About Performance about dsrl HOT 8 CLOSED

Hi,
You can generate the results by running the scripts in the corresponding environment directories. Please let me know if anything does not work.

from dsrl.

Thank you for your reply. My own training is relatively slow, and it takes about 2 weeks for a run (I don't know why). Due to time constraints, I would like to ask about the mean and variance of the TD3 plot in the paper?

from dsrl.

Generally, training SNNs w/ PyTorch is much slower than training ANNs. I don't have access to the data right now, I'll have to search for it on my old workstation. Maybe you can tell me what you're trying to do and I can help you.

from dsrl.

Thank you for your reply. I need these data because one reviewer of my paper asked me to add some methods for comparison. After running the source code, I found that the required time was too long, so I ask you about the original data for help

from dsrl.

OK, I understand. What data do you want to compare against? ANN or SNN or randomized SNN? I will try to find it for you.

from dsrl.

from dsrl.

This might take some time to dig up that data. Also, for a direct comparison with your results, you should use the same random seeds to ensure same initial conditions on the environment side as well as on the network initialization.

from dsrl.

Thank you for your reminder. While searching for your data, I will also try to run the source code directly without any modifications to see if my modifications caused the training speed to be extremely slow

from dsrl.