Options

Import CSV wizard

sgenzersgenzer Administrator, Moderator, Employee, RapidMiner Certified Analyst, Community Manager, Member, University Professor, PM Moderator Posts: 2,959 Community Manager
in Product Feedback - Resolved
Two small but useful feature requests for the Import CSV data wizard:

1. Can we change the default from semicolon to comma (as in COMMA separated values) under "Column Separation"?  I end up having to change it from semicolon to comma dozens of times per day

2. On Step 4, it would be lovely to add a button that would set all the value types to Polynominal and another all to Numerical (next to where it says "Guess value types".  There are many times when I want to force it one way or another, and if I have dozens of attributes, it's a drag.

These would also both be nice in the Read CSV operator for the same reasons.

Thanks.

Scott
0
0 votes

Fixed and Released · Last Updated

Comments

  • Options
    JEdwardJEdward RapidMiner Certified Analyst, RapidMiner Certified Expert, Member Posts: 578 Unicorn
    Regarding Step 2.  - it would be nice if it could be a button with a range. 
    For example: set type from att1 to att29 to polynominal/numeric, etc. 

    Often for a large number of attributes (for example 100s) I setup with basic parameters to get the field names into the process then I copy out the XML of the paramets into notepad or Excel and edit with Find+replace or concat formula before putting them back exactly as I want them. 
    Pretty cumbersome at times. 
  • Options
    David_ADavid_A Administrator, Moderator, Employee, RMResearcher, Member Posts: 297 RM Research
    When switching between numerical and polynomial format why don't use the Numerical to Polynomial Operator?  It even allows a regular expression as filter, so selecting a lot of similar attributes should be easy.
  • Options
    sgenzersgenzer Administrator, Moderator, Employee, RapidMiner Certified Analyst, Community Manager, Member, University Professor, PM Moderator Posts: 2,959 Community Manager
    Yes using the numerical to polynominal works fine - just was looking to eliminate that extra step.  Also that only works when using the Read CSV operator.  If you use the wizard, you have to pull the data, convert, and store it again.  I had to do it several times last night and just realized that maybe this would be nice!

    Good idea about the XML find/replace.  Thanks!

    Scott
  • Options
    JEdwardJEdward RapidMiner Certified Analyst, RapidMiner Certified Expert, Member Posts: 578 Unicorn
    Anyone noticed a bug with the read CSV operator where if the CSV has multiple comment lines at the top the wizard only recognises one column? 
    Just wondering if it's just as I have been using files with funny encoding before I report it as a bug properly.
Sign In or Register to comment.