Views
No views yet
pip install transformers torch1from transformers import AutoTokenizer, AutoModelForTokenClassification, pipeline
2
3# Load model and tokenizer
4model_name = "Mukesh97/Bert-NER"
5tokenizer = AutoTokenizer.from_pretrained(model_name)
6model = AutoModelForTokenClassification.from_pretrained(model_name)
7
8# Create NER pipeline
9ner_pipeline = pipeline(
10 "ner",
11 model=model,
12 tokenizer=tokenizer,
13 aggregation_strategy="simple"
14)
15
16# Example usage
17text = "My name is John Smith and I work at Google in New York."
18results = ner_pipeline(text)
19
20for entity in results:
21 print(f"{entity['word']}: {entity['entity_group']} (confidence: {entity['score']:.3f})")1from transformers import AutoTokenizer, AutoModelForTokenClassification
2import torch
3
4# Load model
5tokenizer = AutoTokenizer.from_pretrained("Mukesh97/Bert-NER")
6model = AutoModelForTokenClassification.from_pretrained("Mukesh97/Bert-NER")
7
8# Tokenize input
9text = "Apple Inc. was founded by Steve Jobs in California."
10inputs = tokenizer(text, return_tensors="pt", truncation=True, padding=True)
11
12# Get predictions
13with torch.no_grad():
14 outputs = model(**inputs)
15 predictions = torch.argmax(outputs.logits, dim=-1)
16
17# Decode predictions
18tokens = tokenizer.convert_ids_to_tokens(inputs["input_ids"][0])
19predicted_labels = [model.config.id2label[pred.item()] for pred in predictions[0]]
20
21# Display results
22for token, label in zip(tokens, predicted_labels):
23 if token not in ['[CLS]', '[SEP]', '[PAD]']:
24 print(f"{token}: {label}")1# Process multiple texts efficiently
2texts = [
3 "Barack Obama was the President of the United States.",
4 "Microsoft is headquartered in Redmond, Washington.",
5 "The Eiffel Tower is in Paris, France."
6]
7
8results = ner_pipeline(texts)
9
10for i, text_results in enumerate(results):
11 print(f"\nText {i+1}: {texts[i]}")
12 for entity in text_results:
13 print(f" - {entity['word']}: {entity['entity_group']}")