-
Couldn't reproduce the result on the Mujoco suite.
Setting: We run the BEAR with the recommend settings: ** `mmd_sigma = 20.0
`, `kernel_type = gaussian
`, `num_samples_match = 5
`,…
-
Hi @aviralkumar2907
Thank you for the code!
I try to reproduce the results reported in D4RL "walker2d-medium-v0" environment.
I run the code with the following command:
"python main.py --eval_f…
-
I'm interested in experimenting with https://github.com/rail-berkeley/d4rl/wiki/Off-Policy-Evaluation
-
Hi,
Thanks for sharing this repository. It is great
I'd like to ask about "Training and Evaluation Task Split" in Appendix D and how results are reported in Tables 1 and 3. I am a bit confused how…
-
1)
hopper-medium-expert-v0 has 1200919 samples.
It doesn't seem to be properly combined.
2)
d4rl/gym_mujoco/__init__.py
kwargs in register 'ant-medium-expert-v0' doesn't have 'ref_min_score' a…
ghost updated
4 years ago
-
Hello, could you help me figure out how to segment the dataset into trajectories/episodes?
For example, for the halfcheetah datasets, there are no terminal states included, so I assumed first 1000 da…
-
And it looks to me like this module is indeed missing.
Traceback:
```
(mopo) capybara@blg4101:~/mopo$ mopo run_local examples.development --config=examples.config.d4rl.halfcheetah_mixed --gpus=1 …
-
Hi Everyone
I am interested in benchmarking **BEAR**, but I cannot execute the corresponding script (**antmaze_bear.py**). I get the error message _"No module named 'rlkit.torch.sac.bear'"_. Unfort…
-
The demo data seems to be separated into different tasks. Were the VR demo users given a desired goal for each task, and is this goal data available?
-
It seems 'next_observations' is removed from the current version of the MuJoCo datasets. Is it possible to include them as the previous version? Are the data points in the correct order such that I ca…