Hello,
I am doing a super simple experiment with ml-agents. I am just taking the example project PushBlock and making a change to the environment and training it again. My issue is that I am a beginer with Python and don’t know C++ at all. It seems like I could just change out simple things like create a new ground and overwrite the old ground but I haven’t been able to do it without breaking something else.
So, is there anyone out there willing to spend an hour or so tutoring me? Willing to pay for your time (I’m not a jerk). I know there is likely someone who could do this whole thing in five minutes, so just looking for some help.
Thanks.
More information:
I took the example PushBlock and created a new Scene. I set up all of my Assents the exact same as the example Agent, Block, Goal, Area. I assumed that in doing so I would be able to run the same model in a different environment. That may be a bad assumption.
I got the following error on running:
NullReferenceException: Object reference not set to an instance of an object
This was followed by a short list of items that all appear in the PushAgentBasic.cs script. The errors all seem to be related to me not assigning something to an object in my scene; however, I can’t seem to figure out where they should be assigned and how to do it. Again, just basically copying the example here, so really surprised this isn’t working.