ok. but if you change the map, you would have to train again. so why shouldnt i put regular code that detects if it was gonna fail? or why shouldnt i use pathfinding? can you think of a case where the neural network would beat the pathfinding and the regular physics test code ???
yes, this case is not the most suitable to use ANN for. in general, the most suitable are cases where there is a complex relationship between the inputs and outputs and where a similar input can vary a little bit. but I just wanted to demonstrate that you can easily write the code for a neural network yourself and can apply it to many situations if you do some creative thinking.
![Seungmin "그렇게, 천천히, 우리(As we are)" | [Stray Kids : SKZ-PLAYER]](http://i.ytimg.com/vi/kAzmhLHePqU/mqdefault.jpg)
ok. but if you change the map, you would have to train again. so why shouldnt i put regular code that detects if it was gonna fail? or why shouldnt i use pathfinding?
can you think of a case where the neural network would beat the pathfinding and the regular physics test code ???
yes, this case is not the most suitable to use ANN for. in general, the most suitable are cases where there is a complex relationship between the inputs and outputs and where a similar input can vary a little bit. but I just wanted to demonstrate that you can easily write the code for a neural network yourself and can apply it to many situations if you do some creative thinking.
@@peterpopma ok man nice work
Looking forward to your next ai unity videos
Interesting, this is the first tutorial on this subject that dosen't use MLAgents