Image Classification On Svhn
Metrics
Percentage error
Results
Performance results of various models on this benchmark
Comparison Table
| Model Name | Percentage error |
|---|---|
| semi-supervised-learning-with-deep-generative-1 | 54.33 |
| auxiliary-deep-generative-models | 22.86 |
| renet-a-recurrent-neural-network-based | 2.4 |
| exact-how-to-train-your-accuracy | 2.21 |
| 190505393 | 1.2 |
| densely-connected-convolutional-networks | 1.59 |
| rethinking-recurrent-neural-networks-and | 1.0 |
| multi-digit-number-recognition-from-street | 2.2 |
| on-the-importance-of-normalisation-layers-in | 2.0 |
| Model 10 | 1.8 |
| competitive-multi-scale-convolution | 1.8 |
| vision-models-are-more-robust-and-fair-when | 13.6 |
| enhanced-image-classification-with-a-fast | 4.0 |
| unsupervised-representation-learning-with-1 | 22.48 |
| residual-networks-of-residual-networks | 1.59 |
| benchopt-reproducible-efficient-and | 2.65 |
| Model 17 | 1.9 |
| improved-techniques-for-training-gans | 8.11 |
| triplenet-a-low-computing-power-platform-of | - |
| batch-normalized-maxout-network-in-network | 1.8 |
| improved-regularization-of-convolutional | 1.30 |
| regularizing-neural-networks-via-adversarial | 2.30 |
| binaryconnect-training-deep-neural-networks | 2.2 |
| deeply-supervised-nets | 1.9 |
| regularizing-neural-networks-via-adversarial | 1.35 |
| wavemix-lite-a-resource-efficient-neural-1 | 1.58 |
| deep-complex-networks | 3.3 |
| mixmatch-a-holistic-approach-to-semi | 2.59 |
| fractalnet-ultra-deep-neural-networks-without | 2.01 |
| unsupervised-representation-learning-with-1 | 66.55 |
| drop-activation-implicit-parameter-reduction | 1.46 |
| fixup-initialization-residual-learning | 1.4 |
| maxout-networks | 2.5 |
| unsupervised-representation-learning-with-1 | 28.87 |
| wide-residual-networks | 1.7 |
| semi-supervised-learning-with-deep-generative-1 | 65.63 |
| eraserelu-a-simple-way-to-ease-the-training | 1.54 |
| wide-residual-networks | 1.54 |
| stochastic-pooling-for-regularization-of-deep | 2.8 |
| threshnet-an-efficient-densenet-using | - |
| auxiliary-deep-generative-models | 16.61 |
| generalizing-pooling-functions-in | 1.7 |
| network-in-network | 2.35 |
| semi-supervised-learning-with-deep-generative-1 | 36.02 |
| semi-supervised-learning-with-deep-generative-1 | 36.02 |
| how-important-is-weight-symmetry-in | 10.16 |
| loss-sensitive-generative-adversarial | 5.98 |
| wavemix-lite-a-resource-efficient-neural | 1.27 |
| stochastic-optimization-of-plain | 1.50 |
| colornet-investigating-the-importance-of | 1.11 |
| Model 51 | 4.9 |
| Model 52 | 1.9 |
| fast-autoaugment | 1.1 |
| deep-competitive-pathway-networks | 1.58 |
| preventing-manifold-intrusion-with-locality | 8.20 |
| deep-networks-with-stochastic-depth | 1.75 |
| training-neural-networks-with-local-error | 1.65 |
| enaet-self-trained-ensemble-autoencoding | 2.22 |
| unsupervised-representation-learning-with-1 | 77.93 |
| connection-reduction-is-all-you-need | - |
| automatic-data-augmentation-via-invariance | - |
| augmented-neural-odes | 16.5 |