I need help with 5-fold cross validation in python without using sklearn.

Next, you need to use **5-fold cross validation** to pick out the best model for the prediction. You should present a chart of the error function $J$, similar to the one in the lecture (see below). You may need to use package like "**tabulate**" to print out the table.

This is the contents of the textfile: the first two numbers are m for lines of data and n for number of features.

19 1 28 12.2 32 11.9 36 11.5 48 11.9 52 11.5 56 11.5 60 11.0 64 11.4 68 11.0 72 11.07 76 11.08 80 11.06 84 10.97 88 10.54 92 10.82 96 10.94 100 11.12 104 10.93 108 10.78

You read in the first column as x and the second as y. Then turn the the x array into a 2D array like [[1 28]

1 32] etc. Then you must find J(w₀,w₁) as well as the weight and fold the data into equal folds [1-5]. You then must take the 5 fold as the test set and the rest of the fold into training set. With those sets you must use a linear, Quadratic, Cubic, and fourth equation to get the mean using J_w(x) on each sub model

This is what should be the output.

You need to plot your training and testing J with respect to the polynomial degree, similar to the one in the lecture. With this data you do 5-fold cross validation to get the means shown above.

\begin{tabular}{lrrrr} & Linear & Quadratic & Cubic & Fourth \\ \hline 1234 & 10.25 & 24.25 & 12.25 & 23.25 \\ 5 & 10.25 & 19.25 & 14.25 & 23.25 \\ 1235 & 10.25 & 15.25 & 19.25 & 23.25 \\ 4 & 10.25 & 10.25 & 20.25 & 23.25 \\ 1245 & 10.25 & 24.25 & 12.25 & 23.25 \\ 3 & 10.25 & 19.25 & 14.25 & 23.25 \\ 1345 & 10.25 & 24.25 & 12.25 & 23.25 \\ 2 & 10.25 & 19.25 & 14.25 & 23.25 \\ 2345 & 10.25 & 24.25 & 12.25 & 23.25 \\ 1 & 10.25 & 19.25 & 14.25 & 23.25 \\ Mean for Training & 0 & 0 & 0 & 0 \\ Mean for Testing & 0 & 0 & 0 & 0 \end{tabular}