humancompatibleai / evaluating-rewards Goto Github PK

I'm currently working on a project (branching from this) where I created a new gym environment adding policies for autonomous driving in a highway. I defined my own reward weights for each of the actions the agent could take, so I would like to evaluate them, in order to get an optimal value for the reward function and, then, perform a DQN. My question is if your tool supports or not custom gym environments and, if so, how could I exploit it. I tried your notebook in colab but there's a problem with the python version, so I installed it on linux but I don't get how to implement it in my code.
Thanks in advance for your help.

This problem may not be related to this specific package, but it can happen if you install this package, so I am opening the issue here too.

When I try to install this package with pip (version 19.0.3 with Python 3.8), I get the error

Could not find a version that satisfies the requirement tensorflow<1.16,>=1.15 (from evaluating-rewards==0.1.1) (from versions: 2.2.0rc3, 2.2.0rc4, 2.2.0, 2.2.1, 2.3.0rc0, 2.3.0rc1, 2.3.0rc2, 2.3.0, 2.3.1)
No matching distribution found for tensorflow<1.16,>=1.15 (from evaluating-rewards==0.1.1)

I am using a 64-bit version of Python on a Mac OS Catalina (10.15.7). Meanwhile, this problem seems to be temporarily solved by changing the specific version of TensorFlow that is installed in the requirement.txt file to tensorflow==2.3.0. However, I have not confirmed that this problem is solved because I was not yet able to run any code in this repo, given that we apparently also need mojuco as a requirement for this package, but mojuco requires a license, and that's quite annoying.

Once regenerated models, remove:

• Wildcard import in src/evaluating_rewards/rewards/__init__.py
• src/evaluating_rewards/analysis/resulst.py -- path rewriting

pylint 2.5.0 introduced a lot of regressions, so I've pinned it to 2.4.x series for now to avoid CI failing.

Once bugfix release comes out, should remove the pin and deal with any new issues it catches.

Some particularly blocking issues for us:

• pylint-dev/pylint#3524
• pylint-dev/pylint#3547

It also seemed to be ignoring pylintrc, which I did not see reported.

How do I resolve the following conflict? Please help.

I have tried both imitation@tf-master and tf-master-compatibility branch

Linux,
Python 3.7
Tensorflow 1.15

The conflict is caused by:
    evaluating-rewards 0.1.1 depends on stable-baselines 2.10.3a0.-WIP- (from git+https://github.com/hill-a/stable-baselines.git)
    imitation 0.1.1 depends on stable-baselines~=2.10.1

To fix this you could try to:
1. loosen the range of package versions you've specified
2. remove package versions to allow pip attempt to solve the dependency conflict