Semi Supervised Image Classification On 2
評価指標
Top 1 Accuracy
Top 5 Accuracy
評価結果
このベンチマークにおける各モデルのパフォーマンス結果
比較表
| モデル名 | Top 1 Accuracy | Top 5 Accuracy |
|---|---|---|
| weakly-supervised-contrastive-learning-1 | 72.0% | 91.2% |
| debiasing-calibrating-and-improving-semi | 74.1% | 91.5 |
| sequencematch-revisiting-the-design-of-weak | 75.2% | 91.9 |
| 190503670 | - | 82.41% |
| 190503670 | - | 83.39% |
| semi-supervised-learning-of-visual-features | 75.5% | - |
| semi-supervised-vision-transformers-at-scale | 83.3% | - |
| 190503670 | - | 82.78% |
| a-simple-framework-for-contrastive-learning | - | 87.8% |
| online-bag-of-visual-words-generation-for | - | 90.7% |
| 190503670 | - | 81.01% |
| 190503670 | - | 81.01% |
| big-self-supervised-models-are-strong-semi | 80.1% | 95.0% |
| 190503670 | - | 78.53% |
| big-self-supervised-models-are-strong-semi | 68.4% | 89.2% |
| unsupervised-feature-learning-via-non-1 | - | 77.40% |
| self-supervised-sequence-to-sequence-asr | - | 83.82% |
| meta-co-training-two-views-are-better-than | 85.8% | - |
| simmatchv2-semi-supervised-learning-with | 76.2% | - |
| semi-supervised-learning-of-visual-features | 77.8% | - |
| debiasing-calibrating-and-improving-semi | 75.3% | 92.6 |
| 190503670 | 73.21% | 91.23% |
| self-supervised-pretraining-of-visual | 77.9% | - |
| flexmatch-boosting-semi-supervised-learning | 64.79% | 86.04% |
| 190503670 | - | 91.23% |
| semi-supervised-learning-of-visual-features | 79.0% | - |
| 190503670 | - | 83.82% |
| 190503670 | - | 83.72% |
| 190503670 | - | 82.78% |
| milking-cowmask-for-semi-supervised-image | 73.94% | 91.24% |
| big-self-supervised-models-are-strong-semi | 80.2% | 95.0% |
| synco-synthetic-hard-negatives-in-contrastive | 66.6% | 88.0% |
| a-simple-framework-for-contrastive-learning | - | 92.6% |
| self-supervised-pretraining-of-visual | 76.7% | - |
| 190503670 | - | 83.39% |
| dual-student-breaking-the-limits-of-the | 63.52% | 83.58% |
| learning-customized-visual-models-with | 85.1% | - |
| repetitive-reprediction-deep-decipher-for | - | 90.48% |
| comatch-semi-supervised-learning-with | 73.7% | 91.4% |
| semi-supervised-vision-transformers-at-scale | 77.1% | - |
| self-supervised-learning-by-estimating-twin-1 | 75.3% | 92.8% |
| unsupervised-feature-learning-via-non-1 | - | 77.4% |
| semi-supervised-vision-transformers | 75.5% | - |
| vne-an-effective-method-for-improving-deep | 69.1 | 89.9 |
| mean-teachers-are-better-role-models-weight | - | 90.89% |
| meta-pseudo-labels | 73.89% | 91.38% |
| semi-supervised-vision-transformers-at-scale | 79.7% | - |
| barlow-twins-self-supervised-learning-via | 69.7% | 89.3 |
| unsupervised-data-augmentation-1 | - | 88.52 |
| np-match-when-neural-processes-meet-semi | 58.22% | - |
| 190503670 | - | 78.53% |
| a-simple-framework-for-contrastive-learning | - | 91.2% |
| semi-supervised-vision-transformers-at-scale | 84.3% | 96.6% |
| simmatch-semi-supervised-learning-with | 74.4% | - |
| big-self-supervised-models-are-strong-semi | 77.5% | 93.4% |
| data-efficient-image-recognition-with | 73.1% | 91.2% |
| big-self-supervised-models-are-strong-semi | 73.9% | 91.9% |
| 190503670 | - | 83.72% |
| exponential-moving-average-normalization-for | 74% | - |
| big-self-supervised-models-are-strong-semi | 80.9% | 95.5% |
| representation-learning-with-contrastive | - | 84.88% |
| 190503670 | - | 82.41% |
| pushing-the-limits-of-self-supervised-resnets | 72.4% | 91.2% |
| with-a-little-help-from-my-friends-nearest | 69.8% | 89.3 |
| large-scale-adversarial-representation | - | 78.8% |
| vision-models-are-more-robust-and-fair-when | 78.8% | - |
| self-supervised-learning-of-pretext-invariant | - | 83.8% |
| fixmatch-simplifying-semi-supervised-learning | - | 89.13% |