ascentai / diy-gym Goto Github PK

Hi @thomascent,

I've been trying to use your envs with a stable_baselines algo (here's the cleaned-up repository) but I had to do some adjustments in order to make them compatible :
1. normalize and make symmetric action space
2. flatten observation space and action space
3. sum the rewards
4. compress the terminal signal
5. vectorize the environment
also, I had to make a quick fix to the observation space boundaries because the reset() method would return an observation that is outside the observation space (?).
These adjustments were detected via the use of the methods check_env() and set_env() from stable_baselines.

I used the example from the readme, and I tried to train a PPO2 model, with 6e5 training steps but unfortunately, this is the result (the values printed on the terminal are the reward).
I believe that the training affects only one joint and not also the others and as a result, the arm stretches only.
Any idea on how to approach this problem?
Also, you mentioned that you tested your environments using other agents. Can you upload a functioning example (e.g. the TD3 that you mentioned) please?

This repo seems fantastic! I am planning to test it extensively in the next few days!

I see that it was created 2 years ago, though. Is this project still active?
Also, did you publish this work on a paper/medium article?

All the best

Tried to install it today! Just a couple of modifications to make it work:

1. after running pip install -r requirements.txt I got the error:
   "Could not find a version that satisfies the requirement unittest"
   which I solved by deleting unittest from the requirements. Solution explained here.
2. when I tried to run python ur_high_5.py I got this other error:
   "segmentation fault (core dumped)"
   which I solved by installing an old version of gym.

To summarize I modified the requirements as follows:

numpy
gym<=0.15.3
pybullet
pyyaml
tqdm
argparse
requests

and now I can launch the test without a problem!
I hope this helps! :)