I'm loading two CSV files. One is a training dataset (with 2 string class
labels) and the other is a test set. The training data is roughly balanced
50:50. I'm getting close to 100% performance on the training data. However,
when i use the classifier to make predictions on the test set i always get
the same prediction. Even if i add a row from the training test (which it
gets correct) to the test set (it gets it wrong).
Sent from: https://weka.8497.n7.nabble.com/