排序

1> 按标签排序

使用 sort_index()对 DataFrame 进行排序,默认情况下,按照升序对行标签机型排序

un_sorted_df = pd.DataFrame(np.random.randn(5,2), index = [1,4,3,2,0],columns = ['col1','col2'])
print(f'排序前:\n{un_sorted_df}')

# 输出结果:
#  排序前:
#         col1      col2
#  1  1.384395  1.761308
#  4 -0.421264 -0.978312
#  3  1.999370  0.127830
#  2  0.856805  0.713296
#  0 -0.441599  1.015312

print(f'排序后:\n{un_sorted_df.sort_index()}')

# 输出结果:
#  排序后:
#         col1      col2
#  0 -0.441599  1.015312
#  1  1.384395  1.761308
#  2  0.856805  0.713296
#  3  1.999370  0.127830
#  4 -0.421264 -0.978312

2> 排序顺序

un_sorted_df = pd.DataFrame(np.random.randn(5,2), index = [1,4,3,2,0],columns = ['col1','col2'])
print(f'排序前:\n{un_sorted_df}')

# 输出结果:
#  排序前:
#         col1      col2
#  1  1.384395  1.761308
#  4 -0.421264 -0.978312
#  3  1.999370  0.127830
#  2  0.856805  0.713296
#  0 -0.441599  1.015312

print(f'排序后:\n{un_sorted_df.sort_index(ascending = False)}')

# 输出结果:
#  排序后:
#         col1      col2
#  4  1.157527 -0.151781
#  3  0.191023  0.766429
#  2  1.112582 -1.256559
#  1  0.276446  0.804178
#  0 -1.394744 -0.462737

3> 按列排序

un_sorted_df = pd.DataFrame(np.random.randn(5,2), index = [1,4,3,2,0],columns = ['col2','col1'])
print(f'排序前:\n{un_sorted_df}')

# 输出结果:
#  排序前:
#         col2      col1
#  1 -0.624151  0.113447
#  4  0.419938 -0.805421
#  3  0.247321  0.260455
#  2 -0.464907 -0.453989
#  0  0.454340 -1.243215

print(f'排序后:\n{un_sorted_df.sort_index(axis = 1)}')

# 输出结果:
#  排序后:
#         col1      col2
#  1  0.113447 -0.624151
#  4 -0.805421  0.419938
#  3  0.260455  0.247321
#  2 -0.453989 -0.464907
#  0 -1.243215  0.45434

4> 按值排序

un_sorted_df = pd.DataFrame({'col1':[2, 1, 1, 1],'col2':[1, 3, 2, 4]})
print(f'排序前:\n{un_sorted_df}')

# 输出结果:
#  排序前:
#     col1  col2
#  0     2     1
#  1     1     3
#  2     1     2
#  3     1     4

print(f'排序后:\n{un_sorted_df.sort_values(by = ["col1","col2"])}')

# 输出结果:
#  排序后:
#     col1  col2
#  2     1     2
#  1     1     3
#  3     1     4
#  0     2     1

5> 排序算法

• mergesort(稳定)
• heapsort
• quicksort

# 输出结果:
#  排序前:
#     col1  col2
#  0     2     1
#  1     1     3
#  2     1     2
#  3     1     4un_sorted_df = pd.DataFrame({'col1':[2, 1, 1, 1],'col2':[1, 3, 2, 4]})
print(f'排序前:\n{un_sorted_df}')

# 输出结果:
#  排序前:
#     col1  col2
#  0     2     1
#  1     1     3
#  2     1     2
#  3     1     4

print(f'排序后:\n{un_sorted_df.sort_values(by = "col2",kind = "mergesort")}')

# 输出结果:
#  排序后:
#     col1  col2
#  0     2     1
#  2     1     2
#  1     1     3
#  3     1     4