Joint class generation on CIFAR 7to3 (test)
94.14AccuracyAC-GAN
Evaluation results
| Method | Links | |||||||
|---|---|---|---|---|---|---|---|---|
| AC-GANOracle=true, Backbone=BigGAN2022.06 | 94.14 | 16.14 | 8.44 | 74 | 63 | 89 | 0.84 | |
| cGAN-PDOracle=true, Backbone=BigGAN2022.06 | 87.6 | 16.55 | 8.39 | 73 | 65 | 89 | 0.85 | |
| cGAN-PD + S2M samplingBackbone=BigGAN, MCMC iterations=2002022.06 | 80.62 | 14.14 | 9.06 | 75 | 66 | 109 | 0.9 | |
| GAN + S2M samplingBackbone=BigGAN, MCMC iterations=2002022.06 | 77.65 | 14.42 | 8.71 | 72 | 67 | 92 | 0.87 | |
| CP-GANBackbone=BigGAN2022.06 | 42.62 | 27.88 | 7.42 | 71 | 66 | 76 | 0.71 | |
| AC-GAN*Oracle=false, Backbone=BigGAN2022.06 | 31.06 | 26.42 | 7.31 | 69 | 66 | 71 | 0.7 | |
| cGAN-PD*Oracle=false, Backbone=BigGAN2022.06 | 27.19 | 21.12 | 7.94 | 71 | 66 | 81 | 0.78 |