MLP 로 Linear Regression 구현하기 (작성중)
Python 3.8, Jupyter 환경 이용하였습니다.
데이터 생성
First generate 10,000 (\(100 \times 100\)) 2-dimension data points based on $ f(x, y) $ where $-1\leq x \leq 1$ and $-1 \leq y \leq 1$.
Add standard Gaussian noise to $f(x, y)$, ie. $\hat{f}(x, y)=f(x,y)+0.1N(0. 1)$ (1)
The input instance $(x, y)$ and the target value form a dataset, namely $\left{[x, y], \hat{f}(x, y)\right}$