microsoft
/

beit-base-patch16-224-pt22k

Image Classification Transformers PyTorch JAX Safetensors

beit vision Inference Endpoints

Model card Files Files and versions Community

nielsr HF staff commited on Aug 14, 2021

Commit

0913454

•

1 Parent(s): d6894c7

Fix typo

Files changed (1) hide show

README.md +0 -1

README.md CHANGED Viewed

@@ -16,7 +16,6 @@ Disclaimer: The team releasing BEiT did not write a model card for this model so
 ## Model description
 The BEiT model is a Vision Transformer (ViT), which is a transformer encoder model (BERT-like). In contrast to the original ViT model, BEiT is pretrained on a large collection of images in a self-supervised fashion, namely ImageNet-21k, at a resolution of 224x224 pixels. The pre-training objective for the model is to predict visual tokens from the encoder of OpenAI's DALL-E's VQ-VAE, based on masked patches.
-Next, the model was fine-tuned in a supervised fashion on ImageNet (also referred to as ILSVRC2012), a dataset comprising 1 million images and 1,000 classes, also at resolution 224x224.
 Images are presented to the model as a sequence of fixed-size patches (resolution 16x16), which are linearly embedded. Contrary to the original ViT models, BEiT models do use relative position embeddings (similar to T5) instead of absolute position embeddings, and perform classification of images by mean-pooling the final hidden states of the patches, instead of placing a linear layer on top of the final hidden state of the [CLS] token.

 ## Model description
 The BEiT model is a Vision Transformer (ViT), which is a transformer encoder model (BERT-like). In contrast to the original ViT model, BEiT is pretrained on a large collection of images in a self-supervised fashion, namely ImageNet-21k, at a resolution of 224x224 pixels. The pre-training objective for the model is to predict visual tokens from the encoder of OpenAI's DALL-E's VQ-VAE, based on masked patches.
 Images are presented to the model as a sequence of fixed-size patches (resolution 16x16), which are linearly embedded. Contrary to the original ViT models, BEiT models do use relative position embeddings (similar to T5) instead of absolute position embeddings, and perform classification of images by mean-pooling the final hidden states of the patches, instead of placing a linear layer on top of the final hidden state of the [CLS] token.