---
license: cc-by-nc-4.0
tags:
  - medical-imaging
  - diabetic-retinopathy
  - domain-generalization
  - pytorch
pipeline_tag: image-classification
library_name: pytorch
---

# GradeEye baseline-3ch

This repository contains GradeEye CORN ordinal diabetic-retinopathy classifiers using three backbones (`convnext_tiny`, `deit3_small_patch16_384`, `maxvit_tiny_tf_384`) at 384x384 resolution, with 3-channel input and a 4-threshold ordinal head. The `_ema.safetensors` file is the **primary** artifact: the paper's reported evaluation metrics were generated using the EMA state dict. The unsuffixed `.safetensors` file is the corresponding raw `model_state_dict` secondary artifact.

## Checkpoints

| Primary EMA weights | Raw secondary weights | Architecture | Held-out fold | Best QWK | Epoch |
|---|---|---|---|---:|---:|
| `lodo_aptos_convnext_tiny_best_ema.safetensors` | `lodo_aptos_convnext_tiny_best.safetensors` | `convnext_tiny` | aptos | 0.7614 | 6 |
| `lodo_aptos_deit3_small_patch16_384_best_ema.safetensors` | `lodo_aptos_deit3_small_patch16_384_best.safetensors` | `deit3_small_patch16_384` | aptos | 0.7064 | 6 |
| `lodo_aptos_maxvit_tiny_tf_384_best_ema.safetensors` | `lodo_aptos_maxvit_tiny_tf_384_best.safetensors` | `maxvit_tiny_tf_384` | aptos | 0.6558 | 11 |
| `lodo_ddr_convnext_tiny_best_ema.safetensors` | `lodo_ddr_convnext_tiny_best.safetensors` | `convnext_tiny` | ddr | 0.7506 | 3 |
| `lodo_ddr_deit3_small_patch16_384_best_ema.safetensors` | `lodo_ddr_deit3_small_patch16_384_best.safetensors` | `deit3_small_patch16_384` | ddr | 0.7252 | 3 |
| `lodo_ddr_maxvit_tiny_tf_384_best_ema.safetensors` | `lodo_ddr_maxvit_tiny_tf_384_best.safetensors` | `maxvit_tiny_tf_384` | ddr | 0.5483 | 14 |
| `lodo_eyepacs_convnext_tiny_best_ema.safetensors` | `lodo_eyepacs_convnext_tiny_best.safetensors` | `convnext_tiny` | eyepacs | 0.7820 | 1 |
| `lodo_eyepacs_deit3_small_patch16_384_best_ema.safetensors` | `lodo_eyepacs_deit3_small_patch16_384_best.safetensors` | `deit3_small_patch16_384` | eyepacs | 0.7684 | 8 |
| `lodo_eyepacs_maxvit_tiny_tf_384_best_ema.safetensors` | `lodo_eyepacs_maxvit_tiny_tf_384_best.safetensors` | `maxvit_tiny_tf_384` | eyepacs | 0.6500 | 17 |
| `lodo_messidor2_convnext_tiny_best_ema.safetensors` | `lodo_messidor2_convnext_tiny_best.safetensors` | `convnext_tiny` | messidor2 | 0.7615 | 10 |
| `lodo_messidor2_deit3_small_patch16_384_best_ema.safetensors` | `lodo_messidor2_deit3_small_patch16_384_best.safetensors` | `deit3_small_patch16_384` | messidor2 | 0.7253 | 5 |
| `lodo_messidor2_maxvit_tiny_tf_384_best_ema.safetensors` | `lodo_messidor2_maxvit_tiny_tf_384_best.safetensors` | `maxvit_tiny_tf_384` | messidor2 | 0.5182 | 8 |

## Preprocessing

1. Resize the RGB fundus image to 384x384 using the same offline preprocessing pipeline.
2. Convert RGB to float in [0,1] and apply ImageNet normalization: mean=(0.485, 0.456, 0.406), std=(0.229, 0.224, 0.225).

## Loading

```python
import json
from modeling import load_model

config = json.load(open('config.json'))
model = load_model('lodo_eyepacs_convnext_tiny_best_ema.safetensors', config)
# model(x) returns CORN logits with shape (batch, 4)
```

Install `torch`, `timm`, and `safetensors`, and make the GradeEye source repository available on `PYTHONPATH`.

## Intended use and limitations

These weights are released for research and reproducibility only. They are not validated for clinical diagnosis or treatment decisions. Performance varies substantially by held-out dataset and should not be interpreted as clinical-grade generalization.

Source code and paper materials: [https://github.com/ahmed-farhanur-rashid/gradeeye](https://github.com/ahmed-farhanur-rashid/gradeeye).