pandas 自己流チートシート

#チートシート（翻訳含む）
英語）https://github.com/pandas-dev/pandas/blob/master/doc/cheatsheet/Pandas_Cheat_Sheet.pdf
日本語）https://qiita.com/s_katagiri/items/4cd7dee37aae7a1e1fc0

上記で大体わかるが、
データ加工など痒い所に手が届かない点があるので、自分なりに書く
（かぶっているところもあるが気にしない）

#詳細（随時更新予定）

import numpy as np
import pandas as pd

df = pd.DataFrame(
    {"Age":[22,33,44],
     "Sex":["man","woman","man"],
     "Embarked":["S","C","Q"],
     "FamilyS" :[1,2,4],
     "Name":["TORO Mr. BU","A Miss. gao","ninnniku"]
    }
)

#Use CSV
df.to_csv("file_name")
df=pd.read_csv("file_name")

### treat data
#checknull
null_val = df.isnull().sum()
percent = 100 * df.isnull().sum()/len(df)
kesson_table = pd.concat([null_val, percent], axis=1)
kesson_table.columns= ['欠損','%']

print(df)
#fillnull
df["Age"] = df["Age"].fillna(df["Age"].median())

#data change
#1. replace
df["Embarked"] = df["Embarked"].replace([0,1,2],[-1,-2,-3])

#2. map
df['Embarked'] = df['Embarked'].map( {'S': 0, 'C': 1, 'Q': 2, 'Unknown': 3} ).astype(int)

#3.apply
df["Embarked"] = df["Embarked"].apply(lambda x:0 if x == "S" else x)
df['Sex'] = df['Sex'].apply(lambda x: 1 if x == 'male' else 0)
def family(x):
    if x < 2:
        return 'Single'
    elif x == 2:
        return 'Couple'
    elif x <= 4:
        return 'InterM'
    else:
        return 'Large'
df['Embarked'] = df['Embarked'].apply(family)

#extract
df['Salutation'] = df.Name.str.extract(' ([A-Za-z]+).', expand=False)

#extract jyoken
df = df[df.Age > 0]

# convert row and columns
df.T

# unique
df["Age"].unique()

# group by
df.groupby(by="Age").sum().sort_values("Embarked")
df.groupby(by="Age").head(1)

#statics
df.corr()      #相関
df.describe()  #概要