cnocr 2.3.0.2__tar.gz → 2.3.0.3__tar.gz
This diff represents the content of publicly available package versions that have been released to one of the supported registries. The information contained in this diff is provided for informational purposes only and reflects changes between package versions as they appear in their respective public registries.
- {cnocr-2.3.0.2 → cnocr-2.3.0.3}/PKG-INFO +13 -13
- {cnocr-2.3.0.2 → cnocr-2.3.0.3}/README.md +12 -12
- {cnocr-2.3.0.2 → cnocr-2.3.0.3}/cnocr/__version__.py +1 -1
- {cnocr-2.3.0.2 → cnocr-2.3.0.3}/cnocr/app.py +1 -1
- {cnocr-2.3.0.2 → cnocr-2.3.0.3}/cnocr.egg-info/PKG-INFO +13 -13
- {cnocr-2.3.0.2 → cnocr-2.3.0.3}/LICENSE +0 -0
- {cnocr-2.3.0.2 → cnocr-2.3.0.3}/cnocr/__init__.py +0 -0
- {cnocr-2.3.0.2 → cnocr-2.3.0.3}/cnocr/classification/__init__.py +0 -0
- {cnocr-2.3.0.2 → cnocr-2.3.0.3}/cnocr/classification/dataset.py +0 -0
- {cnocr-2.3.0.2 → cnocr-2.3.0.3}/cnocr/classification/image_classifier.py +0 -0
- {cnocr-2.3.0.2 → cnocr-2.3.0.3}/cnocr/clf_cli.py +0 -0
- {cnocr-2.3.0.2 → cnocr-2.3.0.3}/cnocr/cli.py +0 -0
- {cnocr-2.3.0.2 → cnocr-2.3.0.3}/cnocr/cn_ocr.py +0 -0
- {cnocr-2.3.0.2 → cnocr-2.3.0.3}/cnocr/consts.py +0 -0
- {cnocr-2.3.0.2 → cnocr-2.3.0.3}/cnocr/data_utils/__init__.py +0 -0
- {cnocr-2.3.0.2 → cnocr-2.3.0.3}/cnocr/data_utils/aug.py +0 -0
- {cnocr-2.3.0.2 → cnocr-2.3.0.3}/cnocr/data_utils/block_shuffle.py +0 -0
- {cnocr-2.3.0.2 → cnocr-2.3.0.3}/cnocr/data_utils/transforms.py +0 -0
- {cnocr-2.3.0.2 → cnocr-2.3.0.3}/cnocr/data_utils/utils.py +0 -0
- {cnocr-2.3.0.2 → cnocr-2.3.0.3}/cnocr/dataset.py +0 -0
- {cnocr-2.3.0.2 → cnocr-2.3.0.3}/cnocr/dataset_utils.py +0 -0
- {cnocr-2.3.0.2 → cnocr-2.3.0.3}/cnocr/gradio_app.py +0 -0
- {cnocr-2.3.0.2 → cnocr-2.3.0.3}/cnocr/gradio_app2.py +0 -0
- {cnocr-2.3.0.2 → cnocr-2.3.0.3}/cnocr/label_cn.txt +0 -0
- {cnocr-2.3.0.2 → cnocr-2.3.0.3}/cnocr/label_number.txt +0 -0
- {cnocr-2.3.0.2 → cnocr-2.3.0.3}/cnocr/line_split.py +0 -0
- {cnocr-2.3.0.2 → cnocr-2.3.0.3}/cnocr/lr_scheduler.py +0 -0
- {cnocr-2.3.0.2 → cnocr-2.3.0.3}/cnocr/models/__init__.py +0 -0
- {cnocr-2.3.0.2 → cnocr-2.3.0.3}/cnocr/models/ctc.py +0 -0
- {cnocr-2.3.0.2 → cnocr-2.3.0.3}/cnocr/models/densenet.py +0 -0
- {cnocr-2.3.0.2 → cnocr-2.3.0.3}/cnocr/models/mobilenet.py +0 -0
- {cnocr-2.3.0.2 → cnocr-2.3.0.3}/cnocr/models/ocr_model.py +0 -0
- {cnocr-2.3.0.2 → cnocr-2.3.0.3}/cnocr/ppocr/__init__.py +0 -0
- {cnocr-2.3.0.2 → cnocr-2.3.0.3}/cnocr/ppocr/consts.py +0 -0
- {cnocr-2.3.0.2 → cnocr-2.3.0.3}/cnocr/ppocr/postprocess/__init__.py +0 -0
- {cnocr-2.3.0.2 → cnocr-2.3.0.3}/cnocr/ppocr/postprocess/rec_postprocess.py +0 -0
- {cnocr-2.3.0.2 → cnocr-2.3.0.3}/cnocr/ppocr/pp_recognizer.py +0 -0
- {cnocr-2.3.0.2 → cnocr-2.3.0.3}/cnocr/ppocr/utility.py +0 -0
- {cnocr-2.3.0.2 → cnocr-2.3.0.3}/cnocr/ppocr/utils/__init__.py +0 -0
- {cnocr-2.3.0.2 → cnocr-2.3.0.3}/cnocr/ppocr/utils/chinese_cht_dict.txt +0 -0
- {cnocr-2.3.0.2 → cnocr-2.3.0.3}/cnocr/ppocr/utils/en_dict.txt +0 -0
- {cnocr-2.3.0.2 → cnocr-2.3.0.3}/cnocr/ppocr/utils/ppocr_keys_v1.txt +0 -0
- {cnocr-2.3.0.2 → cnocr-2.3.0.3}/cnocr/recognizer.py +0 -0
- {cnocr-2.3.0.2 → cnocr-2.3.0.3}/cnocr/serve.py +0 -0
- {cnocr-2.3.0.2 → cnocr-2.3.0.3}/cnocr/trainer.py +0 -0
- {cnocr-2.3.0.2 → cnocr-2.3.0.3}/cnocr/utils.py +0 -0
- {cnocr-2.3.0.2 → cnocr-2.3.0.3}/cnocr.egg-info/SOURCES.txt +0 -0
- {cnocr-2.3.0.2 → cnocr-2.3.0.3}/cnocr.egg-info/dependency_links.txt +0 -0
- {cnocr-2.3.0.2 → cnocr-2.3.0.3}/cnocr.egg-info/entry_points.txt +0 -0
- {cnocr-2.3.0.2 → cnocr-2.3.0.3}/cnocr.egg-info/not-zip-safe +0 -0
- {cnocr-2.3.0.2 → cnocr-2.3.0.3}/cnocr.egg-info/requires.txt +0 -0
- {cnocr-2.3.0.2 → cnocr-2.3.0.3}/cnocr.egg-info/top_level.txt +0 -0
- {cnocr-2.3.0.2 → cnocr-2.3.0.3}/setup.cfg +0 -0
- {cnocr-2.3.0.2 → cnocr-2.3.0.3}/setup.py +0 -0
- {cnocr-2.3.0.2 → cnocr-2.3.0.3}/tests/test_cnocr.py +0 -0
- {cnocr-2.3.0.2 → cnocr-2.3.0.3}/tests/test_dataset.py +0 -0
- {cnocr-2.3.0.2 → cnocr-2.3.0.3}/tests/test_models.py +0 -0
- {cnocr-2.3.0.2 → cnocr-2.3.0.3}/tests/test_ppocr.py +0 -0
- {cnocr-2.3.0.2 → cnocr-2.3.0.3}/tests/test_pytorch.py +0 -0
- {cnocr-2.3.0.2 → cnocr-2.3.0.3}/tests/test_trainer.py +0 -0
- {cnocr-2.3.0.2 → cnocr-2.3.0.3}/tests/test_transforms.py +0 -0
|
@@ -1,6 +1,6 @@
|
|
|
1
1
|
Metadata-Version: 2.1
|
|
2
2
|
Name: cnocr
|
|
3
|
-
Version: 2.3.0.
|
|
3
|
+
Version: 2.3.0.3
|
|
4
4
|
Summary: Python3 package for Chinese/English OCR, with small pretrained models
|
|
5
5
|
Home-page: https://github.com/breezedeus/cnocr
|
|
6
6
|
Author: breezedeus
|
|
@@ -36,7 +36,7 @@ License-File: LICENSE
|
|
|
36
36
|
[](https://pepy.tech/project/cnocr)
|
|
37
37
|
[](https://visitorbadge.io/status?path=https%3A%2F%2Fgithub.com%2Fbreezedeus%2FCnOCR)
|
|
38
38
|
[](./LICENSE)
|
|
39
|
-
[](https://cnocr.readthedocs.io/zh/
|
|
39
|
+
[](https://cnocr.readthedocs.io/zh-cn/stable/?badge=latest)
|
|
40
40
|
[](https://badge.fury.io/py/cnocr)
|
|
41
41
|
[](https://github.com/breezedeus/cnocr)
|
|
42
42
|
[](https://github.com/breezedeus/cnocr)
|
|
@@ -44,12 +44,12 @@ License-File: LICENSE
|
|
|
44
44
|

|
|
45
45
|
[](https://twitter.com/breezedeus)
|
|
46
46
|
|
|
47
|
-
[📖 文档](https://cnocr.readthedocs.io/zh/
|
|
48
|
-
[🛠️ 安装](https://cnocr.readthedocs.io/zh/
|
|
49
|
-
[🧳 可用模型](https://cnocr.readthedocs.io/zh/
|
|
50
|
-
[🕹 模型训练](https://cnocr.readthedocs.io/zh/
|
|
47
|
+
[📖 文档](https://cnocr.readthedocs.io/zh-cn/stable/) |
|
|
48
|
+
[🛠️ 安装](https://cnocr.readthedocs.io/zh-cn/stable/install/) |
|
|
49
|
+
[🧳 可用模型](https://cnocr.readthedocs.io/zh-cn/stable/models/) |
|
|
50
|
+
[🕹 模型训练](https://cnocr.readthedocs.io/zh-cn/stable/train/) |
|
|
51
51
|
[🛀🏻 在线Demo](https://huggingface.co/spaces/breezedeus/CnOCR-Demo) |
|
|
52
|
-
[💬 交流群](https://
|
|
52
|
+
[💬 交流群](https://www.breezedeus.com/article/join-group)
|
|
53
53
|
|
|
54
54
|
</div>
|
|
55
55
|
|
|
@@ -88,7 +88,7 @@ License-File: LICENSE
|
|
|
88
88
|
|
|
89
89
|
|
|
90
90
|
|
|
91
|
-
[**CnOCR**](https://github.com/breezedeus/cnocr) 是 **Python 3** 下的**文字识别**(**Optical Character Recognition**,简称**OCR**)工具包,支持**简体中文**、**繁体中文**(部分模型)、**英文**和**数字**的常见字符识别,支持竖排文字的识别。自带了**20+个** [训练好的模型](https://cnocr.readthedocs.io/zh/
|
|
91
|
+
[**CnOCR**](https://github.com/breezedeus/cnocr) 是 **Python 3** 下的**文字识别**(**Optical Character Recognition**,简称**OCR**)工具包,支持**简体中文**、**繁体中文**(部分模型)、**英文**和**数字**的常见字符识别,支持竖排文字的识别。自带了**20+个** [训练好的模型](https://cnocr.readthedocs.io/zh-cn/stable/models/),适用于不同应用场景,安装后即可直接使用。同时,CnOCR也提供简单的[训练命令](https://cnocr.readthedocs.io/zh-cn/stable/train/)供使用者训练自己的模型。欢迎扫码加小助手为好友,备注 `ocr`,小助手会定期统一邀请大家入群:
|
|
92
92
|
|
|
93
93
|
<div align="center">
|
|
94
94
|
<img src="https://huggingface.co/datasets/breezedeus/cnocr-wx-qr-code/resolve/main/wx-qr-code.JPG" alt="微信群二维码" width="300px"/>
|
|
@@ -315,7 +315,7 @@ $ pip install cnocr[ort-cpu] -i https://mirrors.aliyun.com/pypi/simple
|
|
|
315
315
|
>
|
|
316
316
|
> 请使用 **Python3**(3.7.\*~3.10.\*之间的版本应该都行),没测过Python2下是否ok。
|
|
317
317
|
|
|
318
|
-
更多说明可见 [安装文档](https://cnocr.readthedocs.io/zh/
|
|
318
|
+
更多说明可见 [安装文档](https://cnocr.readthedocs.io/zh-cn/stable/install/)。
|
|
319
319
|
|
|
320
320
|
> **Warning**
|
|
321
321
|
>
|
|
@@ -331,7 +331,7 @@ $ pip install cnocr[ort-cpu] -i https://mirrors.aliyun.com/pypi/simple
|
|
|
331
331
|
$ docker pull breezedeus/cnocr:latest
|
|
332
332
|
```
|
|
333
333
|
|
|
334
|
-
更多说明可见 [安装文档](https://cnocr.readthedocs.io/zh/
|
|
334
|
+
更多说明可见 [安装文档](https://cnocr.readthedocs.io/zh-cn/stable/install/)。
|
|
335
335
|
|
|
336
336
|
|
|
337
337
|
|
|
@@ -421,7 +421,7 @@ print(ocr_out)
|
|
|
421
421
|
|
|
422
422
|
相比于 CnOCR V2.2.* 版本,**V2.3** 中的大部分模型都经过了重新训练和精调,精度比旧版模型更高。同时,加入了两个参数量更多的模型系列:
|
|
423
423
|
|
|
424
|
-
* `*-densenet_lite_246-gru_base`:优先供 **知识星球** [**CnOCR/CnSTD私享群**](https://t.zsxq.com/FEYZRJQ)
|
|
424
|
+
* `*-densenet_lite_246-gru_base`:优先供 **知识星球** [**CnOCR/CnSTD私享群**](https://t.zsxq.com/FEYZRJQ) 会员使用,后续会免费开源。
|
|
425
425
|
* `*-densenet_lite_666-gru_large`:**Pro 模型**,购买后可使用。购买链接见文档:
|
|
426
426
|
|
|
427
427
|
**V2.3** 中的模型按使用场景可以分为以下几大类:
|
|
@@ -433,7 +433,7 @@ print(ocr_out)
|
|
|
433
433
|
|
|
434
434
|
> 注意 ⚠️:以上说明仅供参考,具体选择模型时建议以实际效果为准。
|
|
435
435
|
|
|
436
|
-
更多说明见:[可用模型](https://cnocr.readthedocs.io/zh/
|
|
436
|
+
更多说明见:[可用模型](https://cnocr.readthedocs.io/zh-cn/stable/models/)。
|
|
437
437
|
|
|
438
438
|
| `rec_model_name` | PyTorch 版本 | ONNX 版本 | 模型原始来源 | 模型文件大小 | 支持语言 | 是否支持竖排文字识别 |
|
|
439
439
|
| ------------------------------------------------------------ | ------------ | --------- | ------------ | ------------ | ----------------------------------- | -------------------- |
|
|
@@ -477,7 +477,7 @@ print(ocr_out)
|
|
|
477
477
|
|
|
478
478
|
## 给作者来杯咖啡
|
|
479
479
|
|
|
480
|
-
开源不易,如果此项目对您有帮助,可以考虑 [给作者加点油🥤,鼓鼓气💪🏻](https://cnocr.readthedocs.io/zh/
|
|
480
|
+
开源不易,如果此项目对您有帮助,可以考虑 [给作者加点油🥤,鼓鼓气💪🏻](https://cnocr.readthedocs.io/zh-cn/stable/buymeacoffee/) 。
|
|
481
481
|
|
|
482
482
|
---
|
|
483
483
|
|
|
@@ -6,7 +6,7 @@
|
|
|
6
6
|
[](https://pepy.tech/project/cnocr)
|
|
7
7
|
[](https://visitorbadge.io/status?path=https%3A%2F%2Fgithub.com%2Fbreezedeus%2FCnOCR)
|
|
8
8
|
[](./LICENSE)
|
|
9
|
-
[](https://cnocr.readthedocs.io/zh/
|
|
9
|
+
[](https://cnocr.readthedocs.io/zh-cn/stable/?badge=latest)
|
|
10
10
|
[](https://badge.fury.io/py/cnocr)
|
|
11
11
|
[](https://github.com/breezedeus/cnocr)
|
|
12
12
|
[](https://github.com/breezedeus/cnocr)
|
|
@@ -14,12 +14,12 @@
|
|
|
14
14
|

|
|
15
15
|
[](https://twitter.com/breezedeus)
|
|
16
16
|
|
|
17
|
-
[📖 文档](https://cnocr.readthedocs.io/zh/
|
|
18
|
-
[🛠️ 安装](https://cnocr.readthedocs.io/zh/
|
|
19
|
-
[🧳 可用模型](https://cnocr.readthedocs.io/zh/
|
|
20
|
-
[🕹 模型训练](https://cnocr.readthedocs.io/zh/
|
|
17
|
+
[📖 文档](https://cnocr.readthedocs.io/zh-cn/stable/) |
|
|
18
|
+
[🛠️ 安装](https://cnocr.readthedocs.io/zh-cn/stable/install/) |
|
|
19
|
+
[🧳 可用模型](https://cnocr.readthedocs.io/zh-cn/stable/models/) |
|
|
20
|
+
[🕹 模型训练](https://cnocr.readthedocs.io/zh-cn/stable/train/) |
|
|
21
21
|
[🛀🏻 在线Demo](https://huggingface.co/spaces/breezedeus/CnOCR-Demo) |
|
|
22
|
-
[💬 交流群](https://
|
|
22
|
+
[💬 交流群](https://www.breezedeus.com/article/join-group)
|
|
23
23
|
|
|
24
24
|
</div>
|
|
25
25
|
|
|
@@ -58,7 +58,7 @@
|
|
|
58
58
|
|
|
59
59
|
|
|
60
60
|
|
|
61
|
-
[**CnOCR**](https://github.com/breezedeus/cnocr) 是 **Python 3** 下的**文字识别**(**Optical Character Recognition**,简称**OCR**)工具包,支持**简体中文**、**繁体中文**(部分模型)、**英文**和**数字**的常见字符识别,支持竖排文字的识别。自带了**20+个** [训练好的模型](https://cnocr.readthedocs.io/zh/
|
|
61
|
+
[**CnOCR**](https://github.com/breezedeus/cnocr) 是 **Python 3** 下的**文字识别**(**Optical Character Recognition**,简称**OCR**)工具包,支持**简体中文**、**繁体中文**(部分模型)、**英文**和**数字**的常见字符识别,支持竖排文字的识别。自带了**20+个** [训练好的模型](https://cnocr.readthedocs.io/zh-cn/stable/models/),适用于不同应用场景,安装后即可直接使用。同时,CnOCR也提供简单的[训练命令](https://cnocr.readthedocs.io/zh-cn/stable/train/)供使用者训练自己的模型。欢迎扫码加小助手为好友,备注 `ocr`,小助手会定期统一邀请大家入群:
|
|
62
62
|
|
|
63
63
|
<div align="center">
|
|
64
64
|
<img src="https://huggingface.co/datasets/breezedeus/cnocr-wx-qr-code/resolve/main/wx-qr-code.JPG" alt="微信群二维码" width="300px"/>
|
|
@@ -285,7 +285,7 @@ $ pip install cnocr[ort-cpu] -i https://mirrors.aliyun.com/pypi/simple
|
|
|
285
285
|
>
|
|
286
286
|
> 请使用 **Python3**(3.7.\*~3.10.\*之间的版本应该都行),没测过Python2下是否ok。
|
|
287
287
|
|
|
288
|
-
更多说明可见 [安装文档](https://cnocr.readthedocs.io/zh/
|
|
288
|
+
更多说明可见 [安装文档](https://cnocr.readthedocs.io/zh-cn/stable/install/)。
|
|
289
289
|
|
|
290
290
|
> **Warning**
|
|
291
291
|
>
|
|
@@ -301,7 +301,7 @@ $ pip install cnocr[ort-cpu] -i https://mirrors.aliyun.com/pypi/simple
|
|
|
301
301
|
$ docker pull breezedeus/cnocr:latest
|
|
302
302
|
```
|
|
303
303
|
|
|
304
|
-
更多说明可见 [安装文档](https://cnocr.readthedocs.io/zh/
|
|
304
|
+
更多说明可见 [安装文档](https://cnocr.readthedocs.io/zh-cn/stable/install/)。
|
|
305
305
|
|
|
306
306
|
|
|
307
307
|
|
|
@@ -391,7 +391,7 @@ print(ocr_out)
|
|
|
391
391
|
|
|
392
392
|
相比于 CnOCR V2.2.* 版本,**V2.3** 中的大部分模型都经过了重新训练和精调,精度比旧版模型更高。同时,加入了两个参数量更多的模型系列:
|
|
393
393
|
|
|
394
|
-
* `*-densenet_lite_246-gru_base`:优先供 **知识星球** [**CnOCR/CnSTD私享群**](https://t.zsxq.com/FEYZRJQ)
|
|
394
|
+
* `*-densenet_lite_246-gru_base`:优先供 **知识星球** [**CnOCR/CnSTD私享群**](https://t.zsxq.com/FEYZRJQ) 会员使用,后续会免费开源。
|
|
395
395
|
* `*-densenet_lite_666-gru_large`:**Pro 模型**,购买后可使用。购买链接见文档:
|
|
396
396
|
|
|
397
397
|
**V2.3** 中的模型按使用场景可以分为以下几大类:
|
|
@@ -403,7 +403,7 @@ print(ocr_out)
|
|
|
403
403
|
|
|
404
404
|
> 注意 ⚠️:以上说明仅供参考,具体选择模型时建议以实际效果为准。
|
|
405
405
|
|
|
406
|
-
更多说明见:[可用模型](https://cnocr.readthedocs.io/zh/
|
|
406
|
+
更多说明见:[可用模型](https://cnocr.readthedocs.io/zh-cn/stable/models/)。
|
|
407
407
|
|
|
408
408
|
| `rec_model_name` | PyTorch 版本 | ONNX 版本 | 模型原始来源 | 模型文件大小 | 支持语言 | 是否支持竖排文字识别 |
|
|
409
409
|
| ------------------------------------------------------------ | ------------ | --------- | ------------ | ------------ | ----------------------------------- | -------------------- |
|
|
@@ -447,7 +447,7 @@ print(ocr_out)
|
|
|
447
447
|
|
|
448
448
|
## 给作者来杯咖啡
|
|
449
449
|
|
|
450
|
-
开源不易,如果此项目对您有帮助,可以考虑 [给作者加点油🥤,鼓鼓气💪🏻](https://cnocr.readthedocs.io/zh/
|
|
450
|
+
开源不易,如果此项目对您有帮助,可以考虑 [给作者加点油🥤,鼓鼓气💪🏻](https://cnocr.readthedocs.io/zh-cn/stable/buymeacoffee/) 。
|
|
451
451
|
|
|
452
452
|
---
|
|
453
453
|
|
|
@@ -158,7 +158,7 @@ def main():
|
|
|
158
158
|
|
|
159
159
|
st.markdown('# 开源Python OCR工具 ' '[CnOCR](https://github.com/breezedeus/cnocr)')
|
|
160
160
|
st.markdown('> 详细说明参见:[CnOCR 文档](https://cnocr.readthedocs.io/) ;'
|
|
161
|
-
'欢迎加入 [交流群](https://
|
|
161
|
+
'欢迎加入 [交流群](https://www.breezedeus.com/article/join-group) ;'
|
|
162
162
|
'作者:[breezedeus](https://github.com/breezedeus) 。')
|
|
163
163
|
st.markdown('')
|
|
164
164
|
st.subheader('选择待检测图片')
|
|
@@ -1,6 +1,6 @@
|
|
|
1
1
|
Metadata-Version: 2.1
|
|
2
2
|
Name: cnocr
|
|
3
|
-
Version: 2.3.0.
|
|
3
|
+
Version: 2.3.0.3
|
|
4
4
|
Summary: Python3 package for Chinese/English OCR, with small pretrained models
|
|
5
5
|
Home-page: https://github.com/breezedeus/cnocr
|
|
6
6
|
Author: breezedeus
|
|
@@ -36,7 +36,7 @@ License-File: LICENSE
|
|
|
36
36
|
[](https://pepy.tech/project/cnocr)
|
|
37
37
|
[](https://visitorbadge.io/status?path=https%3A%2F%2Fgithub.com%2Fbreezedeus%2FCnOCR)
|
|
38
38
|
[](./LICENSE)
|
|
39
|
-
[](https://cnocr.readthedocs.io/zh/
|
|
39
|
+
[](https://cnocr.readthedocs.io/zh-cn/stable/?badge=latest)
|
|
40
40
|
[](https://badge.fury.io/py/cnocr)
|
|
41
41
|
[](https://github.com/breezedeus/cnocr)
|
|
42
42
|
[](https://github.com/breezedeus/cnocr)
|
|
@@ -44,12 +44,12 @@ License-File: LICENSE
|
|
|
44
44
|

|
|
45
45
|
[](https://twitter.com/breezedeus)
|
|
46
46
|
|
|
47
|
-
[📖 文档](https://cnocr.readthedocs.io/zh/
|
|
48
|
-
[🛠️ 安装](https://cnocr.readthedocs.io/zh/
|
|
49
|
-
[🧳 可用模型](https://cnocr.readthedocs.io/zh/
|
|
50
|
-
[🕹 模型训练](https://cnocr.readthedocs.io/zh/
|
|
47
|
+
[📖 文档](https://cnocr.readthedocs.io/zh-cn/stable/) |
|
|
48
|
+
[🛠️ 安装](https://cnocr.readthedocs.io/zh-cn/stable/install/) |
|
|
49
|
+
[🧳 可用模型](https://cnocr.readthedocs.io/zh-cn/stable/models/) |
|
|
50
|
+
[🕹 模型训练](https://cnocr.readthedocs.io/zh-cn/stable/train/) |
|
|
51
51
|
[🛀🏻 在线Demo](https://huggingface.co/spaces/breezedeus/CnOCR-Demo) |
|
|
52
|
-
[💬 交流群](https://
|
|
52
|
+
[💬 交流群](https://www.breezedeus.com/article/join-group)
|
|
53
53
|
|
|
54
54
|
</div>
|
|
55
55
|
|
|
@@ -88,7 +88,7 @@ License-File: LICENSE
|
|
|
88
88
|
|
|
89
89
|
|
|
90
90
|
|
|
91
|
-
[**CnOCR**](https://github.com/breezedeus/cnocr) 是 **Python 3** 下的**文字识别**(**Optical Character Recognition**,简称**OCR**)工具包,支持**简体中文**、**繁体中文**(部分模型)、**英文**和**数字**的常见字符识别,支持竖排文字的识别。自带了**20+个** [训练好的模型](https://cnocr.readthedocs.io/zh/
|
|
91
|
+
[**CnOCR**](https://github.com/breezedeus/cnocr) 是 **Python 3** 下的**文字识别**(**Optical Character Recognition**,简称**OCR**)工具包,支持**简体中文**、**繁体中文**(部分模型)、**英文**和**数字**的常见字符识别,支持竖排文字的识别。自带了**20+个** [训练好的模型](https://cnocr.readthedocs.io/zh-cn/stable/models/),适用于不同应用场景,安装后即可直接使用。同时,CnOCR也提供简单的[训练命令](https://cnocr.readthedocs.io/zh-cn/stable/train/)供使用者训练自己的模型。欢迎扫码加小助手为好友,备注 `ocr`,小助手会定期统一邀请大家入群:
|
|
92
92
|
|
|
93
93
|
<div align="center">
|
|
94
94
|
<img src="https://huggingface.co/datasets/breezedeus/cnocr-wx-qr-code/resolve/main/wx-qr-code.JPG" alt="微信群二维码" width="300px"/>
|
|
@@ -315,7 +315,7 @@ $ pip install cnocr[ort-cpu] -i https://mirrors.aliyun.com/pypi/simple
|
|
|
315
315
|
>
|
|
316
316
|
> 请使用 **Python3**(3.7.\*~3.10.\*之间的版本应该都行),没测过Python2下是否ok。
|
|
317
317
|
|
|
318
|
-
更多说明可见 [安装文档](https://cnocr.readthedocs.io/zh/
|
|
318
|
+
更多说明可见 [安装文档](https://cnocr.readthedocs.io/zh-cn/stable/install/)。
|
|
319
319
|
|
|
320
320
|
> **Warning**
|
|
321
321
|
>
|
|
@@ -331,7 +331,7 @@ $ pip install cnocr[ort-cpu] -i https://mirrors.aliyun.com/pypi/simple
|
|
|
331
331
|
$ docker pull breezedeus/cnocr:latest
|
|
332
332
|
```
|
|
333
333
|
|
|
334
|
-
更多说明可见 [安装文档](https://cnocr.readthedocs.io/zh/
|
|
334
|
+
更多说明可见 [安装文档](https://cnocr.readthedocs.io/zh-cn/stable/install/)。
|
|
335
335
|
|
|
336
336
|
|
|
337
337
|
|
|
@@ -421,7 +421,7 @@ print(ocr_out)
|
|
|
421
421
|
|
|
422
422
|
相比于 CnOCR V2.2.* 版本,**V2.3** 中的大部分模型都经过了重新训练和精调,精度比旧版模型更高。同时,加入了两个参数量更多的模型系列:
|
|
423
423
|
|
|
424
|
-
* `*-densenet_lite_246-gru_base`:优先供 **知识星球** [**CnOCR/CnSTD私享群**](https://t.zsxq.com/FEYZRJQ)
|
|
424
|
+
* `*-densenet_lite_246-gru_base`:优先供 **知识星球** [**CnOCR/CnSTD私享群**](https://t.zsxq.com/FEYZRJQ) 会员使用,后续会免费开源。
|
|
425
425
|
* `*-densenet_lite_666-gru_large`:**Pro 模型**,购买后可使用。购买链接见文档:
|
|
426
426
|
|
|
427
427
|
**V2.3** 中的模型按使用场景可以分为以下几大类:
|
|
@@ -433,7 +433,7 @@ print(ocr_out)
|
|
|
433
433
|
|
|
434
434
|
> 注意 ⚠️:以上说明仅供参考,具体选择模型时建议以实际效果为准。
|
|
435
435
|
|
|
436
|
-
更多说明见:[可用模型](https://cnocr.readthedocs.io/zh/
|
|
436
|
+
更多说明见:[可用模型](https://cnocr.readthedocs.io/zh-cn/stable/models/)。
|
|
437
437
|
|
|
438
438
|
| `rec_model_name` | PyTorch 版本 | ONNX 版本 | 模型原始来源 | 模型文件大小 | 支持语言 | 是否支持竖排文字识别 |
|
|
439
439
|
| ------------------------------------------------------------ | ------------ | --------- | ------------ | ------------ | ----------------------------------- | -------------------- |
|
|
@@ -477,7 +477,7 @@ print(ocr_out)
|
|
|
477
477
|
|
|
478
478
|
## 给作者来杯咖啡
|
|
479
479
|
|
|
480
|
-
开源不易,如果此项目对您有帮助,可以考虑 [给作者加点油🥤,鼓鼓气💪🏻](https://cnocr.readthedocs.io/zh/
|
|
480
|
+
开源不易,如果此项目对您有帮助,可以考虑 [给作者加点油🥤,鼓鼓气💪🏻](https://cnocr.readthedocs.io/zh-cn/stable/buymeacoffee/) 。
|
|
481
481
|
|
|
482
482
|
---
|
|
483
483
|
|
|
File without changes
|
|
File without changes
|
|
File without changes
|
|
File without changes
|
|
File without changes
|
|
File without changes
|
|
File without changes
|
|
File without changes
|
|
File without changes
|
|
File without changes
|
|
File without changes
|
|
File without changes
|
|
File without changes
|
|
File without changes
|
|
File without changes
|
|
File without changes
|
|
File without changes
|
|
File without changes
|
|
File without changes
|
|
File without changes
|
|
File without changes
|
|
File without changes
|
|
File without changes
|
|
File without changes
|
|
File without changes
|
|
File without changes
|
|
File without changes
|
|
File without changes
|
|
File without changes
|
|
File without changes
|
|
File without changes
|
|
File without changes
|
|
File without changes
|
|
File without changes
|
|
File without changes
|
|
File without changes
|
|
File without changes
|
|
File without changes
|
|
File without changes
|
|
File without changes
|
|
File without changes
|
|
File without changes
|
|
File without changes
|
|
File without changes
|
|
File without changes
|
|
File without changes
|
|
File without changes
|
|
File without changes
|
|
File without changes
|
|
File without changes
|
|
File without changes
|
|
File without changes
|
|
File without changes
|
|
File without changes
|
|
File without changes
|
|
File without changes
|