import pandas as pd
import numpy as np
import matplotlib.pyplot as plt
import seaborn as sns
import re
import os
from PIL import Image
from pyecharts.charts import Pie
from pyecharts import options as opts
from pyecharts.globals import CurrentConfig, NotebookType
CurrentConfig.NOTEBOOK_TYPE = NotebookType.JUPYTER_LAB
#%matplotlib inline
pd.set_option("display.max_column",None)
sns.set_style('white',{'font.sans-serif':['simhei','Arial']})
df_season = pd.read_pickle("./season.pkl")
df_season.rename(columns={"出手":"总出手","命中":"总命中","命中.1":"三分命中","出手.1":"三分出手","命中.2":"罚球命中","出手.2":"罚球出手"},
inplace=True)
df_season.head()
print('df_season.head()',df_season.head())
#print('df_season.head()',df_season)
season = list(np.sort(df_season["赛季"].unique()))
season = season[2:] + season[:2]
df_season["win"] = df_season["果"].map({"胜": 1, "负": 0})
wining_rate = df_season.groupby(["赛季"])["win"].mean()[season].values
fig,ax = plt.subplots(figsize=(12,8))
fig.text(x=0.1, y=0.9, s=' 常规赛的场均得分以及胜率 ',
fontsize=32, weight='bold', color='white', backgroundcolor='#c5b783')
sns.barplot(x="赛季",y="得分",data=df_season,order=season,ci=None,color="#3c7f99")
plt.tick_params(labelsize=16,axis="both",direction="in")
plt.ylabel("场均得分",fontsize=14)
plt.xlabel("赛季",fontsize=14)
ax2=ax.twinx()
plt.scatter(x=season,y=wining_rate,color="r",marker='d')
plt.plot(season,wining_rate,color="r")
plt.tick_params(labelsize=16,axis="both",direction="in")
ax2.set_yticks([0.2,0.3,0.4,0.5,0.6,0.7,0.8,0.9])
plt.ylim(0.2,1)
plt.ylabel("胜率",fontsize=14)
plt.show()
#单场比赛各个得分段占比
#使用pie绘制饼图
interval = ["20分及以下","20-30","30-40","40-50","大于50"]
value = pd.cut(df_season["得分"],bins=[0,20,30,40,50,100],labels=interval).value_counts(normalize=True)[interval].values
pie = (
Pie()
.add(
"",
[list(z) for z in zip(interval, value)],
center=["50%", "50%"],
# radius=["50%","75%"]
)
.set_global_opts(
title_opts=opts.TitleOpts(title="单场比赛各个得分段占比",pos_left="center"),
legend_opts=opts.LegendOpts(
is_show=True
)
)
.set_series_opts(label_opts=opts.LabelOpts(formatter="{b}: {d}%")))
pie.render_notebook()
#柱形图
pattern = re.compile(r'[u4E00-u9FA5]+')
df_season["team_list"] = df_season["比赛"].str.findall(pattern).map(set)
def get_opp(x,num=0):
period_dict = {0:"公牛",1:"奇才"}
opp = (x-set([period_dict[num]])).pop()
if opp == "人":
return "76人"
elif opp == "凯尔特":
return "凯尔特人"
elif opp == "超音速":
return "雷霆"
elif opp == "子弹":
return "奇才"
else:
return opp
df_season.loc[~df_season["赛季"].isin(["01-02","02-03"]),"opp"] = df_season["team_list"].map(lambda x : get_opp(x,0))
df_season.loc[df_season["赛季"].isin(["01-02","02-03"]),"opp"] = df_season["team_list"].map(lambda x : get_opp(x,1))
fig, ax = plt.subplots(figsize=(12,8))
order = df_season.groupby("opp")["得分"].mean().sort_values(ascending=False).index
sns.barplot(y="opp",x = "得分",order=order,data=df_season,color='#3c7f99',ci=None)
plt.box(False)
fig.text(x=0.04, y=0.90, s=' 对阵联盟各个球队的场均得分 ',
fontsize=32, weight='bold', color='white', backgroundcolor='#c5b783')
plt.tick_params(axis='both', which='major', labelsize=16)
ax.xaxis.grid(which='both', linewidth=0.5, color='#3c7f99')
plt.xlabel('')
plt.ylabel('')
plt.show()
fig, ax = plt.subplots(figsize=(12,8))
order = df_season.groupby("opp")["win"].mean().sort_values(ascending=False).index
sns.barplot(y="opp",x = "win",order=order,data=df_season,color='#3c7f99',ci=None)
plt.box(False)
fig.text(x=0.04, y=0.90, s=' 对阵联盟各个球队的胜率 ',
fontsize=32, weight='bold', color='white', backgroundcolor='#c5b783')
plt.tick_params(axis='both', which='major', labelsize=16)
ax.xaxis.grid(which='both', linewidth=0.5, color='#3c7f99')
# plt.xticks([0.2,0.4,0.6,0.8,1.0])
plt.xlabel('')
plt.ylabel('')
plt.show()
pattern = re.compile("(d+)-(d+)")
df_season["diff"] = df_season["比赛"].str.findall(pattern).map(lambda x : np.abs(int(x[0][0])-int(x[0][1])))
fig, ax = plt.subplots(figsize=(12,8))
sns.swarmplot(x="果",y="diff",data=df_season)
fig.text(x=0.1, y=0.90, s=' 输赢时的分差 ',
fontsize=32, weight='bold', color='white', backgroundcolor='#3c7f99')
plt.tick_params(axis='both', which='major', labelsize=16)
plt.xlabel("分差",fontsize=16)
plt.ylabel("比赛结果",fontsize=16)
plt.show()
df_playoff = pd.read_excel("./playoff.xls")
df_playoff["win"] = df_playoff["果"].map({"胜":1,"负":0})
pattern = re.compile(r'[u4E00-u9FA5]+')
df_playoff["team_list"] = df_playoff["比赛"].str.findall(pattern).map(set)
df_playoff.loc[:,"opp"] = df_playoff["team_list"].map(lambda x : get_opp(x,0))
df_plot = df_playoff.groupby(["赛季","轮次","opp"])["win"].mean().reset_index()
# season_dict = dict(zip(df_plot["赛季"].unique(),[i for i in range(12)]))
# df_plot["season"] = df_plot["赛季"].map(season_dict)
# horizon = [0.11,0.18,0.24,0.30,0.37,0.43,0.50,0.56,0.63,0.69,0.76,0.82]
# vertical = [0.13,0.35,0.58,0.79]
# fig, ax = plt.subplots(figsize=(16, 8))
# fig.text(x=0.03, y=0.92, s=' 乔丹季后赛全景图 ', fontsize=32,
# weight='bold', color='white', backgroundcolor='#3c7f99')
# plt.scatter(df_plot["赛季"], df_plot["轮次"], alpha=0.35)
# plt.tick_params(axis='both', which='major', labelsize=16)
# plt.box(False)
# plt.yticks([1, 2, 3, 4], labels=["第一轮", "第二轮", "分区决赛", "总决赛"])
# plt.show()
# for point in range(len(df_plot)):
# season = df_plot.iloc[point, 4]
# round_ = df_plot.iloc[point, 1] - 1
# team_name = df_plot.iloc[point, 2]
# left, bottom, width, height = horizon[season], vertical[round_], 0.08, 0.08
# ax2 = fig.add_axes([left, bottom, width, height])
# img = Image.open("./picture/{team_name}.jpg".format(team_name=team_name))
# ax2.imshow(img)
# ax2.axis('off') # 关掉坐标轴为 off
#
# pattern = re.compile("(d+)-(d+)")
# df_playoff["diff"] = df_playoff["比赛"].str.findall(pattern).map(lambda x : np.abs(int(x[0][0])-int(x[0][1])))
# color = df_playoff["果"].map({"胜":"#e60000","负":"#73e68c"})
#
# fig, ax = plt.subplots(figsize=(12, 8))
# fig.text(x=0.03, y=0.92, s=' 季后赛各轮次的输赢分差 ', fontsize=32,
# weight='bold', color='white', backgroundcolor='#3c7f99')
# plt.yticks([1,2,3,4],labels=["第一轮","第二轮","分区决赛","总决赛"])
# plt.scatter(df_playoff["diff"], df_playoff["轮次"], alpha=0.35,c=color,s=500)
# plt.scatter(df_playoff["diff"], df_playoff["轮次"], c=color)
# plt.tick_params(axis='both', which='both', length=0)
# plt.tick_params(axis='both', which='major', labelsize=16)
# ax.xaxis.grid(which='both', linewidth=0.75)
# #plt.xticks(np.arange(0,61,10), [str(i)+"k" for i in range(0,61,10)])
# plt.xlabel('分差', fontsize=18)
# plt.box(False)
# plt.show()
#seed_scale_plot()
season_score = df_season.groupby("赛季")["得分"].mean().values[2:]
playoff_score = df_playoff.groupby("赛季")["得分"].mean().values
index = df_playoff.groupby("赛季")["得分"].mean().index
fig, ax = plt.subplots(figsize=(12,8))
sns.barplot(x=index,y=playoff_score-season_score)
fig.text(x=0.06, y=0.90, s=' 季后赛与常规赛场均得分差 ',
fontsize=32, weight='bold', color='white', backgroundcolor='#3c7f99')
plt.tick_params(axis='both', which='major', labelsize=16)
plt.xlabel("赛季",fontsize=16)
plt.ylabel("场均分差",fontsize=16)
plt.box(False)
