|
gptkbp:instance_of
|
gptkb:Transformers
|
|
gptkbp:application
|
Image Classification
|
|
gptkbp:architectural_style
|
gptkb:Transformers
|
|
gptkbp:architecture
|
gptkb:Transformers
|
|
gptkbp:attention_mechanism
|
Multi-head Attention
|
|
gptkbp:coat_of_arms
|
gptkb:32
|
|
gptkbp:collaborator
|
Various Universities
|
|
gptkbp:community_support
|
Active
|
|
gptkbp:computational_cost
|
gptkb:High
|
|
gptkbp:contribution
|
Advancement of Vision Transformers
|
|
gptkbp:developed_by
|
gptkb:Google_Research
|
|
gptkbp:drops
|
0.1
|
|
gptkbp:economic_impact
|
gptkb:Significant
|
|
gptkbp:evaluates
|
F1 Score
|
|
gptkbp:feature_extractor
|
gptkb:Yes
|
|
gptkbp:field_of_study
|
gptkb:Computer_Vision
|
|
gptkbp:future_prospects
|
Ongoing
|
|
gptkbp:gpu
|
gptkb:Yes
|
|
gptkbp:has_publications
|
Vision Transformers for Image Recognition
|
|
https://www.w3.org/2000/01/rdf-schema#label
|
Vi T-H
|
|
gptkbp:hyperparameters
|
Tunable
|
|
gptkbp:influenced_by
|
gptkb:BERT
|
|
gptkbp:initiated_by
|
Ge LU
|
|
gptkbp:input_output
|
224x224
|
|
gptkbp:invention
|
gptkb:None
|
|
gptkbp:is_a_framework_for
|
gptkb:Tensor_Flow
gptkb:Py_Torch
|
|
gptkbp:is_cited_in
|
gptkb:High
|
|
gptkbp:is_open_source
|
gptkb:Yes
|
|
gptkbp:is_optimized_for
|
gptkb:Adam
|
|
gptkbp:is_taught_in
|
0.001
|
|
gptkbp:is_trained_in
|
gptkb:Image_Net
|
|
gptkbp:losses
|
Cross-entropy Loss
|
|
gptkbp:migration
|
Supported
|
|
gptkbp:model
|
Large
Vi T-B, Vi T-L
|
|
gptkbp:num_parameters
|
632 million
|
|
gptkbp:orbital_period
|
over 600 million
|
|
gptkbp:output_activation
|
Softmax
|
|
gptkbp:performance
|
Available
Top-1 Accuracy
|
|
gptkbp:provides_information_on
|
gptkb:CIFAR-10
gptkb:Yes
|
|
gptkbp:related_to
|
gptkb:Deep_Learning
|
|
gptkbp:resolution
|
1000 classes
|
|
gptkbp:sister_channel
|
gptkb:3
|
|
gptkbp:successor
|
Vi T-G
|
|
gptkbp:training
|
Days
Self-supervised Learning
|
|
gptkbp:tuning
|
Possible
|
|
gptkbp:user_base
|
Researchers, Developers
|
|
gptkbp:uses
|
Self-Attention Mechanism
|
|
gptkbp:uses_attention
|
gptkb:Yes
|
|
gptkbp:uses_patch_embedding
|
gptkb:Yes
|
|
gptkbp:uses_positional_encoding
|
gptkb:Yes
|
|
gptkbp:year_established
|
gptkb:2021
|
|
gptkbp:bfsParent
|
gptkb:Transformers
|
|
gptkbp:bfsLayer
|
4
|