-
Currently, in `lasercake -lne151 pressure_tunnel`, the number of ztree objects that's debug-printed is the same from run to run, but varies based on various things (in my quick test) :
g++-4.7 -O0, g…
-
With the model-based tests starting to replace standard static tests for Tendermint-rs, we need to decide on how to organize conformance / integration tests for this repository in particular, and conf…
-
See history:
https://github.com/gpuweb/cts/commits/main/src/resources/cache/webgpu/shader/execution/bitcast.bin
This file keeps getting updated (+0 bytes) on framework changes when none of the oth…
-
If you run `dotnet publish /t:PublishContainer` on a project that hasn't changed since the last container publish, the SDK will build a new image, even though nothing's changed
## Repro Steps
1.…
-
> > > Note that a copy only occurs if the data is shared with another instance, (i.e. if the OrdMap was previously cloned elsewhere) and only the shared parts are copied.
> >
> >
> > Ahhh…
-
In discussion about editorconfig today, a number of scenarios were mentioned that we know or suspect are not properly accounted for in determinism:
- additional files
- rulesets
- included rulesets…
-
- [ ] validate that weak references may cause non-determinism in model checking mode (and add a test)
- [ ] design the approach to support weak references in model checking mode
- [ ] track new we…
eupp updated
6 months ago
-
I have only one NVIDIA GPU, I was training with TensorFlow 2.5.2 because of the bug with GPU and multiprocessing.
- TF2.8 and No Child Process => works but Memory Leak :(
- TF2.8 and Child Proc…
-
### What version of Materialize are you using?
main
### What is the issue?
A general assumption of the compute protocol is that all replicas receive the (semantically) same stream of commands…
-
Hey, as I've described below, I think there are problems training Deepspeed in a multi-node setting when `full_determinism = True` in the `TrainingArguments`. I've replicated this on multiple hardware…