-
Hi,
I recently found your project and have started to build a learning app using Ebisu to time the repetitions. I have a few questions specifically concerned with version 3 of Ebisu.
I'll just li…
-
## 🚀 Feature
We have added MKLDNN+AMD BLIS path for PyTorch and want to upstream our changes to master branch
## Motivation
PyTorch with MKLDNN+AMD BLIS would give higher performance for DNN …
-
cpu信息：
```
Architecture: x86_64
Model name: AMD Ryzen 7 5700G with Radeon Graphics
```
日志：
```
[08/21 13:38:53 fvcore.common.checkpoint]: [Checkpoi…
-
### 🐛 Describe the bug
Given the code below, the custom backend is only invoked during forward graph compilation and not backwards. "MyBackend called" is print only once (logs below)
```
import l…
-
Results should be sorted (see google doc). A solution that seems reasonable is to use [Intl.Collator](https://developer.mozilla.org/fr/docs/Web/JavaScript/Reference/Objets_globaux/Collator) with the `…
eroux updated
3 years ago
-
## Introduction
This RFC provides a solution which completely separates software (drivers, subsystems, RAM/ROM layout etc.) from hardware (physical devices, interrupts, cores, etc.)
This solves:…
-
**Description**
xLARFGP computes a Householder reflector. Given an input vector `x` of dimension `m`, xLARFGP computes `H = I - β vv^*` such that
* `H x = ‖x‖ e1` and
* `v(1) = 1`,
where `β` …
-
- [x] Implement new matrix class `ad.matrix.convex_relu` that takes in `mat` (original matrix), `mask` (D matrices).
- [x] Implement NN for Boston Housing dataset and train with adamw.
- [x] Implem…
-
Please, before you create a new bug report, please make sure you searched in open and closed issues and couldn't find anything that matches.
**Printer type** - [Prusa XL 5 toolhead (upgraded from 2…
-
Hi, one of the developers from Luxonis here :slightly_smiling_face:
We'd be happy to discuss what improvements could be made regarding running RTABMap together with DepthAI-based devices to provide …