Add fimu-docproc-research/vitstr_small_DoctrOcrEngine model
Browse files- README.md +46 -0
- config.json +21 -0
- pytorch_model.bin +3 -0
README.md
ADDED
@@ -0,0 +1,46 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
1 |
+
|
2 |
+
---
|
3 |
+
language: cz
|
4 |
+
---
|
5 |
+
**Optical Character Recognition made seamless & accessible to anyone, powered by PyTorch**
|
6 |
+
|
7 |
+
## Task: recognition
|
8 |
+
### Example usage:
|
9 |
+
|
10 |
+
```python
|
11 |
+
>>> from doctr.io import DocumentFile
|
12 |
+
>>> from doctr.models import ocr_predictor, from_hub
|
13 |
+
|
14 |
+
>>> img = DocumentFile.from_images(['<image_path>'])
|
15 |
+
>>> # Load your model from the hub
|
16 |
+
>>> model = from_hub('mindee/my-model')
|
17 |
+
|
18 |
+
>>> # Pass it to the predictor
|
19 |
+
>>> # If your model is a recognition model:
|
20 |
+
>>> predictor = ocr_predictor(det_arch='db_resnet50',
|
21 |
+
>>> reco_arch=model,
|
22 |
+
>>> pretrained=True)
|
23 |
+
|
24 |
+
>>> # Get your predictions
|
25 |
+
>>> res = predictor(img)
|
26 |
+
```
|
27 |
+
|
28 |
+
Training configuration and logs: https://wandb.ai/xbankov/text-recognition
|
29 |
+
### Run Configuration
|
30 |
+
|
31 |
+
{
|
32 |
+
"hf_dataset_name": "fimu-docproc-research/born_digital_recognition",
|
33 |
+
"name": "vitstr_small_25_512_32_0.03481964801161507_0.05483515188085567_cosine_c368f4a3_92dfd51b",
|
34 |
+
"epochs": 25,
|
35 |
+
"lr": 0.03481964801161507,
|
36 |
+
"weight_decay": 0.05483515188085567,
|
37 |
+
"batch_size": 512,
|
38 |
+
"input_size": 32,
|
39 |
+
"sched": "cosine",
|
40 |
+
"sample": null,
|
41 |
+
"workers": 16,
|
42 |
+
"wb": true,
|
43 |
+
"push_to_hub": "fimu-docproc-research/vitstr_small",
|
44 |
+
"test_only": false,
|
45 |
+
"arch": "vitstr_small"
|
46 |
+
}
|
config.json
ADDED
@@ -0,0 +1,21 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
1 |
+
{
|
2 |
+
"mean": [
|
3 |
+
0.694,
|
4 |
+
0.695,
|
5 |
+
0.693
|
6 |
+
],
|
7 |
+
"std": [
|
8 |
+
0.299,
|
9 |
+
0.296,
|
10 |
+
0.301
|
11 |
+
],
|
12 |
+
"input_shape": [
|
13 |
+
3,
|
14 |
+
32,
|
15 |
+
128
|
16 |
+
],
|
17 |
+
"vocab": "®©äàőôöüűĺľŕĽÖÜߧ0123456789abcdefghijklmnopqrstuvwxyzABCDEFGHIJKLMNOPQRSTUVWXYZ!\"#$%&'()*+,-./:;<=>?@[\\]^_`{|}~°£€¥¢฿áčďéěíňóřšťúůýžÁČĎÉĚÍŇÓŘŠŤÚŮÝŽ",
|
18 |
+
"url": null,
|
19 |
+
"arch": "vitstr_small",
|
20 |
+
"task": "recognition"
|
21 |
+
}
|
pytorch_model.bin
ADDED
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
1 |
+
version https://git-lfs.github.com/spec/v1
|
2 |
+
oid sha256:5b947470496b60c222ce9ac8d1d14ba484aa942018d5b46b92da324fea53a46d
|
3 |
+
size 85800521
|