A simple linear regression model on a simple dataset: A simple linear model establishes a linear relationship between dependent(y) and independent(x) variables. The accuracy of the model is high in simple_linear_better compared to the other because of shuffling the data to remove orderness in the data so the model establish better linear relationship between dependent and independent variables. This shows the importance of shuffling the ordered data. Shuffling is not necesssary for unordered data.