Options

how to use loop in Rapidminer

sgnarkhede2016sgnarkhede2016 Member Posts: 152 Contributor II
edited September 2020 in Help
Hello,
I have below Data,
I want to check first value with all below values and vice versa is greter or not and put True or False flag in  check column

1
Sr   A     Check
1   12     
2    5       True
3    8     True

2
Sr   A     Check
1   12      False
2    5    
3    8       False

3
Sr   A     Check
1   12      false  
2    5       true
3    8    

How can i do this

Answers

  • Options
    BalazsBaranyBalazsBarany Administrator, Moderator, Employee, RapidMiner Certified Analyst, RapidMiner Certified Expert Posts: 955 Unicorn
    Hi,

    is this one example set? Or three different ones? In which form do you have them?

    If you have one table and want to change the Check column, you don't need a loop. Just Generate Attributes with something like this:

    A > Sr
    (this will automatically return true or false)

    I hope I understand you correctly. If not, please answer the questions and specify what the Check logic should be.

    Regards,

    Balázs
  • Options
    sgnarkhede2016sgnarkhede2016 Member Posts: 152 Contributor II
    this is single table but how can i check iterative ways
    like first time i need to check with "1" with 2,3  and then "2" check with 1,3 etc 
  • Options
    sgnarkhede2016sgnarkhede2016 Member Posts: 152 Contributor II
    I want to use multiple for loop in rapidminer as i used in python

    for yr in yr_df['sic'].unique().tolist():
                sic_df=yr_df[yr_df['sic']==yr]
                for index_1,i in sic_df.iterrows():
                    for index_2,j in sic_df.iterrows():
                        if i['name']!=j['name']:                        
                            if i['RevenueTotalR']>=j['below_percnt'] and i['RevenueTotalR']<=j['above_percnt']:
                                print(i['name'],'==',j['name'])
                                print(index_1)
                                i['peer_name']=j['name']
Sign In or Register to comment.