A series of convenience functions make your machine learning project easier
Project description
turnsole
A series of convenience functions make your machine learning project easier
安装方法
Latest release
pip install turnsole
项目暂不开源,因此该安装方法暂时不保证能用
Developer mode
pip install -e .
快速上手
PDF 操作
智能 PDF 文件转图片
智能的把 PDF 文件里面的插图找出来,例如没有插图就将整页 PDF 截图下来,也能智能的将碎图拼接在一起
Example:
# pdf_path 表示 PDF 文件的路径,输出 images 按页码进行汇总输出 images = turnsole.pdf_to_images(pdf_path)
图像操作工具箱
base64_to_bgr / bgr_to_base64
图像和 base64 互相转换
Example:
image = turnsole.base64_to_bgr(img64) img64 = turnsole.bgr_to_base64(image)
image_crop
根据 bbox 在 image 上进行切片,如果指定 perspective 为 True 则切片方式为透视变换(可以切旋转目标)
Example:
im_slice_no_perspective = turnsole.image_crop(image, bbox) im_slice = turnsole.image_crop(image, bbox, perspective=True)
Output:
OCR 引擎模块
OCR 引擎指的是一系列跟 OCR 相关的底层模型,我们提供了这些模型的函数式调用接口和标准 API
- ADC :tada:
- DBNet :tada:
- CRNN :tada:
- Object Detector :tada:
- Signature Detector :tada:
免费试用
import requests
results = requests.post(url=r'http://139.196.149.46:9001/gen_ocr', files={'file': open(file_path, 'rb')}).json()
ocr_results = results['ocr_results']
Prerequisites
由于 OCR 引擎模块依赖于底层神经网络模型,因此需要先用 Docker 挂载底层神经网络模型
首先把 ./model_repository 文件夹和里面的模型放到项目根目录下再启动,如果没有相关模型找 lvkui 要
使用起来非常简单,你只需要启动对应的 Docker 容器即可
docker run --gpus="device=0" --rm -p 8000:8000 -p 8001:8001 -p 8002:8002 -v $PWD/model_repository:/models nvcr.io/nvidia/tritonserver:21.10-py3 tritonserver --model-repository=/models
ADC
通用文件摆正算法
from turnsole.ocr_engine import angle_detector
image_rotated, direction = angle_detector.ADC(image, fine_degree=False)
DBNet
通用文字检测算法
from turnsole.ocr_engine import text_detector
boxes = text_detector.predict(image)
CRNN
通用文字识别算法
from turnsole.ocr_engine import text_recognizer
ocr_result, ocr_time = text_recognizer.predict_batch(image, boxes)
Object Detector
通用文件检测算法
from turnsole.ocr_engine import object_detector
object_list = object_detector.process(image)
Signature Detector
签字盖章二维码检测算法
from turnsole.ocr_engine import signature_detector
signature_list = signature_detector.process(image)
标准 API
python api/ocr_engine_server.py
Project details
Release history Release notifications | RSS feed
Download files
Download the file for your platform. If you're not sure which to choose, learn more about installing packages.
Source Distribution
Built Distribution
Filter files by name, interpreter, ABI, and platform.
If you're not sure about the file name format, learn more about wheel file names.
Copy a direct link to the current filters
File details
Details for the file turnsole-0.0.27.tar.gz.
File metadata
- Download URL: turnsole-0.0.27.tar.gz
- Upload date:
- Size: 56.6 kB
- Tags: Source
- Uploaded using Trusted Publishing? No
- Uploaded via: twine/4.0.1 CPython/3.8.0
File hashes
| Algorithm | Hash digest | |
|---|---|---|
| SHA256 |
e442f4b4b6a83bd713a240be3cffe18749e930027abc4fdb769b44ae9ad3cbdc
|
|
| MD5 |
09e1ae61e7b3afc785da755fadd625fe
|
|
| BLAKE2b-256 |
be9b79810163794c72c9c5f6e5a12f51727b32a8f991d9f006e5eb55a2e86ba2
|
File details
Details for the file turnsole-0.0.27-py3-none-any.whl.
File metadata
- Download URL: turnsole-0.0.27-py3-none-any.whl
- Upload date:
- Size: 62.9 kB
- Tags: Python 3
- Uploaded using Trusted Publishing? No
- Uploaded via: twine/4.0.1 CPython/3.8.0
File hashes
| Algorithm | Hash digest | |
|---|---|---|
| SHA256 |
f7fb6b66e23ff6c9fac23db08b50732824c77df5bd5ea2309d6e5f1c7bd1b772
|
|
| MD5 |
ac0c73c108bacabedf5b195187efd685
|
|
| BLAKE2b-256 |
fa6e30dfec47c9d6a1cfb07215041bcd2a50c6b6715d8496f4eac5424a542762
|