DataFrames Become Invalid When a Connection Context is Recreated

Teradata® Python Package User Guide

brand
Teradata Vantage
prodname
Teradata Python Package
vrm_release
16.20
category
User Guide
featnum
B700-4006-098K

On recreating a connection context using the create_context command, most of the operations on the DataFrames created using an earlier connection context fail.

The users receive a warning when the create_context command is rerun.

For example:

>>> create_context(host = "myhostname", username="myusername", password = "mypassword")
>>> df = DataFrame('table1')
 
 
>>> create_context(host = "myhostname", username="myusername", password = "mypassword")
UserWarning: [Teradata][teradataml](TDML_2002) Overwriting an existing context associated with Teradata connection. Most of the operations on any teradataml DataFrames created before this will not work.
 
 
>>> df.select(['model','phrase_id'])
TeradataMlException: [Teradata][teradataml](TDML_2103) Internal Error: Non-zero status returned from AED.

To work around this limitation, recreate the DataFrames with the new connection context.

For example, recreate the 'df' DataFrame after the new connection context is created in the above example.

>>> df = DataFrame('table1')
>>> df.select(['model','phrase_id'])
      phrase_id
model         
1             1
1             2
1             2
1             1
1             1
1             1
(6, 3)