Skip to main content
Evelina CAMILLERI
ST Employee
July 15, 2021
Solved

To train the MLC Decision tree, can the training dataset be generated based on simulation?

  • July 15, 2021
  • 1 reply
  • 799 views

..

    This topic has been closed for replies.
    Best answer by Evelina CAMILLERI

    Training a sensor using generated data can be tricky because your generation program needs to mimic as much as possible the sensor behavior (in terms of output data rate, full scale, sensitivity, noise, etc...). If, ideally, you have a good sensor model that can emulate the physical sensor, using simulated training data for the MLC is possible.

    1 reply

    Evelina CAMILLERI
    Evelina CAMILLERIAuthorBest answer
    ST Employee
    July 15, 2021

    Training a sensor using generated data can be tricky because your generation program needs to mimic as much as possible the sensor behavior (in terms of output data rate, full scale, sensitivity, noise, etc...). If, ideally, you have a good sensor model that can emulate the physical sensor, using simulated training data for the MLC is possible.