六、 合并

Pandas 提供了大量的方法能够轻松的对Series，DataFrame和Panel对象进行各种符合各种逻辑关系的合并操作。具体请参阅：合并。

Concat

  1. In [73]: df = pd.DataFrame(np.random.randn(10, 4))
  3. In [74]: df
  4. Out[74]: 
  5.           0         1         2         3
  6. 0 -0.548702  1.467327 -1.015962 -0.483075
  7. 1  1.637550 -1.217659 -0.291519 -1.745505
  8. 2 -0.263952  0.991460 -0.919069  0.266046
  9. 3 -0.709661  1.669052  1.037882 -1.705775
  10. 4 -0.919854 -0.042379  1.247642 -0.009920
  11. 5  0.290213  0.495767  0.362949  1.548106
  12. 6 -1.131345 -0.089329  0.337863 -0.945867
  13. 7 -0.932132  1.956030  0.017587 -0.016692
  14. 8 -0.575247  0.254161 -1.143704  0.215897
  15. 9  1.193555 -0.077118 -0.408530 -0.862495
  17. # break it into pieces
  18. In [75]: pieces = [df[:3], df[3:7], df[7:]]
  20. In [76]: pd.concat(pieces)
  21. Out[76]: 
  22.           0         1         2         3
  23. 0 -0.548702  1.467327 -1.015962 -0.483075
  24. 1  1.637550 -1.217659 -0.291519 -1.745505
  25. 2 -0.263952  0.991460 -0.919069  0.266046
  26. 3 -0.709661  1.669052  1.037882 -1.705775
  27. 4 -0.919854 -0.042379  1.247642 -0.009920
  28. 5  0.290213  0.495767  0.362949  1.548106
  29. 6 -1.131345 -0.089329  0.337863 -0.945867
  30. 7 -0.932132  1.956030  0.017587 -0.016692
  31. 8 -0.575247  0.254161 -1.143704  0.215897
  32. 9  1.193555 -0.077118 -0.408530 -0.862495

Join

类似于 SQL 类型的合并，具体请参阅：数据库风格的连接

  1. In [77]: left = pd.DataFrame({'key': ['foo', 'foo'], 'lval': [1, 2]})
  3. In [78]: right = pd.DataFrame({'key': ['foo', 'foo'], 'rval': [4, 5]})
  5. In [79]: left
  6. Out[79]: 
  7.    key  lval
  8. 0  foo     1
  9. 1  foo     2
  11. In [80]: right
  12. Out[80]: 
  13.    key  rval
  14. 0  foo     4
  15. 1  foo     5
  17. In [81]: pd.merge(left, right, on='key')
  18. Out[81]: 
  19.    key  lval  rval
  20. 0  foo     1     4
  21. 1  foo     1     5
  22. 2  foo     2     4
  23. 3  foo     2     5

另一个例子：

  1. In [82]: left = pd.DataFrame({'key': ['foo', 'bar'], 'lval': [1, 2]})
  3. In [83]: right = pd.DataFrame({'key': ['foo', 'bar'], 'rval': [4, 5]})
  5. In [84]: left
  6. Out[84]: 
  7.    key  lval
  8. 0  foo     1
  9. 1  bar     2
  11. In [85]: right
  12. Out[85]: 
  13.    key  rval
  14. 0  foo     4
  15. 1  bar     5
  17. In [86]: pd.merge(left, right, on='key')
  18. Out[86]: 
  19.    key  lval  rval
  20. 0  foo     1     4
  21. 1  bar     2     5

Append

将一行连接到一个DataFrame上，具体请参阅附加：

  1. In [87]: df = pd.DataFrame(np.random.randn(8, 4), columns=['A','B','C','D'])
  3. In [88]: df
  4. Out[88]: 
  5.           A         B         C         D
  6. 0  1.346061  1.511763  1.627081 -0.990582
  7. 1 -0.441652  1.211526  0.268520  0.024580
  8. 2 -1.577585  0.396823 -0.105381 -0.532532
  9. 3  1.453749  1.208843 -0.080952 -0.264610
  10. 4 -0.727965 -0.589346  0.339969 -0.693205
  11. 5 -0.339355  0.593616  0.884345  1.591431
  12. 6  0.141809  0.220390  0.435589  0.192451
  13. 7 -0.096701  0.803351  1.715071 -0.708758
  15. In [89]: s = df.iloc[3]
  17. In [90]: df.append(s, ignore_index=True)
  18. Out[90]: 
  19.           A         B         C         D
  20. 0  1.346061  1.511763  1.627081 -0.990582
  21. 1 -0.441652  1.211526  0.268520  0.024580
  22. 2 -1.577585  0.396823 -0.105381 -0.532532
  23. 3  1.453749  1.208843 -0.080952 -0.264610
  24. 4 -0.727965 -0.589346  0.339969 -0.693205
  25. 5 -0.339355  0.593616  0.884345  1.591431
  26. 6  0.141809  0.220390  0.435589  0.192451
  27. 7 -0.096701  0.803351  1.715071 -0.708758
  28. 8  1.453749  1.208843 -0.080952 -0.264610