I am working on a Regression task, where the goal is to predict one target variable. I have 37 input features with 6700 datapoints.
I pre-processed data and developed a neural network with 3 layers with 64, 32, and 16 nodes. Using SGD+Momentum, L2 regularization, relu activation, 1000 epoches and learning rate of 0.000001 – I was able to get to RMSE of 2.36 and r2 score of 0.73. My goal is to get r2 score above 0.9. I tried many combinations to get upto this score. What else should I try?
New contributor
Abhilash Vyas is a new contributor to this site. Take care in asking for clarification, commenting, and answering.
Check out our Code of Conduct.