Model-J: SupViT Model (model_idx_0518)
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 | train |
| Base Model | google/vit-base-patch16-224 |
| Dataset | CIFAR100 (50 classes) |
Training Hyperparameters
| Parameter | Value |
|---|
| Learning Rate | 9e-05 |
| LR Scheduler | constant |
| Epochs | 5 |
| Max Train Steps | 1665 |
| Batch Size | 64 |
| Weight Decay | 0.03 |
| Seed | 518 |
| Random Crop | False |
| Random Flip | True |
Performance
| Metric | Value |
|---|
| Train Accuracy | 0.9784 |
| Val Accuracy | 0.9261 |
| Test Accuracy | 0.9150 |
Training Categories
The model was fine-tuned on the following 50 CIFAR100 classes:
pickup_truck, dinosaur, whale, road, wolf, sweet_pepper, shark, leopard, sunflower, streetcar, motorcycle, crab, camel, telephone, bee, lamp, tank, trout, tractor, mouse, turtle, rabbit, girl, willow_tree, aquarium_fish, maple_tree, lobster, ray, palm_tree, cup, beetle, house, snail, forest, sea, clock, cattle, woman, orange, bottle, crocodile, baby, bridge, oak_tree, kangaroo, pear, butterfly, lion, bed, shrew
