Matthew Rocklin

Sure, something like that seems nicely in scope. I look forward to seeing what you come up with! On Tue, Apr 23, 2019 at 9:36 AM sephib wrote: > Hi,...

cc @jacobtomlinson for when he gets back. @holdenk I'm curious, how often do you see people using docker on arm for dask-like workloads today? I'm curious how much we should...

cc'ing conda/dask experts @jakirkham @mariusvniekerk in case they have any insight

From my perspective the challenge here is that it's hard to wrap up all of the options in a comprehensive way. Someone will come by with a more custom configuation...

Both are solved by dask-labextension, which anticipated this need. For 1 you can create an initial cluster in the configuration. For 2 you need to use the auto-start dask option...

https://github.com/dask/dask-labextension#configuration-of-dask-cluster-management On Mon, Dec 9, 2019 at 10:51 AM Matthew Rocklin wrote: > Both are solved by dask-labextension, which anticipated this need. > > For 1 you can create an...

There are some active issues though: https://github.com/dask/dask-labextension/issues/84 See also https://github.com/pangeo-data/pangeo/issues/743 On Mon, Dec 9, 2019 at 11:16 AM Tom Augspurger wrote: > Thanks. Looking into that now. > > On...

I think that the PipInstall worker plugin was added to master recently. https://github.com/dask/distributed/pull/3216 On Wed, Nov 18, 2020 at 5:01 AM Jacob Tomlinson wrote: > Overall, I 100% agree that...

How hard is it to support PyTorch/Keras fit/predict APIs? If this is as simple as making a function like the following, then I would be in favor ```python def fit(estimator,...

For serialization I think that we have a decent Pytorch serializer in distributed (early work from @stsievert if I recall correctly). I don't think that we have anything for Keras...