https://community.databricks.com/t5/machine-learning/modules/m-p/33865#M1797 <P>do not reinvent the wheel. If what you need exists already, use it.</P><P>If you only use a few methods of a package you can consider not importing it completely.</P><P>The cost of importing is not huge, but that depends on the amount of imports and the size of the imported packages.</P><P></P> Fri, 03 Dec 2021 09:16:40 GMT -werners- 2021-12-03T09:16:40Z https://community.databricks.com/t5/machine-learning/modules/m-p/33864#M1796 <P>Hello Python People.</P><P>Im still going through the motions learning python and have a general question.</P><P></P><P>example = Im creating basic ETL tasks to practice (<I>SQL, SQLite, Excel etc</I>)</P><P>I can see that to read excel I can use the pyodbc module - or can use Pandas , openpyxl</P><P></P><P>is it ok to just take pieces where needed that you prefer from other modules or would you aim to use as few modules as possible?</P><P></P><P>I suppose I'm not sure of the cost of importing modules</P> Fri, 03 Dec 2021 07:43:40 GMT https://community.databricks.com/t5/machine-learning/modules/m-p/33864#M1796 missyT 2021-12-03T07:43:40Z https://community.databricks.com/t5/machine-learning/modules/m-p/33865#M1797 <P>do not reinvent the wheel. If what you need exists already, use it.</P><P>If you only use a few methods of a package you can consider not importing it completely.</P><P>The cost of importing is not huge, but that depends on the amount of imports and the size of the imported packages.</P><P></P> Fri, 03 Dec 2021 09:16:40 GMT https://community.databricks.com/t5/machine-learning/modules/m-p/33865#M1797 -werners- 2021-12-03T09:16:40Z