We will train NLP model that will extract needed information from text.
The code will prepare your data for training ner model. And it will guide you to train using your terminal, brilliant code BY ME as it makes annotations based on the labels you provide. No need to annotate manually if you have dataset with entity labels.
Result. Trained accuracy is 95%.
That one with real data was trained for extracting the dates and payment numbers from texts. It is much more powerfull than regular expressions in python. As Regex does not work for different string formats, the NER model came to the world.
