|
gptkbp:instance_of
|
gptkb:neural_networks
|
|
gptkbp:application
|
gptkb:Computer_Vision
gptkb:Yes
|
|
gptkbp:architecture
|
Image Classification
|
|
gptkbp:available_in
|
gptkb:Keras
gptkb:Py_Torch
gptkb:Tensor_Flow_Hub
|
|
gptkbp:base_model
|
gptkb:Efficient_Net
|
|
gptkbp:cloud_integration
|
Compatible
|
|
gptkbp:community_support
|
gptkb:Strong
|
|
gptkbp:contribution
|
gptkb:Yes
|
|
gptkbp:developed_by
|
gptkb:Google_AI
|
|
gptkbp:drops
|
0.3
|
|
gptkbp:edge_devices
|
Compatible
|
|
gptkbp:evaluates
|
gptkb:Yes
|
|
gptkbp:feature_extractor
|
gptkb:Yes
|
|
gptkbp:flops
|
9.9 billion
|
|
gptkbp:future_prospects
|
Ongoing
|
|
gptkbp:has_publications
|
Efficient Net: Rethinking Model Scaling for Convolutional Neural Networks
|
|
https://www.w3.org/2000/01/rdf-schema#label
|
Efficient Net B5
|
|
gptkbp:image_classification_task
|
gptkb:Yes
|
|
gptkbp:image_input
|
RGB Images
|
|
gptkbp:impact
|
gptkb:Significant
|
|
gptkbp:input_output
|
456x456
|
|
gptkbp:is_a_framework_for
|
gptkb:Tensor_Flow
|
|
gptkbp:is_cited_in
|
Highly Cited
|
|
gptkbp:is_open_source
|
gptkb:Yes
|
|
gptkbp:is_optimized_for
|
gptkb:Adam
|
|
gptkbp:is_standardized_by
|
Included
|
|
gptkbp:is_taught_in
|
Adaptive
|
|
gptkbp:losses
|
Categorical Crossentropy
|
|
gptkbp:marketed_as
|
gptkb:Yes
|
|
gptkbp:migration
|
Supported
|
|
gptkbp:model
|
gptkb:Yes
12 MB
|
|
gptkbp:model_comparison
|
Compared with Res Net and Dense Net
|
|
gptkbp:model_deployment
|
Supported
|
|
gptkbp:model_download
|
Available
|
|
gptkbp:object_detection_task
|
No
|
|
gptkbp:optimization_technique
|
Compound Scaling
|
|
gptkbp:orbital_period
|
30 million
|
|
gptkbp:output_layer
|
Softmax
|
|
gptkbp:performance
|
Top Performer
FLOPs
|
|
gptkbp:predecessor
|
gptkb:Efficient_Net_B4
|
|
gptkbp:pretrained_models
|
Available
|
|
gptkbp:provides_information_on
|
gptkb:Image_Net
Used
|
|
gptkbp:real_world_usage
|
Widely Used
|
|
gptkbp:regularization
|
L2 Regularization
|
|
gptkbp:release_year
|
gptkb:2019
|
|
gptkbp:semantic_segmentation_task
|
No
|
|
gptkbp:successor
|
gptkb:Efficient_Net_B6
|
|
gptkbp:top1_accuracy
|
88.4%
|
|
gptkbp:top5_accuracy
|
97.1%
|
|
gptkbp:training
|
Longer than Efficient Net B4
|
|
gptkbp:tutorials
|
gptkb:Yes
|
|
gptkbp:uses
|
Swish Activation Function
|
|
gptkbp:bfsParent
|
gptkb:Efficient_Net
|
|
gptkbp:bfsLayer
|
5
|