Update README.md
Browse files
README.md
CHANGED
|
@@ -14,12 +14,6 @@ terms of accuracy vs. image throughput without extra training data !
|
|
| 14 |
class="center">
|
| 15 |
</p>
|
| 16 |
|
| 17 |
-
We introduce a new self-attention mechanism, denoted as Hierarchical
|
| 18 |
-
Attention (HAT), that captures both short and long-range information by learning
|
| 19 |
-
cross-window carrier tokens.
|
| 20 |
-
|
| 21 |
-
|
| 22 |
-
|
| 23 |
Note: Please use the [**latest NVIDIA TensorRT release**](https://docs.nvidia.com/deeplearning/tensorrt/container-release-notes/index.html) to enjoy the benefits of optimized FasterViT ops.
|
| 24 |
|
| 25 |
|
|
|
|
| 14 |
class="center">
|
| 15 |
</p>
|
| 16 |
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 17 |
Note: Please use the [**latest NVIDIA TensorRT release**](https://docs.nvidia.com/deeplearning/tensorrt/container-release-notes/index.html) to enjoy the benefits of optimized FasterViT ops.
|
| 18 |
|
| 19 |
|