tr_x = trans_range*np.random.uniform()-trans_range/2 steer_ang = steer + tr_x/trans_range*2*.2
An augmentation based deep neural network approach to learn human driving behavior
Vivek Yadav

Hi Vivek,

I got the problem with these 2 lines.

When I mathematically insert the formula of tr_x into steer_ang, I get steer_ang = 0.4 * np.random.uniform()-1. Random here is used as a real number, calculated once for tr_x. This formula means, that no matter what tr_x, steering angle is always reduced.

Am I right?

Show your support

Clapping shows how much you appreciated Dima Yanchenko’s story.