I am working on a problem to predict “next year” when treatment on a road would be required. This is what my dataset looks like: Dataset
I pre-processed the data as follows: converted categorical data to numeric using one-hot encoding (where there is no relationship) and label encoding. I started with linear regression and achieved an R² of 0.72.
In my dataset, I am using NEXT_TREATMENT_YEAR as the target variable. One issue is that my dataset doesn’t have any future dates, so I want my model to have extrapolation ability. I also used random forest and XGBoost, but neither of these can extrapolate. I am now working on setting up a neural network using PyTorch. However, I need some guidance to ensure I am on the right track.
Abhilash Vyas is a new contributor to this site. Take care in asking for clarification, commenting, and answering.
Check out our Code of Conduct.