1
$\begingroup$

As per the Orange video tutorial: youtu.be/D6zd7m2aYqU?list=PLmNPvQr9Tf-ZSDLwOzxpvY-HrE0yv-8Fy

I'm trying to make predictions on a new dataset based upon one that's already been 'coded' - both data sets are formatted exactly the same, where the 'training' set has all entries marked as either 1 or 0, see below:

enter image description here

whereas the second set contains no values, as I hope the 'predictions' feature will try to guess the status:

enter image description here

Trouble is, when I look at the predictions widget, it does nothing, displaying 'Data does not have the same target as predictors'

I've only been using Orange for two days, so I imagine it's a really simple error on my part.

Any help would be greatly appreciated.

John

$\endgroup$
2
  • $\begingroup$ Thank you. It worked when I removed the column. The youTube tutorial is misleading. $\endgroup$ Commented Dec 3, 2018 at 2:39
  • $\begingroup$ Please don't add "thank you" as an answer. Once you have sufficient reputation, you will be able to vote up questions and answers that you found helpful. - From Review $\endgroup$
    – Aditya
    Commented Dec 6, 2018 at 4:53

2 Answers 2

3
$\begingroup$

This is because we have to exclude the target column from the second document (testing data-set).

In your example remove the column 'Status' from second file.

Example:

In below example, in training dataset, 'Loan_Status' column is present which is the target field, whereas, in testing dataset, 'Loan_Status' column is not present, and will be computed by the 'Prediction' as shown below:

training and testing data set

Prediction:

prediction value

$\endgroup$
1
$\begingroup$

this may be a bug in orange tool because the data sets look well formatted. I suggest you to put all data (both data sets) in one file and load it onto orange, if it is OK then split the file (separate the data) and reload the two new files and it should work.

$\endgroup$

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service and acknowledge you have read our privacy policy.

Not the answer you're looking for? Browse other questions tagged or ask your own question.