In [6]:

    

%matplotlib inline
from matplotlib import pylab as plt
plt.rcParams['figure.figsize'] = (15.0, 10.0)
import pandas as pd
import seaborn as sns


    

In [4]:

    

data = pd.read_csv("978-3-319-12065-2/chapter-4/teams.csv")
data.head()


    

    
Out[4]:






  

    

      

      
team
      
code
      
league
      
division
      
games
      
wins
      
losses
      
pct
      
payroll
    
  
  

    

      
0
      
Arizona Diamondbacks
      
ARI
      
NL
      
West
      
162
      
81
      
81
      
0.500
      
67069833
    
    

      
1
      
Atlanta Braves
      
ATL
      
NL
      
East
      
162
      
94
      
68
      
0.580
      
86208000
    
    

      
2
      
Baltimore Orioles
      
BAL
      
AL
      
East
      
162
      
93
      
69
      
0.574
      
76704000
    
    

      
3
      
Boston Red Sox
      
BOS
      
AL
      
East
      
162
      
69
      
93
      
0.426
      
110386000
    
    

      
4
      
Chicago Cubs
      
CHC
      
NL
      
Central
      
162
      
61
      
101
      
0.377
      
80422700
    
  

In [19]:

    

data.plot(kind='scatter',x="payroll",y="wins")


    

    
Out[19]:





<matplotlib.axes._subplots.AxesSubplot at 0x7fd8e97f57d0>






    

In [18]:

    

sns.regplot("payroll","wins",data,fit_reg=False)


    

    
Out[18]:





<matplotlib.axes._subplots.AxesSubplot at 0x7fd8e96d7e50>






    

In [22]:

    

def label_point_orig(x, y, val, ax):
    a = pd.concat({'x': x, 'y': y, 'val': val}, axis=1)
    for i, point in a.iterrows():
        ax.text(point['x'], point['y'], str(point['val']))

ax = data.plot(kind='scatter',x="payroll",y="wins")

label_point_orig(data.payroll,data.wins,data.code,ax)


    

In [23]:

    

ax1 = sns.regplot("payroll","wins",data,fit_reg=False)
label_point_orig(data.payroll,data.wins,data.code,ax1)


    

In [37]:

    

data_al = data[data.league == "AL"]
data_nl = data[data.league == "NL"]
fig, ax = plt.subplots()
ax.plot(data_al.payroll, data_al.wins, marker='o',linestyle='',label="AL")
ax.plot(data_nl.payroll, data_nl.wins, marker='s',linestyle='',label="NL")
ax.legend()
ax.set_xlabel("payroll")
ax.set_ylabel("wins")


    

    
Out[37]:





<matplotlib.text.Text at 0x7fd8e912eb10>






    

In [35]:

    

data_groups = data.groupby('league')
markers = ["o","s"]
mindx = 0

fig, ax = plt.subplots()
for name, group in data_groups:
    ax.plot(group.payroll, group.wins, marker=markers[mindx], linestyle='', ms=12, label=name)
    mindx += 1
ax.legend()
ax.set_xlabel("payroll")
ax.set_ylabel("wins")


    

    
Out[35]:





<matplotlib.text.Text at 0x7fd8e91051d0>






    

In [58]:

    

data_gf_al = data[data.pct >= 0.5 ]
data_gf_nl = data[data.pct <= 0.5]
fig, ax = plt.subplots()
ax.plot(data_gf_al.payroll,data_gf_al.wins,marker='o',linestyle='')
ax.plot(data_gf_nl.payroll,data_gf_nl.wins,marker='s',linestyle='')
ax.plot([data.payroll.min(),data.payroll.max()],[81,81])


    

    
Out[58]:





[<matplotlib.lines.Line2D at 0x7fd8e2ba9ed0>]






    

In [64]:

    

data_groups = data.groupby('league')
markers = ["o","^"]
mindx = 0

fig, ax = plt.subplots()
for name, group in data_groups:
    ax.plot(group.payroll, group.wins, marker=markers[mindx], linestyle='', ms=12, label=name)
    mindx += 1
ax.legend()
ax.set_xlabel("payroll")
ax.set_ylabel("wins")


    

    
Out[64]:





<matplotlib.text.Text at 0x7fd8e2cbc9d0>






    

In [65]:

    

data[data.league == "NL"].payroll.sum()


    

    
Out[65]:





1512099665

In [66]:

    

data.groupby("league").payroll.sum()


    

    
Out[66]:





league
AL    1424254675
NL    1512099665
Name: payroll, dtype: int64

In [68]:

    

data.groupby(["league","division"]).payroll.sum()


    

    
Out[68]:





league  division
AL      Central     489326375
        East        530789300
        West        404139000
NL      Central     476248200
        East        547594982
        West        488256483
Name: payroll, dtype: int64

In [71]:

    

data.groupby("league").payroll.sum().plot(kind='bar')


    

    
Out[71]:





<matplotlib.axes._subplots.AxesSubplot at 0x7fd8e92c1990>






    

In [109]:

    

data.groupby(["league","division"]).payroll.sum().plot(kind='bar',stacked=True)


    

    
Out[109]:





<matplotlib.axes._subplots.AxesSubplot at 0x7fd8e2117610>






    

In [116]:

    

data.groupby(["league","division"]).payroll.sum().unstack("division").plot(kind='bar',stacked=True)


    

    
Out[116]:





<matplotlib.axes._subplots.AxesSubplot at 0x7fd8e23f0110>






    

In [120]:

    

data.groupby(["league","division"]).payroll.sum().unstack("division").plot(kind='bar')


    

    
Out[120]:





<matplotlib.axes._subplots.AxesSubplot at 0x7fd8e1e78410>






    

In [121]:

    

data.groupby(["league"]).payroll.sum().plot(kind='pie')


    

    
Out[121]:





<matplotlib.axes._subplots.AxesSubplot at 0x7fd8e1d60350>






    

In [7]:

    

data.groupby(["league","division"]).payroll.sum().plot(kind='pie')


    

    
Out[7]:





<matplotlib.axes._subplots.AxesSubplot at 0x7fc3037836d0>






    

In [10]:

    

data_al = data[data.league == "AL"]
data_nl = data[data.league == "NL"]
fig, ax = plt.subplots(2)
ax[0].plot(data_al.payroll, data_al.wins, marker='o', linestyle='', ms=12, label="AL")
ax[1].plot(data_nl.payroll, data_nl.wins, marker='o', linestyle='', ms=12, label="NL")


    

    
Out[10]:





[<matplotlib.lines.Line2D at 0x7fc302fe5d10>]






    

In [21]:

    

from bokeh.sampledata.iris import flowers
from bokeh.plotting import figure, show, output_file,output_notebook

output_notebook()
colormap = {'AL': 'red', 'NL': 'green'}
data['color'] = data['league'].map(lambda x: colormap[x])
p = figure(title = "Payroll vs Wins")
p.xaxis.axis_label = 'Wins'
p.yaxis.axis_label = 'Payroll'
p.circle(data.payroll,data.wins,color=data["color"],size=10)
show(p)