Felipe Codevilla

Hey @eds89 , no question is really silly :). 1. Yes, during evaluation you need to have a CARLA running. Then you will receive new images and control de car....

Right now the problems in benchmark have been fixed and the results can be easily reproducible by running " run_CIL.py" . Please check out the blog post ! http://carla.org/2018/04/23/release-0.8.2/ I...

The gas is the throttle. For the best version of the data collection please check this repository: https://github.com/carla-simulator/data-collector

Yes, I also think it is unfair ! I will eventually work on that. I think it shouldn't be hard to make it work with tensorflow CPU. However downgrading to...

I am bit short on time to do that. But this should be quite straightforward. Just install tensorflow-cpu or other versions. Then you should probably remove the gpu parameters related...

Hey the crop is like this: image_cut=[115, 510] then you cut the image like this: rgb_image = rgb_image[image_cut[0]:image_cut[1], :] And reshape to 200x88 The angles are camera = { "id":...

Are you running on windows ? I can't reproduce the error =/

The training data contains demonstrations where the agent does not stop for traffic lights. So It is expected for it to not learn this policy. You would need new data...

Yes, Some last minute changes. I already see a pull request to fix that. Thanks for the issue !

Hey . Kabnoory 1. You are right ! left and right cameras needs to be augmented on the fly. This piece of code do the trick https://gist.github.com/felipecode/0c0bf85c6f3864f7f41c16a78b0416a4 2. I am...