mirror of
https://github.com/huggingface/candle.git
synced 2025-06-16 18:48:51 +00:00
Readme updates. (#1134)
This commit is contained in:
20
candle-examples/examples/vit/README.md
Normal file
20
candle-examples/examples/vit/README.md
Normal file
@ -0,0 +1,20 @@
|
||||
# candle-vit
|
||||
|
||||
Vision Transformer (ViT) model implementation following the lines of
|
||||
[vit-base-patch16-224](https://huggingface.co/google/vit-base-patch16-224)
|
||||
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 vit --release -- --image tiger.jpg
|
||||
|
||||
loaded image Tensor[dims 3, 224, 224; f32]
|
||||
model built
|
||||
tiger, Panthera tigris : 100.00%
|
||||
tiger cat : 0.00%
|
||||
jaguar, panther, Panthera onca, Felis onca: 0.00%
|
||||
leopard, Panthera pardus: 0.00%
|
||||
lion, king of beasts, Panthera leo: 0.00%
|
||||
```
|
Reference in New Issue
Block a user