Can you give an example of customizing trainer? It is difficult to make large modifications via yaml file. Currently it seems that the start-kit only supports customizing envs and NN models.
Thank you.
Hello @kaixin
Did you have a look at https://github.com/AIcrowd/neurips2020-procgen-starter-kit/tree/master/algorithms/random_policy?
If you are looking for an even low-level implementation, this should be a good starting point, https://github.com/AIcrowd/neurips2020-procgen-starter-kit/blob/master/algorithms/custom_random_agent/custom_random_agent.py. If you plan to follow this one, please make sure that you are able to train your agent locally with the configuration provided in FAQ: Round 1 evaluations configuration.
Thank you for your quick reply. I will look at them 
@jyotish by the way, can you help look at submission #75308? I just clone and submit the impala-baseline but training failed in one of the environments.