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

plt.figure(figsize=(9,5), # (寬度 , 高度) 單位inch
dpi=120, # 清晰度 dot-per-inch
# facecolor='#CCCCCC', # 畫(huà)布底色
# edgecolor='black',linewidth=0.2,frameon=True, # 畫(huà)布邊框
#frameon=False # 不要畫(huà)布邊框
)
# 設(shè)置全局中文字體
plt.rcParams['font.sans-serif'] = 'KaiTi' # 設(shè)置全局字體為中文 楷體
plt.rcParams['axes.unicode_minus'] = False # 不使用中文減號(hào)
#讀取數(shù)據(jù)
crime=pd.read_csv("crimeRatesByState2005.csv")
print (list(crime.murder))#轉(zhuǎn)化成列表
#刪除state為United States的數(shù)據(jù)
crime2 = crime[crime.state != "United States"]
#刪除state為District of Columbia的數(shù)據(jù)
crime2 = crime2[crime2.state != "District of Columbia" ]
z = list(crime2.population/10000)#取人口數(shù)據(jù)
#colors = np.random.rand(len(list(crime2.murder)))#根據(jù)謀殺率隨機(jī)去顏色
cm = plt.cm.get_cmap('RdYlBu')#使用色譜RdYlBu
plt.scatter(list(crime2.murder), list(crime2.burglary), s=z,c=z,cmap = cm, linewidth = 0.5, alpha = 0.5)#繪制散點(diǎn)圖
plt.xlabel("murder")
plt.ylabel("burglary")
plt.show()

# 讀取數(shù)據(jù)
df = pd.read_csv('iris.csv')
# 平面坐標(biāo)系的位置只能表示2維數(shù)據(jù)
x = df['sepal_length']
y = df['sepal_width']
# 根據(jù)X，Y值畫(huà)散點(diǎn)圖
plt.scatter(x,y)

# 讀取數(shù)據(jù)
df = pd.read_csv('iris.csv')
# 平面坐標(biāo)系的位置只能表示2維數(shù)據(jù)
x = df['sepal_length']
y = df['sepal_width']
c = df['species'].map({'setosa':'r','versicolor':'g','virginica':'b'})
# 根據(jù)X，Y值畫(huà)散點(diǎn)圖, 用不同的顏色標(biāo)識(shí)不同的分類
plt.scatter(x,y, c=c)

# 讀取數(shù)據(jù)
df = pd.read_csv('iris.csv')
# 平面坐標(biāo)系的位置只能表示2維數(shù)據(jù)
x = df['sepal_length']
y = df['sepal_width']
c = df['petal_length']
# 根據(jù)X，Y值畫(huà)散點(diǎn)圖, 用顏色的深淺表示花萼的長(zhǎng)度
plt.scatter(x,y, c=c, cmap=plt.cm.RdYlBu)

# 讀取數(shù)據(jù)
df = pd.read_csv('iris.csv')
# 平面坐標(biāo)系的位置只能表示2維數(shù)據(jù)
x = df['sepal_length'] # x 軸坐標(biāo)
y = df['sepal_width'] # y 軸坐標(biāo)
c = df['petal_length'] # 顏色color
s = df['petal_width'] # 大小size
# 根據(jù)X，Y值畫(huà)散點(diǎn)圖, 用顏色的深淺表示花萼的長(zhǎng)度，用大小表示花萼的寬度
plt.figure(figsize=(5,5),dpi=100)
#plt.scatter(x,y, c=c, s=50) # 可以是標(biāo)量，那么所有的點(diǎn)都一樣
plt.scatter(x,y, c=c, s=s*30)

# 讀取數(shù)據(jù)
df = pd.read_csv('iris.csv')
def get_xycs(df):
# 平面坐標(biāo)系的位置只能表示2維數(shù)據(jù)
x = df['sepal_length'] # x 軸坐標(biāo)
y = df['sepal_width'] # y 軸坐標(biāo)
c = df['petal_length'] # 顏色color
s = df['petal_width'] # 大小size
return x,y,c,s
markers = {'setosa':'o', 'versicolor':'D', 'virginica':'*'}
# 根據(jù)X，Y值畫(huà)散點(diǎn)圖, 用顏色的深淺表示花萼的長(zhǎng)度，用大小表示花萼的寬度, 每組數(shù)據(jù)只能是一種點(diǎn)樣式
plt.figure(figsize=(5,5),dpi=100)
#plt.scatter(x,y, c=c, s=50) # 可以是標(biāo)量，那么所有的點(diǎn)都一樣
for sp in df['species'].unique():
x,y,c,s = get_xycs(df[df['species']==sp])
plt.scatter(x,y, c=c, s=s*30, cmap=plt.cm.seismic, marker=markers[sp],label=sp)
plt.legend()

# 讀取數(shù)據(jù)
df = pd.read_csv('iris.csv')
def get_xycs(df):
# 平面坐標(biāo)系的位置只能表示2維數(shù)據(jù)
x = df['sepal_length'] # x 軸坐標(biāo)
y = df['sepal_width'] # y 軸坐標(biāo)
c = df['petal_length'] # 顏色color
s = df['petal_width'] # 大小size
return x,y,c,s
markers = {'setosa':'o', 'versicolor':'D', 'virginica':'*'}
# 根據(jù)X，Y值畫(huà)散點(diǎn)圖, 用顏色的深淺表示花萼的長(zhǎng)度，用大小表示花萼的寬度, 每組數(shù)據(jù)只能是一種點(diǎn)樣式
plt.figure(figsize=(5,5),dpi=100)
#plt.scatter(x,y, c=c, s=50) # 可以是標(biāo)量，那么所有的點(diǎn)都一樣
for sp in df['species'].unique():
x,y,c,s = get_xycs(df[df['species']==sp])
plt.scatter(x,y, s=s*30, cmap=plt.cm.seismic, marker=markers[sp],label=sp)
plt.legend()

### 在二維坐標(biāo)系上，位置表示(x,y)二維數(shù)據(jù)
x = df.sepal_length # x 表示花瓣長(zhǎng)
y = df.sepal_width # y 表示花瓣寬
s = (df.petal_length * df.petal_width)*np.pi # s(size) 表示花萼面積
c = (df.petal_length * df.petal_width)*np.pi
plt.scatter(x,y,s=s*5, c=c,cmap=plt.cm.RdYlBu_r)
plt.xlabel('sepal_length')
plt.ylabel('sepal_width')

# 在二維坐標(biāo)系上，位置表示(x,y)二維數(shù)據(jù)
x = df.sepal_length # x 表示花瓣長(zhǎng)
y = df.sepal_width # y 表示花瓣寬
s = (df.petal_length * df.petal_width)*np.pi # s(size) 表示花萼面積
#print(df.species)
#colormap = {"setosa":"#FF0000", "versicolor":"green", "virginica":"b"} # 定義一個(gè)字典將species字符串映射到顏色字符串上
colormap = {"setosa":1, "versicolor":5, "virginica":6} # 定義一個(gè)字典將species字符串映射到顏色字符串上
c = df.species.map(colormap)
#print(c)
plt.scatter(x,y,s=s*5, c=c,cmap=plt.cm.coolwarm, alpha=0.7, edgecolors='face')
plt.xlabel('sepal_length')
plt.ylabel('sepal_width')

plt.scatter(df['burglary'], df['larceny_theft'],
s=df['population']*2e-5,
c=df['motor_vehicle_theft'], cmap=plt.cm.coolwarm,
edgecolors='b',
alpha=0.75)

for idx,statename in df['state'].items():
plt.text(x=df['burglary'][idx],y=df['larceny_theft'][idx]-df['population'][idx]*2e-5*0.5,s=statename,fontsize=6,ha='center',va='top')

df.plot.scatter(x='burglary',y='larceny_theft',c='motor_vehicle_theft',cmap=plt.cm.coolwarm,s=df['population']*2e-5)
for i in df.index:
if i in top5_motor_theft_index: # 偷車賊最多的5個(gè)州
plt.text(df.loc[i,'burglary']+10, df.loc[i,'larceny_theft']-10, df.loc[i,'state'], color='red') # 一個(gè)文本框