Spaces:
Sleeping
Sleeping
| from gliner import GLiNER | |
| def run_ner(model, text, labels_list, threshold=0.4): | |
| entities = model.predict_entities(text, labels_list, threshold=threshold) | |
| # Loading the GLiNER model | |
| model = GLiNER.from_pretrained("urchade/gliner_multi-v2.1") | |
| model.eval() # Put the model in evaluation mode | |
| # Initializing the dictionary to store the results | |
| ner_results = {label: [] for label in labels_list} | |
| # Iterating over the recognized entities and storing them in the dictionary | |
| for entity in entities: | |
| if entity['label'] in ner_results: | |
| # Adds the entity's text to the corresponding list for the label | |
| ner_results[entity['label']].append(entity['text']) | |
| return ner_results | |