-
This issue is to track the new design required for flash-attention on bottom-up optimization pipeline.
## Status
The most of the optimization passes has been finished and been checked in llvm-targ…
-
#### Work Expected From The Issue
Provide several optimizations to improve the performance of the video playback by making the following changes:
- [ ] Use an alternative allocator like `mimalloc` t…
-
### Misc discussion on performance
I've been running some simple tests on multi-node parallel pipeline with NCCL. I doubled the bandwidth between the nodes but saw no increase in t/s or throughput.…
-
## Summary
To be able to run performance analysis, we need timestamps between each stage to calculate the delta.
### Basic example
```json
{
...
"timestamp_stages": {
"timestamp_prefilt…
-
## Bevy version
https://github.com/bevyengine/bevy/commit/c5af1335eb48f990c71082787ff566fc793aab33
## Operating system & version
macOS Catalina, 10.15.7
AdapterInfo { name: "AMD Radeon RX 57…
-
Currently, the log pipeline employs dynamic dispatching within its workflow:
```
(dynamic) (dynamic)
Logger::emit(record) -----------…
-
Hey all,
Before I start investigating options for how to improve my build speed, I wanted to ask for feedback here.
The problem is my CI takes about 30 minutes to create a various docker images …
-
Create [benchmark harness](https://pub.dev/packages/benchmark_harness) and run it in the pipeline, so that we can know if changes make the performance worse.
-
**Is your feature request related to a problem? Please describe.**
#376 => Going the way of the user being able to define the individual metrics he wants and segment into periods for all kind of metr…
-
### Description
We have received exclusive data for speech-to-text (STT) from a specific speaker. The task is to fine-tune the model using both this speaker's training data and the base training data…