README.md

---
base_model: facebook/dino-vitb16
library_name: transformers
pipeline_tag: image-classification
tags:
  - probex
  - model-j
  - weight-space-learning
---

# Model-J: DINO Model (model_idx_0976)

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

<p align="center">
    🌐 <a href="https://horwitz.ai/probex" target="_blank">Project</a> | 📃 <a href="https://arxiv.org/abs/2410.13569" target="_blank">Paper</a> | 💻 <a href="https://github.com/eliahuhorwitz/ProbeX" target="_blank">GitHub</a> | 🤗 <a href="https://huggingface.co/ProbeX" target="_blank">Dataset</a>
</p>

![ProbeX](https://raw.githubusercontent.com/eliahuhorwitz/ProbeX/main/imgs/poster.png)

## Model Details

| Attribute | Value |
|---|---|
| **Subset** | DINO |
| **Split** | train |
| **Base Model** | `facebook/dino-vitb16` |
| **Dataset** | CIFAR100 (50 classes) |

## Training Hyperparameters

| Parameter | Value |
|---|---|
| Learning Rate | 0.0001 |
| LR Scheduler | constant_with_warmup |
| Epochs | 2 |
| Max Train Steps | 666 |
| Batch Size | 64 |
| Weight Decay | 0.009 |
| Seed | 976 |
| Random Crop | True |
| Random Flip | False |

## Performance

| Metric | Value |
|---|---|
| Train Accuracy | 0.9101 |
| Val Accuracy | 0.8379 |
| Test Accuracy | 0.8422 |

## Training Categories

The model was fine-tuned on the following 50 CIFAR100 classes:

`bear`, `possum`, `kangaroo`, `palm_tree`, `couch`, `oak_tree`, `snake`, `beaver`, `sweet_pepper`, `lamp`, `crocodile`, `camel`, `rose`, `cup`, `cockroach`, `fox`, `trout`, `bottle`, `mushroom`, `bed`, `squirrel`, `telephone`, `plain`, `spider`, `wolf`, `bicycle`, `ray`, `apple`, `keyboard`, `porcupine`, `lion`, `cattle`, `mouse`, `pear`, `dolphin`, `castle`, `crab`, `otter`, `rocket`, `chimpanzee`, `lawn_mower`, `plate`, `beetle`, `bee`, `table`, `tulip`, `shark`, `bridge`, `can`, `cloud`