ronnief1 commited on
Commit
77309bf
·
verified ·
1 Parent(s): 185234b

Push model using huggingface_hub.

Browse files
Files changed (2) hide show
  1. README.md +3 -68
  2. model.safetensors +2 -2
README.md CHANGED
@@ -3,73 +3,8 @@ pipeline_tag: image-classification
3
  tags:
4
  - model_hub_mixin
5
  - pytorch_model_hub_mixin
6
- - OCT
7
- - retinal-imaging
8
- - classification
9
  ---
10
 
11
- ## RetFiner-UrFound
12
- This repo contains the weights of RetFiner-UrFound from the paper [RetFiner: A Vision-Language Refinement Scheme for Retinal Foundation Models](https://arxiv.org/abs/2506.22149).
13
-
14
- Project page: [RetFiner](https://github.com/ronnief1/RetFiner)
15
-
16
- ### Required Libraries
17
- This model requires specific Python libraries:
18
- ```bash
19
- torch==2.4.1+cu118
20
- timm==0.4.12
21
- torchvision==0.19.1+cu118
22
- ```
23
-
24
- To use this model, please download the ViT implementation from [vit.py](https://github.com/ronnief1/RetFiner/blob/main/RetFiner/vit.py).
25
-
26
- Note: if you are using this for downstream inference, adjust the num_classes based on the target dataset.
27
-
28
- ```python
29
- import torch
30
- import torch.nn as nn
31
- from huggingface_hub import PyTorchModelHubMixin
32
- from vit import vit_base_patch16
33
-
34
-
35
- class RetFiner(nn.Module, PyTorchModelHubMixin):
36
- """RetFiner: Fine-tuned ViT models for retinal image analysis"""
37
-
38
- def __init__(self, model_name: str = "RetFiner-UrFound", num_classes: int = 2, **kwargs):
39
- super().__init__()
40
-
41
- self.model = vit_base_patch16(
42
- img_size=224,
43
- num_classes=num_classes,
44
- drop_path_rate=0.1,
45
- global_pool=False,
46
- use_proj=False,
47
- **kwargs
48
- )
49
-
50
- self.config = {
51
- "num_classes": num_classes,
52
- **kwargs
53
- }
54
-
55
- def forward(self, x):
56
- return self.model(x)
57
-
58
- model = RetFiner.from_pretrained('ronnief1/RetFiner-UrFound')
59
- ```
60
-
61
- ## Citation
62
-
63
- Please cite the original paper if you use this model:
64
-
65
- ```python
66
- @misc{fecso2025retfinervisionlanguagerefinementscheme,
67
- title={RetFiner: A Vision-Language Refinement Scheme for Retinal Foundation Models},
68
- author={Ronald Fecso and José Morano and Ursula Schmidt-Erfurth and Hrvoje Bogunović},
69
- year={2025},
70
- eprint={2506.22149},
71
- archivePrefix={arXiv},
72
- primaryClass={cs.CV},
73
- url={https://arxiv.org/abs/2506.22149},
74
- }
75
- ```
 
3
  tags:
4
  - model_hub_mixin
5
  - pytorch_model_hub_mixin
 
 
 
6
  ---
7
 
8
+ This model has been pushed to the Hub using the [PytorchModelHubMixin](https://huggingface.co/docs/huggingface_hub/package_reference/mixins#huggingface_hub.PyTorchModelHubMixin) integration:
9
+ - Library: ronnief1/RetFiner
10
+ - Docs: [More Information Needed]
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
model.safetensors CHANGED
@@ -1,3 +1,3 @@
1
  version https://git-lfs.github.com/spec/v1
2
- oid sha256:c96ebb8751171760197e19a4896f23ceb53a78113b9bae8cb36d64f7f5d98bf7
3
- size 343215776
 
1
  version https://git-lfs.github.com/spec/v1
2
+ oid sha256:9727f3662ef3fd159888882fb7078582695dc552ee4aeb40f84105d412fd44e2
3
+ size 343209448