鸢尾花数据集可视化分析

在搜索鸢尾花数据集时看到一篇基于鸢尾花数据集的文章，其中数据可视化部分做的很好，所以自己在此复现一下，原文链接如下：https://www.jianshu.com/p/52b86c774b0b

一：导入需要的库

%matplotlib inline
import matplotlib.pyplot as plt
import numpy as np
import pandas as pd
import seaborn as sns

 

二：读取数据，查看数据集的基本信息

data = pd.read_csv('iris.data', header = None)
data.head()

查看前5行数据：

      0	  1	  2	  3	  4
0	5.1	3.5	1.4	0.2	Iris-setosa
1	4.9	3.0	1.4	0.2	Iris-setosa
2	4.7	3.2	1.3	0.2	Iris-setosa
3	4.6	3.1	1.5	0.2	Iris-setosa
4	5.0	3.6	1.4	0.2	Iris-setosa

 查看数据集的总体信息：

data.info()

<class 'pandas.core.frame.DataFrame'>
RangeIndex: 150 entries, 0 to 149
Data columns (total 5 columns):
0    150 non-null float64
1    150 non-null float64
2    150 non-null float64
3    150 non-null float64
4    150 non-null object
dtypes: float64(4), object(1)
memory usage: 5.9+ KB

四个属性值，一个标签属性，没有缺失值

 查看数据集的统计信息：

data.describe()

	0	1	2	3
count	150.000000	150.000000	150.000000	150.000000
mean	5.843333	3.054000	3.758667	1.198667
std	0.828066	0.433594	1.764420	0.763161
min	4.300000	2.000000	1.000000	0.100000
25%	5.100000	2.800000	1.600000	0.300000
50%	5.800000	3.000000	4.350000	1.300000
75%	6.400000	3.300000	5.100000	1.800000
max	7.900000	4.400000	6.900000	2.500000

 

三：绘制小提琴图，观察数据分布

设置颜色

antV = ['#1890FF', '#2FC25B', '#FACC14', '#223273', '#8543E0', '#13C2C2', '#3436C7', '#F04864']

 以特征为单位分离数据：

分离SepalLengthCm数据

data_setosa0 = data[data[4]=='Iris-setosa'][0]
data_versicolor0 = data[data[4]=='Iris-versicolor'][0]
data_virginica0 = data[data[4]=='Iris-virginica'][0]

data_SepalLengthCm = pd.DataFrame({'setosa':data_setosa0, 'versicolor':data_versicolor0, 'virginica':data_virginica0})

分离SepalWidthCm数据

data_setosa1 = data[data[4]=='Iris-setosa'][1]
data_versicolor1 = data[data[4]=='Iris-versicolor'][1]
data_virginica1 = data[data[4]=='Iris-virginica'][1]

data_SepalWidthCm = pd.DataFrame({'setosa':data_setosa1, 'versicolor':data_versicolor1, 'virginica':data_virginica1})

分离PetalLengthCm数据 

data_setosa2 = data[data[4]=='Iris-setosa'][2]
data_versicolor2 = data[data[4]=='Iris-versicolor'][2]
data_virginica2 = data[data[4]=='Iris-virginica'][2]

data_PetalLengthCm = pd.DataFrame({'setosa':data_setosa2, 'versicolor':data_versicolor2, 'virginica':data_virginica2})

分离PetalWidthCm数据

data_setosa3 = data[data[4]=='Iris-setosa'][3]
data_versicolor3 = data[data[4]=='Iris-versicolor'][3]
data_virginica3 = data[data[4]=='Iris-virginica'][3]

data_PetalWidthCm = pd.DataFrame({'setosa':data_setosa3, 'versicolor':data_versicolor3, 'virginica':data_virginica3})

 

绘制小提琴图：

f, axes = plt.subplots(2, 2, figsize = (8, 8), sharex = True)
sns.despine(left = True)
sns.violinplot(data=data_SepalLengthCm, linewidth=1, width=0.8, palette=antV, ax=axes[0, 0])
axes[0, 0].set_xlabel('Kinds')
axes[0, 0].set_ylabel('SepalLengthCm')
sns.violinplot(data=data_SepalWidthCm, linewidth=1, width=0.8, palette=antV, ax=axes[0, 1])
axes[0, 1].set_xlabel('Kinds')
axes[0, 1].set_ylabel('SepalWidthCm')
sns.violinplot(data=data_PetalLengthCm, linewidth=1, width=0.8, palette=antV, ax=axes[1, 0])
axes[1, 0].set_xlabel('Kinds')
axes[1, 0].set_ylabel('PetalLengthCm')
sns.violinplot(data=data_PetalWidthCm, linewidth=1, width=0.8, palette=antV, ax=axes[1, 1])
axes[1, 1].set_xlabel('Kinds')
axes[1, 1].set_ylabel('PetalWidthCm')

plt.show()

 

 

四：绘制 pointplot图

f, axes = plt.subplots(2, 2, figsize=(8, 8), sharex=True)
sns.despine(left = True)

sns.pointplot(data=data_SepalLengthCm, color=antV[2], ax=axes[0, 0])
axes[0, 0].set_xlabel('Kinds')
axes[0, 0].set_ylabel('SepalLengthCm')
sns.pointplot(data=data_SepalWidthCm, color=antV[2], ax=axes[0, 1])
axes[0, 1].set_xlabel('Kinds')
axes[0, 1].set_ylabel('SepalWidthCm')
sns.pointplot(data=data_PetalLengthCm, color=antV[2], ax=axes[1, 0])
axes[1, 0].set_xlabel('Kinds')
axes[1, 0].set_ylabel('PetalLengthCm')
sns.pointplot(data=data_PetalWidthCm, color=antV[2], ax=axes[1, 1])
axes[1, 1].set_xlabel('Kinds')
axes[1, 1].set_ylabel('PetalWidthCm')

plt.show()

 

 

五：绘制各个特征之间的关系矩阵

g = sns.pairplot(data=data, palette=antV, hue=4)

 

 

六：绘制Andrews Curves

使用 Andrews Curves 将每个多变量观测值转换为曲线并表示傅立叶级数的系数，这对于检测时间序列数据中的异常值很有用。

plt.subplots(figsize = (10, 8))
pd.plotting.andrews_curves(data, 4, colormap = 'cool')
plt.show()

 

 

七： 绘制热图，观察数据特征之间的先关性

fig = plt.gcf()
fig.set_size_inches(12, 8)
fig = sns.heatmap(data.corr(), annot=True, cmap='GnBu', linewidths=1, 
                  linecolor='k', square=True, mask=False, vmin=1, 
                 cbar_kws={"orientation":"vertical"}, cbar=True)