|
gptkbp:instanceOf
|
deep convolutional neural network architecture
|
|
gptkbp:activatedBy
|
gptkb:ReLU
|
|
gptkbp:architecture
|
residual network
|
|
gptkbp:author
|
gptkb:Shaoqing_Ren
gptkb:Xiangyu_Zhang
gptkb:Jian_Sun
gptkb:Kaiming_He
|
|
gptkbp:citation
|
over 100,000 (as of 2024)
|
|
gptkbp:developedBy
|
gptkb:Microsoft_Research
|
|
gptkbp:digestSize
|
224x224 pixels
|
|
gptkbp:hasVariant
|
gptkb:ResNet-101
gptkb:ResNet-152
gptkb:ResNet-18
gptkb:ResNet-34
|
|
https://www.w3.org/2000/01/rdf-schema#label
|
ResNet-50
|
|
gptkbp:influenced
|
gptkb:DenseNet
gptkb:ResNeXt
gptkb:ResNet-101
gptkb:ResNet-152
|
|
gptkbp:introduced
|
gptkb:Deep_Residual_Learning_for_Image_Recognition
|
|
gptkbp:introducedIn
|
2015
|
|
gptkbp:lastLayer
|
fully connected layer
|
|
gptkbp:level
|
50
|
|
gptkbp:normalization
|
batch normalization
|
|
gptkbp:notableFor
|
skip connections
enabling very deep networks
|
|
gptkbp:openSource
|
gptkb:GitHub
|
|
gptkbp:outputClasses
|
1000 (ImageNet)
|
|
gptkbp:parameter
|
~25.6 million
|
|
gptkbp:platform
|
gptkb:TensorFlow
gptkb:Keras
gptkb:PyTorch
|
|
gptkbp:pretrainedWeightsAvailable
|
gptkb:ImageNet
|
|
gptkbp:trainer
|
gptkb:ImageNet
|
|
gptkbp:trainingOptimizer
|
SGD
|
|
gptkbp:usedFor
|
feature extraction
image classification
object detection
|
|
gptkbp:usedIn
|
autonomous vehicles
transfer learning
face recognition
medical image analysis
video analysis
fine-tuning tasks
|
|
gptkbp:usedInBenchmarks
|
gptkb:COCO
gptkb:CIFAR-10
gptkb:CIFAR-100
|
|
gptkbp:uses
|
residual blocks
|
|
gptkbp:won
|
ILSVRC 2015 classification task
|
|
gptkbp:bfsParent
|
gptkb:Residual_Network
gptkb:EfficientNet-B2
|
|
gptkbp:bfsLayer
|
7
|