Model-J: SupViT Model (model_idx_0939)
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 | val |
| 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 | 0.01 |
| Seed | 939 |
| Random Crop | True |
| Random Flip | False |
Performance
| Metric | Value |
|---|
| Train Accuracy | 0.9865 |
| Val Accuracy | 0.9451 |
| Test Accuracy | 0.9408 |
Training Categories
The model was fine-tuned on the following 50 CIFAR100 classes:
keyboard, can, leopard, apple, spider, porcupine, bed, beaver, rocket, plain, mouse, lizard, telephone, seal, fox, skyscraper, shark, plate, tiger, clock, house, crocodile, lamp, streetcar, palm_tree, elephant, man, orange, wardrobe, poppy, road, girl, caterpillar, tractor, hamster, woman, tank, cattle, bicycle, butterfly, orchid, bus, motorcycle, snake, bridge, lobster, sea, snail, camel, possum
