- Home /
How do i create a SVM in unity using tensorflow?
I created an AI car in unity and I want my car to be able to avoid obstacles.
To do this i recorded some data which stores information about the scenario my car faced and my reaction to that (steering, brake and throttle input).
Then i am trying to feed this data to train a SVM model and generate a prediction for the AI car so that it can avoid obstacles, just like I did.
In fact my friend created a SVM in python for this, and the predictions were quite accurate. But, I don't have any idea of doing this in Unity and generate real time predictions.
Any ideas on how i can achieve this in Unity?
I have Tensorflow Sharp library and Unity ML agents integrated in Unity. However I have no idea where do I start? Is it possible to create an entire SVM, train it and generate predictions in one or more C# scripts using the Tensorflow sharp library?
Please Help.
Hi, did you have any luck doing this? I have a similar problem, I really don't know how to start
Take a look at Accord.net, it works well with unity..:)
Your answer

Follow this Question
Related Questions
How to set up PC for Unity ML Agent Sample Projects 0 Answers
Unity stutters when training with Machine Learning Agents. 1 Answer
Why Does this Barracuda error suddenly pops and stops training - unity ml agents 1 Answer
A good way to attach a mechanical part to another object at runtime and let it rotate? Like an axle. 0 Answers