Create README.md
Browse files
README.md
ADDED
@@ -0,0 +1,80 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
1 |
+
---
|
2 |
+
library_name: transformers
|
3 |
+
base_model: 1aurent/phikon-finetuned-lora-kather2016
|
4 |
+
tags:
|
5 |
+
- feature-extraction
|
6 |
+
- image-classification
|
7 |
+
- biology
|
8 |
+
- cancer
|
9 |
+
- owkin
|
10 |
+
- histology
|
11 |
+
model-index:
|
12 |
+
- name: owkin_pancancer
|
13 |
+
results:
|
14 |
+
- task:
|
15 |
+
type: image-classification
|
16 |
+
name: Image Classification
|
17 |
+
dataset:
|
18 |
+
name: 1aurent/Kather-texture-2016
|
19 |
+
type: image-classification
|
20 |
+
metrics:
|
21 |
+
- type: accuracy
|
22 |
+
value: 0.932
|
23 |
+
name: accuracy
|
24 |
+
verified: false
|
25 |
+
license: other
|
26 |
+
license_name: owkin-non-commercial
|
27 |
+
license_link: https://github.com/owkin/HistoSSLscaling/blob/main/LICENSE.txt
|
28 |
+
pipeline_tag: image-classification
|
29 |
+
datasets:
|
30 |
+
- 1aurent/Kather-texture-2016
|
31 |
+
metrics:
|
32 |
+
- accuracy
|
33 |
+
widget:
|
34 |
+
- src: >-
|
35 |
+
https://datasets-server.huggingface.co/assets/1aurent/Kather-texture-2016/--/default/train/0/image/image.jpg
|
36 |
+
example_title: adipose
|
37 |
+
---
|
38 |
+
|
39 |
+
# Model card for phikon-distil-vit-tiny-patch16-224-kather2016
|
40 |
+
|
41 |
+
This model is a distilled version of [owkin/phikon](https://huggingface.co/owkin/phikon) to a TinyViT on the [1aurent/Kather-texture-2016](https://huggingface.co/datasets/1aurent/Kather-texture-2016) dataset.
|
42 |
+
|
43 |
+
## Model Usage
|
44 |
+
|
45 |
+
### Image Classification
|
46 |
+
```python
|
47 |
+
from transformers import AutoModelForImageClassification, AutoImageProcessor
|
48 |
+
from urllib.request import urlopen
|
49 |
+
from PIL import Image
|
50 |
+
|
51 |
+
# get example histology image
|
52 |
+
img = Image.open(
|
53 |
+
urlopen(
|
54 |
+
"https://datasets-server.huggingface.co/assets/1aurent/Kather-texture-2016/--/default/train/0/image/image.jpg"
|
55 |
+
)
|
56 |
+
)
|
57 |
+
|
58 |
+
# load image_processor and model from the hub
|
59 |
+
model_name = "1aurent/phikon-distil-vit-tiny-patch16-224-kather2016"
|
60 |
+
image_processor = AutoImageProcessor.from_pretrained(model_name)
|
61 |
+
model = AutoModelForImageClassification.from_pretrained(model_name)
|
62 |
+
|
63 |
+
inputs = image_processor(img, return_tensors="pt")
|
64 |
+
outputs = model(**inputs)
|
65 |
+
```
|
66 |
+
|
67 |
+
## Citation
|
68 |
+
```bibtex
|
69 |
+
@article{Filiot2023.07.21.23292757,
|
70 |
+
author = {Alexandre Filiot and Ridouane Ghermi and Antoine Olivier and Paul Jacob and Lucas Fidon and Alice Mac Kain and Charlie Saillard and Jean-Baptiste Schiratti},
|
71 |
+
title = {Scaling Self-Supervised Learning for Histopathology with Masked Image Modeling},
|
72 |
+
elocation-id = {2023.07.21.23292757},
|
73 |
+
year = {2023},
|
74 |
+
doi = {10.1101/2023.07.21.23292757},
|
75 |
+
publisher = {Cold Spring Harbor Laboratory Press},
|
76 |
+
url = {https://www.medrxiv.org/content/early/2023/09/14/2023.07.21.23292757},
|
77 |
+
eprint = {https://www.medrxiv.org/content/early/2023/09/14/2023.07.21.23292757.full.pdf},
|
78 |
+
journal = {medRxiv}
|
79 |
+
}
|
80 |
+
```
|