[Python] Making a package

In this next section, you'll convert the Distributions code into a Python package. A package is a collection of Python modules. Although the previous code might already seem like it was a Python package because it contained multiple files, a Python package also needs an __init__.py file. In this section, you'll learn how to create this __init__.py file and then pip install the package into your local Python installation.

Folder structure:

|- distribution

　　| - __init__.py

　　| - Gaussiandistribution.py

　　| - Generaldistribution.py

|- setup.py

Gaussiandistribution.py:

100

101

102

103

104

105

106

107

108

109

110

111

112

113

114

115

116

117

118

119

120

121

122

123

124

125

126

127

128

129

130

131

132

133

134

135

136

137

138

139

140

141

142

143

144

145

146

147

148

149

150

151

152

153

154

155

156

157

158

159

160

161

162

163

164

165

166

167

168

169

170

171

172

173

174

175

176

177

178

179

180

181

182

183

184

185

186

187

188

189

190

191

192

193

194

195

196

197

198

199

200

201

202

203

204

205

206

import math
import matplotlib.pyplot as plt
from .Generaldistribution import Distribution # need a . for python 3
 
class Gaussian(Distribution):
    """ Gaussian distribution class for calculating and 
    visualizing a Gaussian distribution.
     
    Attributes:
        mean (float) representing the mean value of the distribution
        stdev (float) representing the standard deviation of the distribution
        data_list (list of floats) a list of floats extracted from the data file
             
    """
    def __init__(self, mu=0, sigma=1):
         
        Distribution.__init__(self, mu, sigma)
 
     
    def calculate_mean(self):
     
        """Function to calculate the mean of the data set.
         
        Args: 
            None
         
        Returns: 
            float: mean of the data set
     
        """
                     
        avg = 1.0 * sum(self.data) / len(self.data)
         
        self.mean = avg
         
        return self.mean
 
 
 
    def calculate_stdev(self, sample=True):
 
        """Function to calculate the standard deviation of the data set.
         
        Args: 
            sample (bool): whether the data represents a sample or population
         
        Returns: 
            float: standard deviation of the data set
     
        """
 
        if sample:
            n = len(self.data) - 1
        else:
            n = len(self.data)
     
        mean = self.mean
     
        sigma = 0
     
        for d in self.data:
            sigma += (d - mean) ** 2
         
        sigma = math.sqrt(sigma / n)
     
        self.stdev = sigma
         
        return self.stdev
         
 
    def read_data_file(self, file_name, sample=True):
     
        """Function to read in data from a txt file. The txt file should have
        one number (float) per line. The numbers are stored in the data attribute. 
        After reading in the file, the mean and standard deviation are calculated
                 
        Args:
            file_name (string): name of a file to read from
         
        Returns:
            None
         
        """
             
        with open(file_name) as file:
            data_list = []
            line = file.readline()
            while line:
                data_list.append(int(line))
                line = file.readline()
        file.close()
     
        self.data = data_list
        self.mean = self.calculate_mean()
        self.stdev = self.calculate_stdev(sample)
         
         
    def plot_histogram(self):
        """Function to output a histogram of the instance variable data using 
        matplotlib pyplot library.
         
        Args:
            None
             
        Returns:
            None
        """
        plt.hist(self.data)
        plt.title('Histogram of Data')
        plt.xlabel('data')
        plt.ylabel('count')
         
         
         
    def pdf(self, x):
        """Probability density function calculator for the gaussian distribution.
         
        Args:
            x (float): point for calculating the probability density function
             
         
        Returns:
            float: probability density function output
        """
         
        return (1.0 / (self.stdev * math.sqrt(2*math.pi))) * math.exp(-0.5*((x - self.mean) / self.stdev) ** 2)
         
 
    def plot_histogram_pdf(self, n_spaces = 50):
 
        """Function to plot the normalized histogram of the data and a plot of the 
        probability density function along the same range
         
        Args:
            n_spaces (int): number of data points 
         
        Returns:
            list: x values for the pdf plot
            list: y values for the pdf plot
             
        """
         
        mu = self.mean
        sigma = self.stdev
 
        min_range = min(self.data)
        max_range = max(self.data)
         
         # calculates the interval between x values
        interval = 1.0 * (max_range - min_range) / n_spaces
 
        x = []
        y = []
         
        # calculate the x values to visualize
        for i in range(n_spaces):
            tmp = min_range + interval*i
            x.append(tmp)
            y.append(self.pdf(tmp))
 
        # make the plots
        fig, axes = plt.subplots(2,sharex=True)
        fig.subplots_adjust(hspace=.5)
        axes[0].hist(self.data, density=True)
        axes[0].set_title('Normed Histogram of Data')
        axes[0].set_ylabel('Density')
 
        axes[1].plot(x, y)
        axes[1].set_title('Normal Distribution for \n Sample Mean and Sample Standard Deviation')
        axes[0].set_ylabel('Density')
        plt.show()
 
        return x, y
         
    def __add__(self, other):
         
        """Function to add together two Gaussian distributions
         
        Args:
            other (Gaussian): Gaussian instance
             
        Returns:
            Gaussian: Gaussian distribution
             
        """
         
        result = Gaussian()
        result.mean = self.mean + other.mean
        result.stdev = math.sqrt(self.stdev ** 2 + other.stdev ** 2)
         
        return result
         
         
    def __repr__(self):
     
        """Function to output the characteristics of the Gaussian instance
         
        Args:
            None
         
        Returns:
            string: characteristics of the Gaussian
         
        """
         
        return "mean {}, standard deviation {}".format(self.mean, self.stdev)

Generaldistribution.py

class Distribution:
     
    def __init__(self, mu=0, sigma=1):
     
        """ Generic distribution class for calculating and 
        visualizing a probability distribution.
     
        Attributes:
            mean (float) representing the mean value of the distribution
            stdev (float) representing the standard deviation of the distribution
            data_list (list of floats) a list of floats extracted from the data file
            """
         
        self.mean = mu
        self.stdev = sigma
        self.data = []
 
 
    def read_data_file(self, file_name):
     
        """Function to read in data from a txt file. The txt file should have
        one number (float) per line. The numbers are stored in the data attribute.
                 
        Args:
            file_name (string): name of a file to read from
         
        Returns:
            None
         
        """
             
        with open(file_name) as file:
            data_list = []
            line = file.readline()
            while line:
                data_list.append(int(line))
                line = file.readline()
        file.close()
     
        self.data = data_list

__init__.py:

from .Gaussiandistribution import Gaussian

setup.py:

from setuptools import setup

setup(name='distribution',
version='1.0',
description='Gaussian distributions',
packages=['distributions'],
zip_safe=False)