Classification (data consolidation and grouping aggregation)

First read in the data ,

Ideas ： Reconstruct an all 0 Array of , The column name is classification , If a classification appears in a piece of data , let 0 Turn into 1

List of statistical classifications df['Genre"].str.split(',').tolist()

df.join()# Merge data with the same index ,

Data consolidation .

Merge row indexes .join（index）

Column index for merging merge(coloms)

df3=pd.DatsFrame(np.zeros(3,3)),columns=list("fax"))

df1.merge(df3,on="a") according to a Column index for operation

merge Combine

df3.loc[1,"a"]=1

 

 merge outer Combine nan completion

merge left On the left nan completion

merge right On the right ,nan completion

Data packet aggregation ：df.groupby()

DataFrameGroupby It can be traversed

Index and composite index

  Index methods and properties print(grouped1.index)

df.index=['x','y']

df1.reindex['a','f']# To reset index

df.set_index("a")# Specify a column as an index

df.set_index("a”,unique)

matplotlib Map the top 10% of the total number of stores in the country

Prepare the data df.hroupby=(by=Country').count().sort_values(ascending=False)[:10])

_x=data1.index()