View Source Bumblebee.Vision.Vit (Bumblebee v0.5.3)

ViT model family.

Architectures

:base - plain ViT without any head on top
:for_image_classification - ViT with a classification head. The head consists of a single dense layer on top of the pooled features
:for_masked_image_modeling - ViT with a language modeling head on top for predicting visual tokens

"pixel_values" - {batch_size, image_size, image_size, num_channels}
Featurized image pixel values.
"patch_mask" - {batch_size, num_patches}
Mask to nullify selected embedded patches.

:image_size - the size of the input spatial dimensions. Defaults to 224
:num_channels - the number of channels in the input. Defaults to 3
:patch_size - the size of the patch spatial dimensions. Defaults to 16
:hidden_size - the dimensionality of hidden layers. Defaults to 768
:num_blocks - the number of Transformer blocks in the encoder. Defaults to 12
:num_attention_heads - the number of attention heads for each attention layer in the encoder. Defaults to 12
:use_qkv_bias - whether to use bias in query, key, and value projections. Defaults to true
:activation - the activation function. Defaults to :gelu
:dropout_rate - the dropout rate for encoder and decoder. Defaults to 0.0
:attention_dropout_rate - the dropout rate for attention weights. Defaults to 0.0
:layer_norm_epsilon - the epsilon used by the layer normalization layers. Defaults to 1.0e-12
:initializer_scale - the standard deviation of the normal initializer used for initializing kernel parameters. Defaults to 0.02
:output_hidden_states - whether the model should return all hidden states. Defaults to false
:output_attentions - whether the model should return all attentions. Defaults to false
:num_labels - the number of labels to use in the last layer for the classification task. Defaults to 2
:id_to_label - a map from class index to label. Defaults to %{}