Setting random seed

[izkula]

Dear Danijar,

I'm running into an issue that may be a non-issue, and I thought it was worth checking.

Are you able to reproduce training runs using a fixed random seed? There is a 'seed' flag in the config file, but I cannot find where it is actually being used, and my runs do not appear fixed to a seed.

Additionally, I am running into what appears to be a weird bug, and I am wondering if you have insight. The model does not train properly if I try to manually set the random seeds by adding, before any other code: np.random.seed(config.seed) tf.random.set_seed(config.seed) print(f'--> Setting random seed to {config.seed}')

For example, using dmc_walker_run, here is a training curve if I do not set the seed

Whereas here is a training curve if the only change I make is to add the above three lines.

This has been a consistent finding. I also see it if I try setting the random seed at other locations in the code. Otherwise, I am getting consistent success training without setting a random seed ( --> congratulations and thank you for the wonderful codebase and algorithm :-)

Is this a known problem? And/or do you have any insight into why this might be the case. Is there a reason to give up trying to set a random seed?

Note: I have been using the original version of your repo (i.e. from March). Is this something you have knowingly fixed with subsequent updates?

Thank you so much.

Best,

Isaac

[danijar]

Hi Isaac, setting the seed is not supported in the code base. I don't know why it breaks with your two added lines. I could imagine that it's related to these two pieces of code that are there to make sampling faster on GPU and that removing them could help. Sorry that I can't help much more with debugging this but please feel free to post progress here for other people to see who might be interested in fixing the seed.