|
gptkbp:instance_of
|
gptkb:Transformers
|
|
gptkbp:application
|
Image Classification
Object Detection
Semantic Segmentation
|
|
gptkbp:architecture
|
gptkb:Hierarchical_Transformer
|
|
gptkbp:developed_by
|
gptkb:Microsoft_Research
|
|
gptkbp:evaluates
|
m AP
Io U
|
|
gptkbp:feature
|
gptkb:stage_adaptation
gptkb:Few-Shot_Learning
gptkb:Zero-Shot_Learning
gptkb:AI_technology
Scalability
Data Augmentation
Robustness
Knowledge Distillation
Domain Adaptation
Attention Mechanism
Ensemble Learning
Model Interpretability
Fine-tuning
Flexible Architecture
Data Efficiency
Model Compression
Efficient Computation
Multi-Task Learning
Self-Supervised Learning
Residual Connections
Contrastive Learning
Layer Normalization
Cross-Scale Attention
Dynamic Windowing
Generative Learning
Global Context Attention
Hierarchical Feature Maps
Multi-Scale Representation
Patch Embedding
|
|
https://www.w3.org/2000/01/rdf-schema#label
|
Swin-L
|
|
gptkbp:input_output
|
224x224
|
|
gptkbp:license
|
MIT License
|
|
gptkbp:orbital_period
|
over 200 million
|
|
gptkbp:performance
|
gptkb:PASCAL_VOC
gptkb:ADE20_K
gptkb:COCO
Cityscapes
Top-1 Accuracy
Top-5 Accuracy
LVIS
|
|
gptkbp:predecessor
|
gptkb:Swin-T
|
|
gptkbp:provides_information_on
|
gptkb:Image_Net
|
|
gptkbp:related_to
|
gptkb:Computer_Vision
gptkb:neural_networks
gptkb:Deep_Learning
gptkb:Swin_Transformer
gptkb:Vision_Transformers
|
|
gptkbp:repository
|
gptkb:Git_Hub
|
|
gptkbp:successor
|
gptkb:Swin-V
|
|
gptkbp:training
|
Supervised Learning
|
|
gptkbp:training_programs
|
gptkb:Py_Torch
|
|
gptkbp:uses
|
Shifted Windowing
|
|
gptkbp:year_established
|
gptkb:2021
|
|
gptkbp:bfsParent
|
gptkb:Swin_Transformer
|
|
gptkbp:bfsLayer
|
5
|