Classify a coordinate point, check accuracy with K Fold
Train/test has its own limitations. It is possible to end up overfitting to a specific train/test split. This may be the case when the training dataset isn't really representative of the entire dataset and too much stuff ended up in your training data set that skews things.
1. Check your model accuracy with K FOLD cross validation
2. Distill 4 dimentional that data down to two dimensions and display it on the scatter plot