07753102,
0.29583006, -0.12788538, -0.19780031,
-0.00999457,-0.39952485, -0.02154428, -0.01719269]]).astype('float32')
#19.1
定義網絡進行預測
# 定義網絡
x = fluid.layers.data(name="x", shape=[13], dtype='float32')
y = fluid.layers.data(name="y", shape=[1], dtype='float32')
y_predict = fluid.layers.fc(input=x, size=1, act=None)
# 參數初始化
cpu = fluid.core.CPUPlace()
exe = fluid.Executor(cpu)
param_path = "./infer_bhp_model"
prog = fluid.default_main_program()
fluid.io.load_params(executor=exe, dirname=param_path,
main_program=prog)
outs = exe.run(
feed={'x': new_x2},
fetch_list=[y_predict.name])
print(outs)
完整代碼
import paddle
import paddle.fluid as fluid
import numpy
# 顯示數據
train_reader = paddle.dataset.uci_housing.train()
test_reader = paddle.dataset.uci_housing.test()
print(paddle.dataset.uci_housing.feature_names)
print("-----------train---------------")
for i, data in enumerate(train_reader()):
print(i, data)
print("-----------test---------------")
for i, data in enumerate(test_reader()):
print(i, data)
#
list = []
for data in train_reader():
list.append(data)
# 定義網絡
x = fluid.layers.data(name="x", shape=[13], dtype='float32')
y = fluid.layers.data(name="y", shape=[1], dtype='float32')
y_predict = fluid.layers.fc(input=x, size=1, act=None)
# 定義損失函數
cost = fluid.layers.square_error_cost(input=y_predict, label=y)
avg_cost = fluid.layers.mean(cost)
sgd_optimizer = fluid.optimizer.SGD(learning_rate=0.001)
sgd_optimizer.minimize(avg_cost)
# 參數初始化
cpu = fluid.core.CPUPlace()
exe = fluid.Executor(cpu)
exe.run(fluid.default_startup_program())
feeder = fluid.DataFeeder(place=cpu, feed_list=[x, y])
for i in range(1000):
avg_loss_value = exe.run(
None,
feed=feeder.feed(list),
fetch_list=[avg_cost])
print(avg_loss_value)
model_path = "./infer_bhp_model"
fluid.io.save_params(executor=exe, dirname=model_path, main_program=None)
import paddle.fluid as fluid
import numpy
# 定義數據
new_x = numpy.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]]).astype('float32')
#24
new_x1 = numpy.array([[-0.03894423, -0.11363636, -0.0944567 ,
-0.06916996, -0.07138901,0.08476061,
0.11663336, -0.09250268, -0.19780031,
-0.04625411,0.26004962, 0.09603603, -0.08921256]]).astype('float32')
#27.5
new_x2 = numpy.array([[-0.01460746, -0.11363636, 0.30950225,
-0.06916996, 0.10350811,-0.07753102,
0.29583006, -0.12788538, -0.19780031,
-0.00999457,-0.39952485, -0.02154428, -0.01719269]]).astype('float32')
#19.1
# 定義網絡
x = fluid.layers.data(name="x", shape=[13], dtype='float32')
y = fluid.layers.data(name="y", shape=[1], dtype='float32')
y_predict = fluid.layers.fc(input=x, size=1, act=None)
# 參數初始化
cpu = fluid.core.CPUPlace()
exe = fluid.Executor(cpu)
param_path = "./infer_bhp_model"
prog = fluid.default_main_program()
fluid.io.load_params(executor=exe, dirname=param_path,
main_program=prog)
outs = exe.run(
feed={'x': new_x2},
fetch_list=[y_predict.name])
print(outs)
結果：
