Muhammad Ubaid Raza

Well this sounds great! Just to update you GPU support is already on the way which will make brain.js super fast both in browser and nodejs environment, without requiring anything...

> Thanks @mubaidr, this is based on parameter averaging and data parallelization. It's probably the most naive implementation possible, but that's a good start because it's easy to test, and...

Not really. But newer version has GPU integration which is much faster than this implementation.

Its in Beta state, expected to release very soon. We do have plans to implement this too. This might help users with powerful CPUs without GPU.

Yes, exactly. But how much it will actually effect performance when using GPU (GPU is already many times faster than cpu) is yet to be seen.

That sounds great. Would love to see as contributor though.

I can confirm this. Chaining trim with extract fails for me too (without any error, just the trim is not performed). Using two pipelines works fine as suggested above.

``` I think, for consistency, changing every types' name to PascalCase would be better in general. ``` This, output should be consistent.

Make sure you donot have `env` variable `COMPOSE_PROJECT_NAME` set.

Also setting ENV variable in `manager` service: `${COMPOSE_PROJECT_NAME:-citus}_master`