Model-J: SupViT Model (model_idx_0698)
This model is part of the Model-J dataset, introduced in:
Learning on Model Weights using Tree Experts (CVPR 2025) by Eliahu Horwitz*, Bar Cavia*, Jonathan Kahana*, Yedid Hoshen
ProbeXModel Details
| Attribute | Value |
|---|
| Subset | SupViT |
| Split | test |
| Base Model | google/vit-base-patch16-224 |
| Dataset | CIFAR100 (50 classes) |
Training Hyperparameters
| Parameter | Value |
|---|
| Learning Rate | 0.0001 |
| LR Scheduler | cosine |
| Epochs | 6 |
| Max Train Steps | 1998 |
| Batch Size | 64 |
| Weight Decay | 0.03 |
| Seed | 698 |
| Random Crop | True |
| Random Flip | True |
Performance
| Metric | Value |
|---|
| Train Accuracy | 0.9991 |
| Val Accuracy | 0.9376 |
| Test Accuracy | 0.9364 |
Training Categories
The model was fine-tuned on the following 50 CIFAR100 classes:
maple_tree, bed, shark, sweet_pepper, mouse, otter, mushroom, rabbit, bowl, squirrel, man, dolphin, fox, forest, crab, hamster, pear, possum, dinosaur, train, cockroach, pickup_truck, keyboard, bear, girl, turtle, flatfish, seal, table, cup, lobster, lizard, tiger, camel, bottle, butterfly, whale, skunk, bus, ray, chimpanzee, sunflower, lawn_mower, willow_tree, plate, shrew, beaver, tulip, snake, bicycle
