Trainable Activations for Image Classification

Open paper
Benchmark Model Rank Results
image-classification-on-mnistDNN-5 (Trainable Activations)#44Percentage error: 2.8Accuracy: 97.2Trainable Parameters: 575051
image-classification-on-mnistDNN-3 (Trainable Activations)#45Percentage error: 3.0Accuracy: 97.0Trainable Parameters: 386719
image-classification-on-mnistDNN-2 (Trainable Activations)#46Percentage error: 3.6Accuracy: 96.4Trainable Parameters: 311651
image-classification-on-cifar-10ResNet-26 (Trainable Activations)#226Percentage correct: 91.1
image-classification-on-cifar-10ResNet-32 (Trainable Activations)#227Percentage correct: 90.9
image-classification-on-cifar-10ResNet-44 (Trainable Activations)#231Percentage correct: 90.5
image-classification-on-cifar-10ResNet-20 (Trainable Activations)#232Percentage correct: 90.4
image-classification-on-cifar-10ResNet-14 (Trainable Activations)#237Percentage correct: 89.0
image-classification-on-cifar-10ResNet-56 (Trainable Activations)#240Percentage correct: 88.8
image-classification-on-cifar-10ResNet-8 (Trainable Activations)#248Percentage correct: 86.5

Rank counts only results with a code link.