Project Assignment Tasks

The data file for this project is available from the Seaborn website. To load this data into a DataFrame, run

this code: $mpg=$ sns$.$load$\_$dataset$(\text{'}$mpg$\text{'})(2$ pts$.)$

Display the first five rows of data $(2$ pts$.).$

Get the correlation data for the mpg column. Sort the results to make it easier to see the columns with the

stronger correlations $(2$ pts$.).$

Use the column with the strongest positive or negative correlation to the mpg column to split the data into

test and training datasets, where the test dataset consists of $20\%$ of the total dataset. Be sure to specify a

value for the random$\_$state parameter $(4$ pts$.).$

Create a linear regression model from the training dataset $(3$ pts$.).$

Score the model using the test dataset $(2$ pts$.).$

Score the model using the training dataset $(2$ pts$.).$

Predict the $y$ values based on the $x$ values in the test dataset, and store the results in a variable. Then, put the

predicted values in a new DataFrame $(3$ pts$.).$

Join the y$\_$test and predicted data with the $x_{-}$test data, save the combined data in a new DataFrame, and

then display the first five rows of data $(3$ pts$.).$

Add the residuals to the new DataFrame created in step $9(2$ pts$.).$

Plot the residuals in a Seaborn KDE plot $(2$ pts$.).$

Melt the actual and predicted MPG columns together, assigning appropriate values to the variable and value

names $(2$ pts$.).$

Use a Seaborn scatter plot to plot the predicted data $(2$ pts$.).$