littlemtman

7 ms·

31.
by littlemtman 7y ago
Yes currently those models are just specialised in some special thing. If there really was a model that could understand the world like a human - well that would be artificial general intelligence
32.
by littlemtman 7y ago
In the end those are just mathematical models and not intelligence. Humans interact based off on their experience and it is really hard to find training data with a snow storm.
33.
by littlemtman 7y ago
Thank you. I have a list of things and I want to add. And it is really long
34.
by littlemtman 7y ago
It is not yet production code only ugly test code
35.
by littlemtman 7y ago
RELU is really good for mostly positive weights. ELU is good for both
36.
by littlemtman 7y ago
Cameras, it works for humans
37.
by littlemtman 7y ago
The risk is extremely high.
38.
by littlemtman 7y ago
No this is the only article about the comma ai model. Many people including me,in the discord server tried to extract the weights from the snpe and carry them over to a keras model
39.
by littlemtman 7y ago
I hope this gets me hired somewhere although I dont even have a high school degree yet
40.
by littlemtman 7y ago
I dont need LIDAR for it. It is very expensive and doesnt have a lot of image points to train on, at least compared to images so 480p for example.
41.
by littlemtman 7y ago
Thank you
42.
by littlemtman 7y ago
Because you can still find the best gradient with ELU. Say you use sigmoid and you have big weights then your gradient can get stuck
43.
by littlemtman 7y ago
Yes for higher autonomy levels more and more networks are needed
44.
by littlemtman 7y ago
Again I am not testing on real roads. The sentence When we compare this to what" I will show you, you are gonna see that this is insane" should mean that those numbers from Tesla are insane. And my project is not insane. That&#x27
45.
by littlemtman 7y ago
No, I am not testing this on real roads only testing with videos so far. So no real cars only videos in cars