0
$\begingroup$

I am a self-learning machine learning, so this question may be a bit trivial. I've been reading "Python Machine Learning" by Sebastian Raschka and Vahid Mirjalili; however, I am still uncertain what model to use.

My end goal is to be able to classify maneuvers based on sensor data. For example, if I were testing a bicycle doing different maneuvers such as left turn, right turn, front flip, and braking, then I could determine which maneuver it is based on lets say a couple accelerometers on the bicycle.

My current reading is leading me to a couple topics. I'd do dimensionality reduction (with PCA) to reduce the amount of redundancy in my data since there will likely be multiple sensors that return similar results.

I'd like to then use an LSTM to classify these maneuvers.

Are there examples of these kind of project being done before?

Improve this question
$\endgroup$

1 Answer 1

Reset to default
0
$\begingroup$

SVM, DBN, MLP, stacked autoencoders. Although SVM tends to be the best classifier for time-series, and it's nice and simple.

LSTM, RNN and GRU are for regression, not classification.

Improve this answer
$\endgroup$
1
  • $\begingroup$ How could an SVM classify different time-series? The time-series are likely not going to be separable let alone linearly separable. $\endgroup$
    – Michael Choi
    Commented Sep 10, 2018 at 15:56

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.