Select the most correct answers If all the non-zero numbers in your confusion matrix are in one column it means that your model is predicting all one class If your model is fully trained this can be caused by your learning rate is too small your model is not well trained your model is predicting all one class your dataset has all training examples in one class your model has underfitt your model has overfit Check Select the most correct answers If all the non-zero numbers in your confusion matrix are in one column it means that your model is predicting all one class If your model is fully trained this can be caused by your learning rate is too small Check your learning rate is too big your learning rate is too small your weights are too large your dataset is too small your dataset is unbalanced Question 2 In one word only (all lowercase, no punctuation) what is it called if your test loss is much higher than your training loss? Not complete Marked out of 2.00 What would be likely to improve your model? P Flag question Oa. early stopping Ob. increasing the number of neurons Oc. increasing training data Od. decreasing the number of neurons Oe, training for longer Of. b and/or e Og. a, c and/or d Oh. a, b and/or Oh, all of a, b, c, d and/or e Check