data cleaning

mrzh

i have 726 protein sequences and 1253 attribute, when i do removal of duplicates, useless attributes, and correlated features by default setting, only 748 attributes remain. after for decision tree, for some model, i cant gain tree.

is there relationship between number of attribute and drawing decision tree? please help me

varunm1

Hello @mrzh

Are you getting an error or just you are unable to see a decision tree? Did you try the decision tree by removing "Pruning option"?

You can set a breakpoint before decision tree and observe the data going into it to see if all the 748 attributes are going into it with 726 samples or not.

Hope this helps.