1aurent commited on
Commit
bf39e9d
1 Parent(s): a22385b

Update README.md

Browse files
Files changed (1) hide show
  1. README.md +62 -1
README.md CHANGED
@@ -1,8 +1,69 @@
1
  ---
2
  tags:
3
- - image-classification
4
  - timm
 
 
5
  library_name: timm
6
  license: apache-2.0
7
  ---
 
8
  # Model card for vit_large_patch14_224.dinobloom
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
  ---
2
  tags:
 
3
  - timm
4
+ - feature-extraction
5
+ - image-classification
6
  library_name: timm
7
  license: apache-2.0
8
  ---
9
+
10
  # Model card for vit_large_patch14_224.dinobloom
11
+
12
+ ![](https://github.com/marrlab/DinoBloom/blob/9ea2f950e1f016cd7f899b3ed025d12b6a355d9f/media/overview.png?raw=true)
13
+
14
+ ## Model Details
15
+
16
+ - **Model Type:** Feature backbone
17
+ - **Model Stats:**
18
+ - Params: 304M (large)
19
+ - Image size: 224 x 224 x 3
20
+ - Patch size: 14 x 14 x 3
21
+ - **Repository:** [github.com:marrlab/DinoBloom](https://github.com/marrlab/DinoBloom)
22
+ - **Original Weights:** [Zenodo](https://zenodo.org/records/10908163)
23
+ - **License:** [Apache License 2.0](https://github.com/marrlab/DinoBloom/blob/main/LICENSE)
24
+ - **Papers:**
25
+ - [DinoBloom: A Foundation Model for Generalizable Cell Embeddings in Hematology](https://arxiv.org/abs/2404.05022)
26
+
27
+ ## Model Usage
28
+
29
+ ### Image Embeddings
30
+
31
+ ```python
32
+ from urllib.request import urlopen
33
+ from PIL import Image
34
+ import timm
35
+
36
+ # get example histology image
37
+ img = Image.open(
38
+ urlopen(
39
+ "https://raw.githubusercontent.com/zxaoyou/segmentation_WBC/master/Dataset%201/001.bmp"
40
+ )
41
+ )
42
+
43
+ # load model from the hub
44
+ model = timm.create_model(
45
+ model_name="hf-hub:1aurent/vit_large_patch14_224.dinobloom",
46
+ pretrained=True,
47
+ ).eval()
48
+
49
+ # get model specific transforms (normalization, resize)
50
+ data_config = timm.data.resolve_model_data_config(model)
51
+ transforms = timm.data.create_transform(**data_config, is_training=False)
52
+
53
+ data = transforms(img).unsqueeze(0) # input is a (batch_size, num_channels, img_size, img_size) shaped tensor
54
+ output = model(data) # output is a (batch_size, num_features) shaped tensor
55
+ ```
56
+
57
+
58
+ ## Citation
59
+
60
+ ```bibtex
61
+ @misc{koch2024dinobloom,
62
+ title = {DinoBloom: A Foundation Model for Generalizable Cell Embeddings in Hematology},
63
+ author = {Valentin Koch and Sophia J. Wagner and Salome Kazeminia and Ece Sancar and Matthias Hehr and Julia Schnabel and Tingying Peng and Carsten Marr},
64
+ year = {2024},
65
+ eprint = {2404.05022},
66
+ archivePrefix = {arXiv},
67
+ primaryClass = {cs.CV}
68
+ }
69
+ ```