Spaces:
Sleeping
Sleeping
| from .model import GLiNER | |
| # Initialize GLiNER with the base model | |
| model = GLiNER.from_pretrained("urchade/gliner_mediumv2.1") | |
| # Sample text for entity prediction | |
| text = """ | |
| lenskart m: (0)9428002330 Lenskart Store,Surat m: (0)9723817060) e:lenskartsurat@gmail.com Store Address UG-4.Ascon City.Opp.Maheshwari Bhavan,Citylight,Surat-395007""" | |
| # Labels for entity prediction | |
| # # Most GLiNER models should work best when entity types are in lower case or title case | |
| # labels = ["Person", "Mail", "Number", "Address", "Organization","Designation"] | |
| # # Perform entity prediction | |
| # entities = model.predict_entities(text, labels, threshold=0.5) | |
| def NER_Model(text): | |
| labels = ["Person", "Mail", "Number", "Address", "Organization","Designation","Link"] | |
| # Perform entity prediction | |
| entities = model.predict_entities(text, labels, threshold=0.5) | |
| # Initialize the processed data dictionary | |
| processed_data = { | |
| "Name": [], | |
| "Contact": [], | |
| "Designation": [], | |
| "Address": [], | |
| "Link": [], | |
| "Company": [], | |
| "Email": [], | |
| "extracted_text": "", | |
| } | |
| for entity in entities: | |
| print(entity["text"], "=>", entity["label"]) | |
| #loading the data into json | |
| if entity["label"]==labels[0]: | |
| processed_data['Name'].extend([entity["text"]]) | |
| if entity["label"]==labels[1]: | |
| processed_data['Email'].extend([entity["text"]]) | |
| if entity["label"]==labels[2]: | |
| processed_data['Contact'].extend([entity["text"]]) | |
| if entity["label"]==labels[3]: | |
| processed_data['Address'].extend([entity["text"]]) | |
| if entity["label"]==labels[4]: | |
| processed_data['Company'].extend([entity["text"]]) | |
| if entity["label"]==labels[5]: | |
| processed_data['Designation'].extend([entity["text"]]) | |
| if entity["label"]==labels[6]: | |
| processed_data['Link'].extend([entity["text"]]) | |
| processed_data['Address']=[', '.join(processed_data['Address'])] | |
| processed_data['extracted_text']=[text] | |
| return processed_data | |
| # result=NER_Model(text) | |
| # print(result) | |