-
### Describe the problem the feature is intended to solve
TensorFlow's pluggable device architecture offers a plugin mechanism for registering devices with TensorFlow without the need to make chan…
-
I'm still not quite understanding why it's acceptable to have TLA silently change the evaluation order.
If you `import defer` a module graph, subgraphs that contain TLA are eagerly evaluated. So, w…
-
Hi, I am doing some inference on large arrays. I have some workers for GPU tasks and others for the generic ones, mainly reshaping and I/O. I notice some GPU-related code is executed on workers withou…
mdrio updated
5 months ago
-
I was reading over Cycle's implementation and noticed an unexpected behavior while trying out some of the examples: it seems that components are re-created any time their parent changes, causing signi…
-
**Describe the bug**
It seems `Kokkos::Experimental::when_all` is not ready for runtime graph creation. Indeed, as shown below, it requires that predecessors are known at compile time.
https://g…
-
## 1. Systems
| system | single or distributed | multi-threads | performance or flexibility | static or dynamic |
| -------- | -------- | -------- | ------- | ---- |
| NetworkX | single | …
-
Hi
I'm learning about image dehazing so I was testing your code.
The open dataset was too many so I used only 500 image pairs(hazy and real image) in RESIDEv0-SOTS_indoor (gt folder and hazy folder)…
-
Today the lib does two things: talks to a mint following NUT specs and manages local wallet storage and proof management.
All apps will probably want to hit the APIs in the same way, while have more …
-
There is a lot of commented out "dead" code in VASim. There is also a lot of "method2()" left over from lazy feature addition and debug cycles. VASim needs a massive cleanup to make it easier to read…
-
## (CORRECTION) GPMA Node labelling should start from 1
The node labelling in GPMA should start from 1, the reason is because when GPMA is initialized by default a sentinel value is inserted for ever…