My original Dataset:
After applying the document vector:
I have only 2 columns in my dataset - Data and Category. My first 1000 rows are the training model but because of this sorting, I’m unable to train the TOP 1000 rows using partitioning and decision tree.
It is really urgent and I’m unable to figure out an alternative way to solve this problem.
What about partitioning your dataset first, then using the Document Vector node in your training branch, and the Document Vector (Apply) node in your test branch?
If I am misunderstanding something - all too likely! - maybe post your workflow and someone can take a look.
Thanks for the suggestion. I have reworked my workflow as below:
But, the Scorer is not displaying results now:
Something’s wrong with either the way the Scorer is configured, or the data being fed to it. It looks like only true values of “1” are being included.
Can you post your actual workflow (and data - IF it’s not business confidential) instead of screenshots?
I have rebuild the model, provided two different input files - learner and predictor and concatenated them later. It is working now.
This topic was automatically closed 182 days after the last reply. New replies are no longer allowed.