1
$\begingroup$

Arguably, the MNIST or titanic examples are probably the most common and instructive examples of classification problems solved using machine learning.

Are there any 'iconic' examples of machine learning models used to predict continuous variables?

Notes

  • By 'continuous', I mean continuous within some range, not limited to negative infinity to positive infinity, 0 to positive infinity, or 0 - 1. E.g. predicting a property value or temperature.
  • By 'example' I mean something accessible on the web, with easy to download data and code (preferably available in the main data science languages)
Improve this question
$\endgroup$
2
  • $\begingroup$ You open by referring to data but then inquire about models: which do you mean? Please note that inquiries about data are not on topic here. $\endgroup$
    – whuber ♦
    Commented Oct 14, 2019 at 16:32
  • 1
    $\begingroup$ @whuber I thought just model, but I guess both (basically after an MRE just like titanic/MNIST effectively are). It's not a complete example without both. MNIST is great because it's doable in just a few lines of code, in many languages, with an easily accessible data set. $\endgroup$
    – stevec
    Commented Oct 14, 2019 at 16:32

2 Answers 2

Reset to default
2
$\begingroup$

On "Deep Learning with Python" by F. Chollet (https://www.manning.com/books/deep-learning-with-python) he uses the "Boston Housing Price" dataset for an introductory example in regression. In Python you can easily import the data using:

from keras.datasets import boston_housing
Improve this answer
$\endgroup$
2
  • $\begingroup$ interestingly, it's the only regression example in entire book, if my memory doesn't fail me. $\endgroup$
    – Aksakal
    Commented Oct 14, 2019 at 16:54
  • $\begingroup$ This post answers a different question: namely, what datasets might be available. The question concerns the models: what model(s) does Chollet use? $\endgroup$
    – whuber ♦
    Commented Oct 14, 2019 at 18:45
0
$\begingroup$

Anything related to physical machinery control will have continuous variables. For instance, double pendulum control, see "Control of Inverted Double Pendulum using Reinforcement Learning". It's a detailed description of the classroom project. see the video, it's easy to implement. enter image description here This demo is also great at showing and discussing the limitations of machine learning applications to this sort of problem. I've seen this type of demo on Google's event once, and the presenter made it sound as if any control problem would be easily solved with deep learning, didn't mention any difficulties with feature selection.

Improve this answer
$\endgroup$

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.