-
Not sure if this is on rune or something upstream but posting here:
Environment: Ubuntu 20.04 on WSL 2
Command: cargo install hotg-rune-cli
```
blackrat@SLIVER:~/workspace/hot-g/rune-cli$ car…
-
There are two vague issues with the compiler frontend as it currently stands:
1. Accumulation of technical debt due to code churn/prototyping - makes it hard to follow the flow of logic and transfo…
-
We'd like to use `librunecoral` as Rune's main TensorFlow Lite inference backend while still allowing hardware acceleration (e.g. using `libedgetpu`) when available.
-
To help people troubleshoot slow Runes, we need a way to see how much time is spent on each stage of the Rune's pipeline.
My initial thoughts are to modify the code that is generated by `rune build…
-
I tried to test the NLP mobilebert_qa.tflite model with the new `Rune librunecoral version`, but I'm still getting the same old ops error. Here is all the [codebase](https://github.com/hotg-ai/test-ru…
-
While running `Rune model-info model.tflite` for [Pose Estimation](https://tfhub.dev/tensorflow/lite-model/posenet/mobilenet/float/075/1/default/1) model, it gives
```
Inputs:
sub_2: Float3…
-
https://www.tensorflow.org/lite/performance/gpu_advanced
On Mobile devices and even on devices like raspberry pi this can bring us significant speed ups and power savings.
But this needs to be par…
-
With some Rune files using models with many labels, the build time seems to almost *exponentially* increase in time complexity.
Currently `rune build` seems to be single-threaded. Making the build …
-
I'm getting the following error during `make` inside the Docker container:
```
ERROR: /home/jonas/.cache/bazel/_bazel_jonas/9c55534d4233b9cdc493d7c1499d7160/external/org_tensorflow/tensorflow/core…
-
We can't publish the `rune-XXX` crates to crates.io because [a Rune project already exists](https://crates.io/search?q=rune) and it already has a bunch of `rune-XXX` crates.
I'm guessing the best s…