Add RepVGG model. (#1561)

* Add RepVGG model.

* Add RepVGG README

* Extract var to top level

* Replace hashmap with a match

* Add a variant for the model kind + avoid some unnecessary config cloning.

---------

Co-authored-by: Laurent <laurent.mazare@gmail.com>
This commit is contained in:
Jani Monoses
2024-01-11 08:07:40 +02:00
committed by GitHub
parent 63944714f2
commit 2480c5dbdd
4 changed files with 438 additions and 0 deletions

View File

@ -0,0 +1,20 @@
# candle-repvgg
A candle implementation of inference using a pre-trained [repvgg](https://arxiv.org/abs/2101.03697).
This uses a classification head trained on the ImageNet dataset and returns the
probabilities for the top-5 classes.
## Running an example
```
$ cargo run --example repvgg --release -- --image candle-examples/examples/yolo-v8/assets/bike.jpg
loaded image Tensor[dims 3, 224, 224; f32]
model built
mountain bike, all-terrain bike, off-roader: 61.70%
bicycle-built-for-two, tandem bicycle, tandem: 33.14%
unicycle, monocycle : 4.88%
crash helmet : 0.15%
moped : 0.04%
```