In this notebook, we will be shown how we can train a higher resolution image for Vision Transformer model. Usually transformer based models with high resolution may not fit into our Graphic Processing Unit (GPU). In such cases, we can adopt a Trainable Resizer mechanism as a backbone of the transformer models and have it perform as a joint learning of the image resizer and recognition models simultaneously.
-
Notifications
You must be signed in to change notification settings - Fork 0
In this notebook, we will be shown how we can train a higher resolution image for **Vision Transformer** model. Usually transformer based models with high resolution may not fit into our Graphic Processing Unit (GPU). In such cases, we can adopt a **Trainable Resizer** mechanism as a backbone of the transformer models and have it perform as a jo…
cjayjonathan/Learning-to-Resize-Images-for-Vision-Transformer
Folders and files
| Name | Name | Last commit message | Last commit date | |
|---|---|---|---|---|
Repository files navigation
About
In this notebook, we will be shown how we can train a higher resolution image for **Vision Transformer** model. Usually transformer based models with high resolution may not fit into our Graphic Processing Unit (GPU). In such cases, we can adopt a **Trainable Resizer** mechanism as a backbone of the transformer models and have it perform as a jo…
Resources
Stars
Watchers
Forks
Releases
No releases published
Packages 0
No packages published