Dirk Groeneveld
Dirk Groeneveld
Stas' code is here: https://github.com/stas00/toolbox/blob/master/pytorch/all_reduce_bench.py We should run this on a single node, and multiple nodes.
TODO - [x] Doesn't run on AMD like this
### 🚀 The feature, motivation and pitch The torch trainer has LR scheduling. The flax trainer should as well. ### Alternatives * None ### Additional context The trainer takes an...
### 🐛 Describe the bug You have to do this in a catwalk context, on commit `e7c5d158b9e8f1c925b4894037b5371a1efdeab7`. ```Python from tango import StepGraph sg = StepGraph.from_file("experiments/everything/everything.jsonnet") from tango import Workspace ws...
Something must have gone wrong with the most recent changes.  When I drag tool5 to the blue X, everything is as expected. When I drag it to the red...
I have this configuration:  There is no way to reach this configuration: 