|
gptkbp:instance_of
|
gptkb:television_channel
|
|
gptkbp:bfsLayer
|
5
|
|
gptkbp:bfsParent
|
gptkb:Google_Net
gptkb:Inception-v2
|
|
gptkbp:architectural_style
|
Lightweight
|
|
gptkbp:data_usage
|
gptkb:theorem
|
|
gptkbp:developed_by
|
gptkb:Alexey_Dosovitskiy
gptkb:Forest_Agostinelli
gptkb:Deep_Scale
Dmitry Kolesnikov
Trevor Darrell
|
|
gptkbp:established
|
Re LU
|
|
gptkbp:has_method
|
1.24 million
|
|
gptkbp:has_programs
|
gptkb:Autonomous_Vehicles
Object Detection
Facial Recognition
Image Segmentation
Medical Imaging
|
|
gptkbp:has_variants
|
gptkb:Squeeze_Net_1.0
gptkb:Squeeze_Net_1.1
|
|
https://www.w3.org/2000/01/rdf-schema#label
|
Squeeze Net
|
|
gptkbp:industry
|
gptkb:musician
gptkb:software
gptkb:helicopter
Smart Cameras
Real-time Image Processing
|
|
gptkbp:influenced_by
|
gptkb:Alex_Net
|
|
gptkbp:input_output
|
1000 classes
227x227 pixels
|
|
gptkbp:is_analyzed_in
|
gptkb:streaming_service
gptkb:Neural_Architecture_Search
Model Compression
|
|
gptkbp:is_available_in
|
gptkb:Caffe_framework
Tensor Flow framework
Py Torch framework
|
|
gptkbp:is_compared_to
|
gptkb:Mobile_Net
gptkb:Res_Net
gptkb:Alex_Net
VGG Net
|
|
gptkbp:is_evaluated_by
|
gptkb:CIFAR-10
gptkb:Pascal_VOC
gptkb:SVHN
gptkb:COCO
gptkb:MNIST
gptkb:CIFAR-100
Image Net validation set
|
|
gptkbp:is_implemented_in
|
gptkb:C++
gptkb:Library
|
|
gptkbp:is_influenced_by
|
gptkb:Goog_Le_Net
gptkb:Alex_Net
VGG Net
|
|
gptkbp:is_known_for
|
gptkb:military_unit
Efficiency
High Accuracy
Low Memory Usage
|
|
gptkbp:is_open_source
|
gptkb:theorem
|
|
gptkbp:is_optimized_for
|
gptkb:smartphone
Mobile and embedded devices
|
|
gptkbp:is_part_of
|
gptkb:viewpoint
Deep Learning Models
CNN Family
|
|
gptkbp:is_supported_by
|
gptkb:Graphics_Processing_Unit
gptkb:Keras
gptkb:Py_Torch
gptkb:ARM_processors
NVIDIAGP Us
Intel CP Us
|
|
gptkbp:is_used_in
|
gptkb:robot
gptkb:computer
gptkb:Telecommunications_company
gptkb:helicopter
Environmental monitoring
Image Classification
Autonomous vehicles
Fashion industry
Image segmentation
Security surveillance
Feature extraction
Augmented reality applications
Sports analytics
Gaming industry
Real-time applications
Social media applications
Manufacturing automation
Object detection
Retail analytics
Agriculture technology
Healthcare imaging
|
|
gptkbp:license
|
Apache License 2.0
|
|
gptkbp:module
|
gptkb:theorem
|
|
gptkbp:notable_feature
|
Small model size
|
|
gptkbp:orbital_period
|
50x Fewer Parameters than Alex Net
|
|
gptkbp:performance
|
Model compression techniques
Computer Vision Tasks
Alex Net-level Accuracy
Top-5 accuracy of 57.5% on Image Net
|
|
gptkbp:primary_source
|
Image Classification
|
|
gptkbp:related_model
|
<0.5 MB
|
|
gptkbp:release_year
|
gptkb:2016
|
|
gptkbp:successor
|
gptkb:Squeeze_Net_1.1
|
|
gptkbp:training
|
gptkb:Image_Net_dataset
gptkb:Image_Net_Dataset
|
|
gptkbp:uses
|
Global Average Pooling
Fire modules
|
|
gptkbp:year_created
|
gptkb:2016
|