td_intersect | SET Operations | Teradata Package for Python - td_intersect - Teradata Package for Python

Teradata® Package for Python User Guide

Product
Teradata Package for Python
Release Number
17.10
Published
May 2022
Language
English (United States)
Last Update
2022-08-18
dita:mapPath
rsu1641592952675.ditamap
dita:ditavalPath
ayr1485454803741.ditaval
dita:id
B700-4006
lifecycle
previous
Product Category
Teradata Vantage

Use the td_intersect() function to find the data at the intersection of the list of teradataml DataFrames or GeoDataFrames along the index axis, and returns a DataFrame or a GeoDataFrame with rows common to all input DataFrames or GeoDataFrames.

This function should be applied to data frames of the same type: either all teradataml DataFrames, or all GeoDataFrames.

Example Prerequisites

>>> from teradataml import load_example_data
>>> load_example_data("dataframe", "setop_test1")
>>> load_example_data("dataframe", "setop_test2")
>>> from teradataml.dataframe import dataframe
>>> from teradataml.dataframe.setop import td_intersect

Example 1: Run td_intersect() on rows from two DataFrames, using default signature

This example gets the intersection of rows from two teradataml DataFrames when using default signature of the function.

>>> df1 = DataFrame('setop_test1')
>>> df1
   masters   gpa     stats programming  admitted
id                                             
62      no  3.70  Advanced    Advanced         1
53     yes  3.50  Beginner      Novice         1
69      no  3.96  Advanced    Advanced         1
61     yes  4.00  Advanced    Advanced         1
58      no  3.13  Advanced    Advanced         1
51     yes  3.76  Beginner    Beginner         0
68      no  1.87  Advanced      Novice         1
66      no  3.87    Novice    Beginner         1
60      no  4.00  Advanced      Novice         1
59      no  3.65    Novice      Novice         1
>>> df2 = DataFrame('setop_test2')
>>> df2
   masters   gpa     stats programming  admitted
id                                             
12      no  3.65    Novice      Novice         1
15     yes  4.00  Advanced    Advanced         1
14     yes  3.45  Advanced    Advanced         0
20     yes  3.90  Advanced    Advanced         1
18     yes  3.81  Advanced    Advanced         1
17      no  3.83  Advanced    Advanced         1
13      no  4.00  Advanced      Novice         1
11      no  3.13  Advanced    Advanced         1
60      no  4.00  Advanced      Novice         1
19     yes  1.98  Advanced    Advanced         0
>>> idf = td_intersect([df1, df2])
>>> idf
   masters   gpa     stats programming  admitted
id                                             
64     yes  3.81  Advanced    Advanced         1
60      no  4.00  Advanced      Novice         1
58      no  3.13  Advanced    Advanced         1
68      no  1.87  Advanced      Novice         1
66      no  3.87    Novice    Beginner         1
60      no  4.00  Advanced      Novice         1
62      no  3.70  Advanced    Advanced         1

Example 2: Run td_intersect() on rows from two DataFrames, discarding duplicate rows

This examples applies the intersect operation on rows from the two teradataml DataFrames from previous example, discarding duplicate rows from the result by passing allow_duplicates = False.

>>> idf = td_intersect([df1, df2], allow_duplicates=False)
>>> idf
   masters   gpa     stats programming  admitted
   id                                             
   64     yes  3.81  Advanced    Advanced         1
   60      no  4.00  Advanced      Novice         1
   58      no  3.13  Advanced    Advanced         1
   68      no  1.87  Advanced      Novice         1
   66      no  3.87    Novice    Beginner         1
   62      no  3.70  Advanced    Advanced         1

Example 3: Run td_intersect() on more than two DataFrames

This example shows what happens when td_intersect is used on more than two teradataml DataFrames. In this example, you have three teradataml DataFrames as df1, df2 & df3, the operation is applied on df1 & df2 first, and then the operation is applied again on the result & df3.

>>> df3 = df1[df1.gpa <= 3.5]
>>> df3
   masters   gpa     stats programming  admitted
id                                             
58      no  3.13  Advanced    Advanced         1
67     yes  3.46    Novice    Beginner         0
54     yes  3.50  Beginner    Advanced         1
68      no  1.87  Advanced      Novice         1
53     yes  3.50  Beginner      Novice         1
>>> # Effective operation here would be, (df1-df2)-df3
>>> idf = td_intersect([df1, df2, df3])
>>> idf
   masters   gpa     stats programming  admitted
id                                             
58      no  3.13  Advanced    Advanced         1
68      no  1.87  Advanced      Novice         1

Example 4: Perform intersection of two GeoDataFrames

  • Create GeoDataFrames
    >>> geo_dataframe = GeoDataFrame('sample_shapes')
    >>> geo_dataframe1 = geo_dataframe[geo_dataframe.skey == 1004].select(['skey','linestrings'])
    
    >>> geo_dataframe1
    skey            linestrings
    1004  LINESTRING (10 20 30,40 50 60,70 80 80)
    >>> geo_dataframe2 = geo_dataframe[geo_dataframe.skey < 1010].select(['skey','linestrings'])
    
    >>> geo_dataframe2
    skey                        linestrings
    1009            MULTILINESTRING ((10 20 30,40 50 60),(70 80 80,90 100 110))
    1005                                         LINESTRING (1 3 6,3 0 6,6 0 1)
    1004                                LINESTRING (10 20 30,40 50 60,70 80 80)
    1002                                               LINESTRING (1 3,3 0,0 1)
    1001                                           LINESTRING (1 1,2 2,3 3,4 4)
    1003                       LINESTRING (1.35 3.6456,3.6756 0.23,0.345 1.756)
    1007                            MULTILINESTRING ((1 1,1 3,6 3),(10 5,20 1))
    1006           LINESTRING (1.35 3.6456 4.5,3.6756 0.23 6.8,0.345 1.756 8.9)
    1008  MULTILINESTRING ((1 3,3 0,0 1),(1.35 3.6456,3.6756 0.23,0.345 1.756))
    
  • Perform intersection
    >>> td_intersect([geo_dataframe1,geo_dataframe2])
    skey            linestrings
    1004  LINESTRING (10 20 30,40 50 60,70 80 80)