|
gptkbp:instance_of
|
gptkb:television_channel
|
|
gptkbp:application
|
gptkb:Autonomous_Vehicles
gptkb:software
Image Recognition
Object Detection
Facial Recognition
Medical Imaging
Video Analysis
|
|
gptkbp:architectural_style
|
Image Classification
|
|
gptkbp:base
|
gptkb:Efficient_Net
|
|
gptkbp:content_type
|
Depthwise Separable Convolution
|
|
gptkbp:developed_by
|
gptkb:Google_AI
|
|
gptkbp:first_performance
|
84.4%
|
|
gptkbp:has_achievements
|
State-of-the-art performance
|
|
https://www.w3.org/2000/01/rdf-schema#label
|
Efficient Net B7
|
|
gptkbp:influenced_by
|
gptkb:Neural_Architecture_Search
Model Compression Techniques
Transfer Learning Techniques
|
|
gptkbp:input_output
|
600x600
|
|
gptkbp:is_available_in
|
gptkb:Graphics_Processing_Unit
gptkb:Py_Torch
|
|
gptkbp:is_compared_to
|
gptkb:Inception
gptkb:Dense_Net
gptkb:Res_Net
|
|
gptkbp:is_designed_for
|
Mobile and Edge Devices
|
|
gptkbp:is_evaluated_by
|
Top-1 Error Rate
Top-5 Error Rate
|
|
gptkbp:is_optimized_for
|
Compound Scaling
|
|
gptkbp:notable_feature
|
Low Latency
High Accuracy
Reduced Model Size
Efficient Scaling
|
|
gptkbp:orbital_period
|
66 million
|
|
gptkbp:performance
|
FLO Ps
|
|
gptkbp:rank
|
97.1%
|
|
gptkbp:related_to
|
gptkb:Efficient_Net_B0
gptkb:Efficient_Net_B1
gptkb:Efficient_Net_B2
gptkb:Efficient_Net_B3
gptkb:Efficient_Net_B4
gptkb:Efficient_Net_B5
gptkb:Efficient_Net_B6
|
|
gptkbp:release_year
|
gptkb:2019
|
|
gptkbp:research_areas
|
gptkb:viewpoint
gptkb:Artificial_Intelligence
gptkb:Deep_Learning
|
|
gptkbp:successor
|
Efficient Net B8
|
|
gptkbp:supports
|
gptkb:streaming_service
|
|
gptkbp:training
|
gptkb:Image_Net
Data Augmentation
Label Smoothing
Stochastic Depth
|
|
gptkbp:uses
|
gptkb:Batch_Normalization
gptkb:Dropout
Global Average Pooling
Squeeze-and-Excitation Blocks
Swish Activation Function
|
|
gptkbp:bfsParent
|
gptkb:Efficient_Net
|
|
gptkbp:bfsLayer
|
5
|