0

Very similar to the situation in this problem: What is neural network good accuracy but I don't have enough reputation to comment so I'm asking here again.

The NN I implemented has 9 inputs, 15 hidden units in 1 layer and 1 sigmoid output. The dataset I was using is about (1) if the annual income of ~50K people $\geq$ or $<$ a certain amount and (2) their education, gender, ethnics, etc (total raw features < 20).

I tried altering the regularization coefficient, no. of iterations, adding 1 more hidden layer (a lot slower smh) and even tried swapping the training and test sets but the 0-1 accuracy always topped at around 83%. Is the networks underfitting my data? What else can I do to improve my NN?

Straightforward logistic regression gives around 81% accuracy, so I bet NN is the way to tackle this prediction problem.

Btw, yes, I'm learning this topic from Prof. Andrew Ng as many others did.

Yuki.F
  • 101
  • 3
  • The answers on the other question seem perfectly applicable to your case. Can you clarify what additional questions you have that are not answered by the thread you link? See also: https://stats.stackexchange.com/questions/312780/why-is-accuracy-not-the-best-measure-for-assessing-classification-models/312787?r=SearchResults&s=6|0.0000#312787 – Sycorax Jun 25 '20 at 12:44
  • So @Sycorax do you have any ideas to possibly make my NN predict closer to the data I have given I have tried the above measures? I understand accuracy is a flawed metric but I won't give too much damn about it for this dataset. – Yuki.F Jun 25 '20 at 13:19
  • Since you've clarified that your question is not about whether your accuracy score is good or bad, but instead how to improve the generalization of your network, you should review the suggestions in the duplicate thread. – Sycorax Jun 25 '20 at 13:24
  • I afraid my NN is actually __underfitting__ given the accuracy on the training set, so the overfitting thread you suggested is obviously useful but might not be the most suitable this time. – Yuki.F Jun 25 '20 at 13:28
  • Please edit your question to explain what you know, what you've tried, and where you're stuck and we'll be able to help you get unsuck. – Sycorax Jun 25 '20 at 13:41

0 Answers0