decision-diffuser

decision-diffuser copied to clipboard

When I run `python train`, there is a error: > Traceback (most recent call last): > File "train.py", line 3, in > from analysis import RUN > ModuleNotFoundError: No module...

Chenrf1121

Different results of halfcheetah

7

Hello, I tried the same config with the repo and got the same good performance with the paper. However, when I tried the env halfcheetah and the testing score is...

rainbow979

Where is the code for history conditioning proposed in the paper?

7

Decision Diffuser needs to maintain a history of length K and then condition on it, to constrain the plan to be consistent, which is described in the paper. However, I...

tinnerhrhe

Thanks for the great work! Where can I find the code for the skill composition task of the quadrupedal robots in the paper Sec. 4.3? I did not find them...

liuzuxin

Why choose total reward of entire trajectory as label?

1

I have some confusions about 'label'(rewards) setting, the horizon of state and action sequences are fixed, but rewards array is from 'start' to end of entire episode. Why?

Chenrf1121

Cannot get packages on Windows

1

I've been trying to run the line **conda env create -f environment.yml** to get the required packages for the code, but I keep running into issues with patchelf. I'm trying...

ritvikpasham

ImportError: cannot import name 'instr' from 'ml_logger'

3

There's a problem when I run train.py. It shows that ImportError: cannot import name 'instr' from 'ml_logger'. Besides, it also cannot import 'needs_relaunch'.

Niloiv111

Metrics about hopper-medium-v2

1

hello, when I train and eval the model in hopper-medium-v2 task. I got results below:  {'average_ep_reward': 3269.0621404844796, 'std_ep_reward': 679.6445333781064} But in paper, I see the metirc value in the...

19373254