use cnn_to_mlp to allow training of racecar using (extremely simplified) ZED camera pixel data using OpenAI baselines.

add a red sphere to make training a bit easier for now.
2017-06-13 18:33:32 -07:00
parent ee8fd56c5e
commit a0ded43a69
6 changed files with 3793 additions and 29 deletions
--- a/examples/pybullet/gym/enjoy_pybullet_zed_racecar.py
+++ b/examples/pybullet/gym/enjoy_pybullet_zed_racecar.py
@@ -7,7 +7,7 @@ from baselines import deepq
 def main():
    
    env = RacecarZEDGymEnv(renders=True)
-    act = deepq.load("racecar_model.pkl")
+    act = deepq.load("racecar_zed_model.pkl")
    print(act)
    while True:
        obs, done = env.reset(), False