add random agent

agents/deepracer_base_agent.py

+class DeepracerAgent():
+    def __init__(self):
+        self.agent_type = None
+
+    def register_reset(self, observations):
+        raise NotImplementedError
+
+    def compute_action(self, observations, info):
+        raise NotImplementedError
\ No newline at end of file

agents/random_agent.py

+import numpy as np
+from deepracer_base_agent import DeepracerAgent
+
+class RandomDeepracerAgent(DeepracerAgent):
+    def __init__(self):
+        pass
+
+    def register_reset(self, observations):
+        action = np.random.randint(5)
+        return action
+
+    def compute_action(self, observations, info):
+        action = np.random.randint(5)
+        return action

aicrowd.json

 {
-  "challenge_id": "evaluations-api-deepracer",
-  "grader_id": "evaluations-api-deepracer",
+  "challenge_id": "neurips-2021-aws-deepracer-ai-driving-olympics-challenge",
   "authors": ["aicrowd-bot"],
   "tags": "change-me",
-  "description": "Random agent for AWS Deep Racer",
+  "description": "Random agent for AWS Deep Racer"
 }
 

submission_config.py

+from agents.random_agent import RandomDeepracerAgent
+
+class LocalEvaluationConfig:
+    submission_agent = RandomDeepracerAgent