0
$\begingroup$

For a curve generated from dataset points, split the curve into parts and obtain the best-fit degree of polynomial,coeffcients and the interval/range of the split through implementation in python.I am new to ML, can someone suggest me a few concepts to be applied. Thanks in advance :)

Improve this question
$\endgroup$

1 Answer 1

Reset to default
0
$\begingroup$

Such a technique is called Spline regression: apply piecewise polynomial functions, in python you can use the generalized linear models module in statsmodels. As for how to choose the degrees and where to split I can't think of any method other than cross-validation.

Improve this answer
$\endgroup$
1
  • 1
    $\begingroup$ some links to provide the questioner with additional resources would be useful. It woudl also be a more complete answer if you gave a brief beginners explanation (couple of bulltet points) of spline regressions and cross validation. See stats.stackexchange.com/help/how-to-answer for advice $\endgroup$
    – ReneBt
    Commented Mar 6, 2019 at 11:01

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service and acknowledge you have read our privacy policy.

Not the answer you're looking for? Browse other questions tagged or ask your own question.