nateraw commited on
Commit
c7ce0d9
1 Parent(s): 37b0c2c

Update README.md

Browse files
Files changed (1) hide show
  1. README.md +2 -4
README.md CHANGED
@@ -14,6 +14,8 @@ thumbnail: https://avatars3.githubusercontent.com/u/32437151?s=460&u=4ec59abc8d2
14
 
15
  Vision Transformer (ViT) model pre-trained on ImageNet-21k (14 million images, 21,843 classes) and **fine-tuned on CIFAR10** at resolution 224x224.
16
 
 
 
17
  ## Model description
18
 
19
  The Vision Transformer (ViT) is a transformer encoder model (BERT-like) pretrained on a large collection of images in a supervised fashion, namely ImageNet-21k, at a resolution of 224x224 pixels.
@@ -21,7 +23,3 @@ Images are presented to the model as a sequence of fixed-size patches (resolutio
21
  Note that this model does not provide any fine-tuned heads, as these were zero'd by Google researchers. However, the model does include the pre-trained pooler, which can be used for downstream tasks (such as image classification).
22
 
23
  By pre-training the model, it learns an inner representation of images that can then be used to extract features useful for downstream tasks: if you have a dataset of labeled images for instance, you can train a standard classifier by placing a linear layer on top of the pre-trained encoder. One typically places a linear layer on top of the [CLS] token, as the last hidden state of this token can be seen as a representation of an entire image.
24
-
25
- ## Training procedure
26
-
27
- The training procedure can be seen in [my github repo](https://github.com/nateraw/huggingface-vit-finetune)
 
14
 
15
  Vision Transformer (ViT) model pre-trained on ImageNet-21k (14 million images, 21,843 classes) and **fine-tuned on CIFAR10** at resolution 224x224.
16
 
17
+ Check out the code at my [my Github repo](https://github.com/nateraw/huggingface-vit-finetune).
18
+
19
  ## Model description
20
 
21
  The Vision Transformer (ViT) is a transformer encoder model (BERT-like) pretrained on a large collection of images in a supervised fashion, namely ImageNet-21k, at a resolution of 224x224 pixels.
 
23
  Note that this model does not provide any fine-tuned heads, as these were zero'd by Google researchers. However, the model does include the pre-trained pooler, which can be used for downstream tasks (such as image classification).
24
 
25
  By pre-training the model, it learns an inner representation of images that can then be used to extract features useful for downstream tasks: if you have a dataset of labeled images for instance, you can train a standard classifier by placing a linear layer on top of the pre-trained encoder. One typically places a linear layer on top of the [CLS] token, as the last hidden state of this token can be seen as a representation of an entire image.