turhancan97
commited on
Commit
·
42d8924
1
Parent(s):
b94f7a8
chore: Add bottom 10% and top 60% pretrained models for CIFAR10 and STL10 datasets
Browse files
README.md
CHANGED
@@ -15,7 +15,9 @@ This repository contains models trained using Masked Autoencoders (MAE) on CIFAR
|
|
15 |
### Models Included:
|
16 |
- **No Mode:** MAE trained on original images.
|
17 |
- **Bottom 25% Variance:** MAE trained to reconstruct images using components with the lowest 25% variance.
|
|
|
18 |
- **Top 75% Variance:** MAE trained to reconstruct images using components with the highest 75% variance.
|
|
|
19 |
|
20 |
## Model Details
|
21 |
### Dataset
|
|
|
15 |
### Models Included:
|
16 |
- **No Mode:** MAE trained on original images.
|
17 |
- **Bottom 25% Variance:** MAE trained to reconstruct images using components with the lowest 25% variance.
|
18 |
+
- **Bottom 10% Variance:** MAE trained to reconstruct images using components with the lowest 10% variance.
|
19 |
- **Top 75% Variance:** MAE trained to reconstruct images using components with the highest 75% variance.
|
20 |
+
- **Top 60% Variance:** MAE trained to reconstruct images using components with the highest 60% variance.
|
21 |
|
22 |
## Model Details
|
23 |
### Dataset
|