robotics-rl-srl
robotics-rl-srl copied to clipboard
araffin
S-RL Toolbox: Reinforcement Learning (RL) and State Representation Learning (SRL) for Robotics
for some network reasons, i download this code, and installed related package. When testing, I got a error. **Describe the bug** Traceback (most recent call last): File "/home/kai/anaconda3/envs/tf_torch/lib/python3.7/runpy.py", line 193,...
Thanks for the excellent toolbox! I'm trying to run the environment CarRacingGymEnv-v0: `python -m rl_baselines.train --algo sac --no-vis --num-cpu 4 --num-timesteps 10000 --env CarRacingGymEnv-v0` I get an error: > train.py:...
when using anaconda installed enviroment.yml in Window 10. I tried training a RL agent on a specific SRL model, error happened in function 'rl_baseline/visyalize.py/loadCSV()' error is cant load file '0.monitor.csv'...
This PR is a draft - the code is still in development: - add of tasks for CL: circular and square shaped move around a target, reaching a target (fex...
### Describe the bug Taking MobileRobotGymEnv-v0 as example: When setting RELATIVE_POS=False, the method `getGroundTruthDim` still return 2 as ground truth dimension, only the robot position is returned. In particular, this...
**Describe the bug** A normal autoencoder or VAE for images should have either sigmoid or tanh activation, but in the current master version, we have linear activation (i.e. no activation)....
As stable-baselines has now integrated tensorflow support, that could be cool to enable it in the toolbox repo (and to add it to the pytorch RL algo: SAC, CMA-ES, ARS).
**Describe the bug** In the current master version, the script ./replay/gather_results.py is used to gather experiments and produce CSV table. Recently, when I tried to reproduce the results of your...
**Describe the bug** In current master branch, we calculate the GTC by calling srl_zoo/plotting/representation_plot.py. In particular, in the function `plotCorrelation`, we call numpy to compute the correlation. ```python corr =...
**Describe the bug** I hope I'm wrong but it seems that the current training mechanism uses the validation set to update the weights during training which is unwanted. The problem...