/adou

Just some typical approaches for document understanding :book:

Primary LanguagePythonMIT LicenseMIT

Just some typical Approaches for Document Understanding and related tasks 📖


Image

  • Document Layout Analysis.
  • OCR pipeline.

Text ETL Process (Extract, Transform, Load)

  • Preprocess.
  • Vietnamese specific text mining toolbox.

Design Patterns

class MyModel(adou.Model, metaclass=adou.ModelType):
    __doc__ = ...
    schema = {'input':..., 'output':...}

    def load(self, *args, **kwargs)
    def summary(self, *args, **kwargs)
    def train(self, *args, **kwargs)
    def predict(self, *args, **kwargs)
class MyModelTestCase(unittest.TestCase):
    def test_case_01(self, *args, **kwargs)
    def test_case_02(self, *args, **kwargs)
    ...