Neural Network Training

Modified on Thu, 14 Dec, 2023 at 1:46 PM

NEURAL NETWORK TRAINING

Neural Network Trainer is an online tool from HP Tuners that allows tuners to configure the complex neural networks used by some modern vehicle controllers to model Volumetric Efficiency.

NOTE: Select 2010+ FCA and 2019+ GM vehicles are eligible for neural network training

Neural networking is an artificial intelligence concept that can be used to "train" computing systems. Such systems "learn" by recognizing patterns in sample input and output data and then apply that learning to determine a reasonable result from any input data received.

In the past, aftermarket calibration tools have been unable to configure neural networks because the calibration data produced by a neural network is often too complex to configure directly.

Neural Network Trainer takes a different approach: it allows you to specify the sample data used to train the neural network. You teach the network. It learns from you.

CONFIGURATION OVERVIEW

NOTE: It takes 25 Tuner Tools Tokens to license a vehicle for one month of neural network training. Tokens can be purchased from the My Account page of the Tuner Tools website.

1.Use VCM Editor to read the vehicle's current tune.
2.Export a training file.
3.Upload the training file to the Tuner Tools website. Neural Network Trainer will process the base file and produce files containing configurable VE tables.
4.Download the VE table files produced by the Tuner Tools website and open them in the Neural Network Trainer window of VCM Editor.
5.Modify the VE tables as desired and export a new training file.
6.Upload the new training file to the Tuner Tools website. Neural Network Trainer will process the base file and produce trained files.
7.Open the trained files in VCM Editor and flash the updated tune to the vehicle.

See Detailed Training Procedure for more information.

Was this article helpful?

That’s Great!

Thank you for your feedback

Sorry! We couldn't be helpful

Thank you for your feedback

Let us know how can we improve this article!

Select at least one of the reasons
CAPTCHA verification is required.

Feedback sent

We appreciate your effort and will try to fix the article