NAME: Stephen Lim
email: [email protected]
From the graph there is no clear overfitting.
Overfitting would show as a continuous divergence between training and test accuracy,
with test accuracy declining after an initial peak. The graph doesn't indicate that.
While the training accuracy is slightly higher than the test accuracy, the gap is
small and stable. This indicates the model generalizes reasonably well to unseen data.
Instructions:
javac *.java
java Logistic <learning rate> <iterations> <.train file> <.test file>
It is assumed that there is a .train file and a .test file in the same directory as the other files.
The csv file is created, and a sample pdf of the graph is provided from previous tests. The graph is generated through tkinter whenever the code is run.
