pandas查询手册

删除重复值

drop_duplicates(subset='author_id',keep='first',inplace=True)

subset：若该字段下样本值相同，则判为重复样本
keep: first 、 last
被判为重复样本的样本中，取第一个样本（first）或最后一个样本（last）作为删除重复样本后的结果
inplace:是否在原dataframe进行改动

删除满足条件的行

df = df.drop(df[条件].index)

删除列

data.drop(['Age'],axis=1,inplace=True)

改列名

data.columns = ['Age','Female','Male']
Xtrain = Xtrain.rename(columns={'Date':'Month'})

重置索引

locafinal = locafinal.set_index(keys="Location")

pandas实现vlookup操作

localfinal的索引与Xtrain[‘location’]进行匹配

Xtrain["Location"] = Xtrain["Location"].map(locafinal.iloc[:,0])

去掉逗号

#re.sub(希望替换的值，希望被替换成的值，要操作的字符串)
Xtrain["Location"] = Xtrain["Location"].apply(lambda x:re.sub(",","",x.strip()))

isin()与not in

data_ = data_[~(data_['Embarked'].isin())]

concat

data = pd.concat([data,age_],axis=1)

删除缺失值

data.dropna(inplace=True) 

pandas填补缺失值

data['Age'] = data['Age'].fillna(data['Age'].mean())

merge

df_merge = pd.merge(left=df_embedding,right=df_movie,left_on='word',right_on='movie_id')

df_merge = pd.merge(left=df_embedding,right=df_movie,on='movie_id')

pandas的group by操作

grouped = pd.DataFrame(data['发币金额'].groupby(data['系统号/组织编码']).sum().reset_index())