Ryan Tremblay
Ryan Tremblay
@Michael24884 is there some other log output you were looking for that could point us in the right direction?
@Michael24884 Hardcoding the source to that seems to have gotten me a new error, which is progress, so thanks! However, I get no video and this log output: ``` 2020-11-11...
> It's probably just the ffmpeg looking ahead into the video data. How are you laying out the component for the player? It might be due to the insufficient size...
> @zaptrem did you just set the resize mode to fix this error? I'm having a similar issue with a blank screen when loading the react native component inside a...
> As far as I am aware, Dreamer experiments relied on the older version of MineRL (v0.4.4). Make sure to install those. > > Also, double-check your Java versions by...
> The "ticks per second" thing does not quite tell the step count, but yes, it is running bit slow. MineRL in general runs pretty slow, but that amount of...
> MineRL has never had official support for multi-agent code even though there are some remnants on some implementation for it in the code-base, so those features are really finnicky...
> The whole multi-agent code side does not work. Note that the environment code file is called "multiagent" but it still does not really support it. You should create the...
Same issue. It appears pyllamacpp also doesn't have an option to disable this. Worse yet, working around it by just ignoring the first X tokens results in a long pause...
> Thanks for the answer! Both models are trained with 16bit precision which is set there: https://github.com/Lightning-Universe/lit-GPT/blob/main/train.py#L90 > > Could it be related to moving from fairscale to native implementation...