gptkbp:instance_of
|
gptkb:television_channel
|
gptkbp:bfsLayer
|
5
|
gptkbp:bfsParent
|
gptkb:Efficient_Net
|
gptkbp:application
|
gptkb:viewpoint
|
gptkbp:architectural_style
|
Image Classification
|
gptkbp:base
|
gptkb:Efficient_Net
|
gptkbp:class
|
gptkb:television_series
|
gptkbp:collaborations
|
gptkb:battle
|
gptkbp:community_support
|
Active
|
gptkbp:compression_ratio
|
Supported
|
gptkbp:content_type
|
Depthwise Separable Convolution
|
gptkbp:contribution
|
gptkb:battle
|
gptkbp:deployment
|
Cloud and Edge
|
gptkbp:developed_by
|
gptkb:Google_AI
|
gptkbp:established
|
Swish
|
gptkbp:features
|
gptkb:battle
|
gptkbp:first_performance
|
79.6%
|
gptkbp:has_documentation
|
Available
|
https://www.w3.org/2000/01/rdf-schema#label
|
Efficient Net B1
|
gptkbp:hyper_threading
|
Tunable
|
gptkbp:image
|
RGB Images
Top-1 Accuracy on Image Net
|
gptkbp:impact
|
gptkb:High
|
gptkbp:input_output
|
Softmax
Tensor
240x240
Normalization and Resizing
|
gptkbp:is_a_framework_for
|
gptkb:Graphics_Processing_Unit
|
gptkbp:is_evaluated_by
|
Accuracy
Cross-Validation
|
gptkbp:is_optimized_for
|
gptkb:Adam
|
gptkbp:license
|
Apache License 2.0
|
gptkbp:localization
|
gptkb:Batch_Normalization
|
gptkbp:migration
|
Supported
|
gptkbp:notable_models
|
gptkb:board_game
|
gptkbp:orbital_period
|
7.8 million
|
gptkbp:performance
|
FLO Ps
|
gptkbp:predecessor
|
gptkb:Efficient_Net_B0
|
gptkbp:provides_information_on
|
gptkb:Image_Net
gptkb:battle
|
gptkbp:publishes
|
Efficient Net: Rethinking Model Scaling for Convolutional Neural Networks
|
gptkbp:rank
|
94.5%
|
gptkbp:reduces
|
0.2
|
gptkbp:related_model
|
gptkb:Efficient_Net
gptkb:Tensor_Flow_Model_Garden
Grad-CAM
< 10 MB
B0, B2, B3, B4, B5, B6, B7
Mobile Net, Res Net
|
gptkbp:release_year
|
gptkb:2019
|
gptkbp:scales
|
Compound Scaling
|
gptkbp:sister_channel
|
gptkb:3
|
gptkbp:speed
|
Fast
|
gptkbp:successor
|
gptkb:Efficient_Net_B2
|
gptkbp:training
|
gptkb:streaming_service
gptkb:Keras
0.001
< 1 day
|
gptkbp:tuning
|
gptkb:battle
|
gptkbp:updates
|
gptkb:Regular
|
gptkbp:use_case
|
Image Recognition
|