Model-J: DINO Model (model_idx_0779)
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 | DINO |
| Split | train |
| Base Model | facebook/dino-vitb16 |
| Dataset | CIFAR100 (50 classes) |
Training Hyperparameters
| Parameter | Value |
|---|
| Learning Rate | 0.0005 |
| LR Scheduler | cosine |
| Epochs | 9 |
| Max Train Steps | 2997 |
| Batch Size | 64 |
| Weight Decay | 0.005 |
| Seed | 779 |
| Random Crop | True |
| Random Flip | False |
Performance
| Metric | Value |
|---|
| Train Accuracy | 0.5094 |
| Val Accuracy | 0.4053 |
| Test Accuracy | 0.4096 |
Training Categories
The model was fine-tuned on the following 50 CIFAR100 classes:
can, woman, possum, butterfly, skunk, sweet_pepper, leopard, bed, dolphin, girl, whale, tulip, clock, sea, table, streetcar, lawn_mower, shrew, fox, bicycle, dinosaur, mountain, tiger, lion, cockroach, boy, camel, tractor, snake, cloud, pear, baby, keyboard, beaver, mouse, television, worm, trout, caterpillar, raccoon, orchid, rocket, maple_tree, bottle, house, bee, elephant, train, ray, castle
