神经网络之预测共享单车使用情况

最新推荐文章于 2024-05-29 16:02:29 发布

芥末的无奈

最新推荐文章于 2024-05-29 16:02:29 发布

阅读量1w

点赞数 5

分类专栏：深度学习神经网络 python 文章标签：神经网络深度学习

本文链接：https://blog.csdn.net/weiwei9363/article/details/78274859

版权

深度学习同时被 3 个专栏收录

14 篇文章 2 订阅

订阅专栏

神经网络

5 篇文章 5 订阅

订阅专栏

python

5 篇文章 0 订阅

订阅专栏

神经网络之预测共享单车使用情况

该项目为优达学城Deep Learning Foundation Nanodegree Program的Neural Network阶段项目
原始数据和代码可以在 https://github.com/udacity/deep-learning/tree/master/first-neural-network 进行下载
本文为该项目的个人总结以及解决方案，主要在代码层面上，所有代码(运行在jupyter notebook)和数据集可以在我的下载-CNN_SharingBike中获取

整体流程

整个程序主要分为以下几块

对数据进行预处理
- 导入数据
- 处理Dummies(哑变量)，这部分可以参考 Pandas 处理 dummy variable
- 标准化数据
- 分割数据成train set, test set, validation set
建立神经网络
- 该神经网络只有一个隐层
- 利用反向传播进行权重更新
测试用例
- 确保神经网络编写正确
训练网络
- 调整参数：迭代次数，学习速率，隐层个数

接下来我们开始我们的项目，首先，导入必要的库

%matplotlib inline
%config InlineBackend.figure_format = 'retina'

import pandas as pd
import numpy as np
import matplotlib.pyplot as plt

Step 1 预处理

Reading the data
Processing the dummies
Standardizing the variables
Spliting the data into training, testing and validation sets

### 导入数据 Reading the data

data_path = 'Bike-Sharing-Dataset/hour.csv'
rides = pd.read_csv(data_path)
rides.head()

	instant	dteday	season	mnth	hr	weekday	weathersit	temp	atemp	hum	casual	registered	cnt
0	1	2011-01-01	1	1	0	6	1	0.24	0.2879	0.81	3	13	16
1	2	2011-01-01	1	1	1	6	1	0.22	0.2727	0.80	8	32	40
2	3	2011-01-01	1	1	2	6	1	0.22	0.2727	0.80	5	27	32
3	4	2011-01-01	1	1	3	6	1	0.24	0.2879	0.75	3	10	13
4	5	2011-01-01	1	1	4	6	1	0.24	0.2879	0.75	0	1	1

Plotting some data

rides[:24*10].plot(x='dteday', y='cnt')

处理哑变量 Process Dummies

More details about the dataset https://archive.ics.uci.edu/ml/datasets/Bike+Sharing+Dataset
More details about the process Pandas 处理 dummy variable

dummies_fields = ['season', 'hr', 'mnth', 'weekday', 'weathersit']

for each in dummies_fields:
    dummies = pd.get_dummies( rides.loc[:, each], prefix=each )
    rides = pd.concat( [rides, dummies], axis=1 )

drop_fields = [ 'season', 'hr', 'mnth', 'weekday', 'weathersit', 'instant', 'dteday', 'atemp' ]
data = rides.drop(drop_fields, axis=1)
data.head()

	temp	hum	casual	registered	cnt	season_1	…	weekday_6	weathersit_1
0	0.24	0.81	3	13	16	1	…	1	1
1	0.22	0.80	8	32	40	1	…	1	1
2	0.22	0.80	5	27	32	1	…	1	1
3	0.24	0.75	3	10	13	1	…	1	1
4	0.24	0.75	0	1	1	1	…	1	1

5 rows × 60 columns

标准化数据 Standardizing

将mean, std保存，以便后来数据的恢复

standard_field = ['temp', 'hum', 'windspeed', 'casual', 'registered', 'cnt']
scaled_feature = {}
for each in standard_field:
    mean, std = data[each].mean(), data[each].std()
    scaled_feature[each] = [mean, std]
    data.loc[:, each] = (data.loc[:,each] - mean) / std

data.head()

	temp	hum	windspeed	casual	registered	cnt	season_1	…	weekday_6	weathersit_1
0	-1.334609	0.947345	-1.553844	-0.662736	-0.930162	-0.956312	1	…	1	1
1	-1.438475	0.895513	-1.553844	-0.561326	-0.804632	-0.823998	1	…	1	1
2	-1.438475	0.895513	-1.553844	-0.622172	-0.837666	-0.868103	1	…	1	1
3	-1.334609	0.636351	-1.553844	-0.662736	-0.949983	-0.972851	1	…	1	1
4	-1.334609	0.636351	-1.553844	-0.723582	-1.009445	-1.039008	1	…	1	1

5 rows × 60 columns

分割数据 Spliting the data

# Save data for approximately 21 days
test_data = data[-21*24:]

# Drop the test data
data = data[:-21*24]

# Separate the data into features and targets
target_fields = ['cnt', 'casual', 'registered']

features = data.drop( target_fields, axis=1 )
targets = data.loc[:, target_fields]

test_features = test_data.drop( target_fields, axis=1 )
test_targets = test_data.loc[:, target_fields]

# Hold out the last 60 days or so of the remaining data as a validation set
train_features, train_targets = features[:-60*24], targets[:-60*24]
val_features, val_targets = features[-60*24:], targets[-60*24:]

Step 2 建立神经网络 Building the network

Only one hidden layer
The activation function between input to hidden is sigmoid function $f(x) = 1/(1 + e^{-x})$
The activation function between hiiden to output is $f(x) = x$

class NeuralNetwork(object):
    def __init__( self, input_nodes, hidden_nodes, output_nodes, learning_rate ):
        # Set the number of nodes in input, hidden and output layers
        self.input_nodes = input_nodes
        self.hidden_nodes = hidden_nodes
        self.output_nodes = output_nodes

        # Set the learning rate
        self.lr = learning_rate

        # Initialize weights
        self.weights_input_to_hidden = np.random.normal( 0.0, self.input_nodes**-0.5, (self.input_nodes, self.hidden_nodes) )
        self.weights_hidden_to_output = np.random.normal( 0.0, self.hidden_nodes**-0.5, (self.hidden_nodes, self.output_nodes) )

        # Activation function
        self.activation_function = lambda x : 1 / ( 1 + np.exp(-x) )

    def train(self, features, targets):
        '''
        Arguments
        ---------

        features: 2D array, each row is one data record, each column is a feature
        targets: 1D array of target values

        '''

        n_record = features.shape[0]
        delta_weights_i_h = np.zeros( self.weights_input_to_hidden.shape )
        delta_weights_h_o = np.zeros( self.weights_hidden_to_output.shape )

        for X, y in zip( features, targets ):
            hidden_inputs = np.dot( X, self.weights_input_to_hidden )
            hidden_outputs = self.activation_function( hidden_inputs )

            final_inputs = np.dot( hidden_outputs, self.weights_hidden_to_output )
            final_outputs = final_inputs

            error = y - final_outputs

            output_error_term = error # error * 1

            hidden_error = np.dot( self.weights_hidden_to_output, output_error_term )

            hidden_error_term = hidden_error * hidden_outputs * (1 - hidden_outputs) # f'(hidden_input)

            # Weight step
            delta_weights_i_h += hidden_error_term * X[:,None]


            delta_weights_h_o += output_error_term * hidden_outputs[:,None]

        # Update the weights
        self.weights_input_to_hidden += self.lr * delta_weights_i_h/n_record
        self.weights_hidden_to_output += self.lr * delta_weights_h_o/n_record


    def run(self, features):
        hidden_inputs = np.dot( features, self.weights_input_to_hidden )
        hidden_output = self.activation_function( hidden_inputs )

        final_inputs = np.dot( hidden_output, self.weights_hidden_to_output )
        final_outputs = final_inputs

        return final_outputs

def MSE(y, Y):
    return np.mean( (y - Y)**2 )

Step 3 单元测试 Unit tests

import unittest

inputs = np.array([[0.5, -0.2, 0.1]])
targets = np.array([[0.4]])
test_w_i_h = np.array([[0.1, -0.2],
                       [0.4, 0.5],
                       [-0.3, 0.2]])
test_w_h_o = np.array([[0.3],
                       [-0.1]])

class TestMethods(unittest.TestCase):

    ##########
    # Unit tests for data loading
    ##########

    def test_data_path(self):
        # Test that file path to dataset has been unaltered
        self.assertTrue(data_path.lower() == 'bike-sharing-dataset/hour.csv')

    def test_data_loaded(self):
        # Test that data frame loaded
        self.assertTrue(isinstance(rides, pd.DataFrame))

    ##########
    # Unit tests for network functionality
    ##########

    def test_activation(self):
        network = NeuralNetwork(3, 2, 1, 0.5)
        # Test that the activation function is a sigmoid
        self.assertTrue(np.all(network.activation_function(0.5) == 1/(1+np.exp(-0.5))))

    def test_train(self):
        # Test that weights are updated correctly on training
        network = NeuralNetwork(3, 2, 1, 0.5)
        network.weights_input_to_hidden = test_w_i_h.copy()
        network.weights_hidden_to_output = test_w_h_o.copy()

        network.train(inputs, targets)

        self.assertTrue(np.allclose(network.weights_hidden_to_output, 
                                    np.array([[ 0.37275328], 
                                              [-0.03172939]])))
       # print(network.weights_input_to_hidden)
        self.assertTrue(np.allclose(network.weights_input_to_hidden,
                                    np.array([[ 0.10562014, -0.20185996], 
                                              [0.39775194, 0.50074398], 
                                              [-0.29887597, 0.19962801]])))

    def test_run(self):
        # Test correctness of run method
        network = NeuralNetwork(3, 2, 1, 0.5)
        network.weights_input_to_hidden = test_w_i_h.copy()
        network.weights_hidden_to_output = test_w_h_o.copy()

        self.assertTrue(np.allclose(network.run(inputs), 0.09998924))

suite = unittest.TestLoader().loadTestsFromModule(TestMethods())
unittest.TextTestRunner().run(suite)

.....
----------------------------------------------------------------------
Ran 5 tests in 0.016s

OK





<unittest.runner.TextTestResult run=5 errors=0 failures=0>

Step 4 训练网络 Trainning the network

Choose the number of iterations
Choose the learning rate
Choose the number of hidden nodes

import sys

### Set the hyperparameters here ###
iterations = 2000
learning_rate = 0.8
hidden_nodes = 12
output_nodes = 1

N_i = train_features.shape[1]
network = NeuralNetwork(N_i, hidden_nodes, output_nodes, learning_rate)

losses = {'train':[], 'validation':[]}
for ii in range(iterations):
    # Go through a random batch of 128 records from the training data set
    batch = np.random.choice(train_features.index, size=128)
    X, y = train_features.iloc[batch].values, train_targets.iloc[batch]['cnt']

    network.train(X, y)

    # Printing out the training progress
    train_loss = MSE(network.run(train_features).T, train_targets['cnt'].values)
    val_loss = MSE(network.run(val_features).T, val_targets['cnt'].values)

    sys.stdout.write("\rProgress: {:2.1f}".format(100 * ii/float(iterations)) \
                     + "% ... Training loss: " + str(train_loss)[:5] \
                     + " ... Validation loss: " + str(val_loss)[:5])
    sys.stdout.flush()

    losses['train'].append(train_loss)
    losses['validation'].append(val_loss)

Progress: 100.0% ... Training loss: 0.073 ... Validation loss: 0.154

plt.plot(losses['train'], label='Traning loss')
plt.plot(losses['validation'], label='Validation loos')
plt.legend()

<matplotlib.legend.Legend at 0x2282dc0a2e8>

这里写图片描述

Check out your predication

fig, ax = plt.subplots(figsize=(8,4))

mean, std = scaled_feature['cnt']
predictions = network.run(test_features).T*std + mean
ax.plot(predictions[0], label='Prediction')
ax.plot((test_targets['cnt']*std + mean).values, label='Data')
ax.set_xlim(right=len(predictions))
ax.legend()

dates = pd.to_datetime(rides.loc[test_data.index]['dteday'])
dates = dates.apply(lambda d: d.strftime('%b %d'))
ax.set_xticks(np.arange(len(dates))[12::24])
_ = ax.set_xticklabels(dates[12::24], rotation=45)

prediction

芥末的无奈

关注

5
点赞
踩
57

收藏

觉得还不错? 一键收藏
15
评论
神经网络之预测共享单车使用情况

神经网络之预测共享单车使用情况该项目为优达学城Deep Learning Foundation Nanodegree Program的Neural Network阶段项目原始数据和代码可以在 https://github.com/udacity/deep-learning/tree/master/first-neural-network 进行下载本文为该项目的个人总结以及解决方案，主要在代码层
复制链接

扫一扫

专栏目录