added description

README.md

@@ -5,17 +5,17 @@ tags:
 - Architecture
 ---
 
-## Model description
+# Tensorflow Keras implementation of : [Learning to tokenize in Vision Transformers](https://keras.io/examples/vision/token_learner/)
 
-More information needed
+The full credit goes to: [Aritra Roy Gosthipaty](https://twitter.com/ariG23498), [Sayak Paul](https://twitter.com/RisingSayak)
 
-## Intended uses & limitations
+## Short description:
 
-More information needed
+ViT and other Transformer based architectures represent the images as patches. As we increase the resolution of the images, the number of patches increase as well. To tackle this, Ryoo et al. introduced a new module called TokenLearner which can help reduce the number of patches used. The full paper can be found [here](https://openreview.net/forum?id=z-l1kpDXs88)
 
-## Training and evaluation data
+## Model and Dataset used
 
-More information needed
+The Dataset used here is CIFAR-10. The model used here is a mini ViT model with the TokenLearner module.
 
 ## Training procedure
 
@@ -36,6 +36,8 @@ The following hyperparameters were used during training:
 | exclude_from_weight_decay | None |
 | training_precision | float32 |
 
+## Training Metrics
+After 20 Epocs, the test accuracy of the model is 55.9% and the Top 5 test accuracy is 95.06%
 
  ## Model Plot