phobert-ner-address

This model is a fine-tuned version of vinai/phobert-base on an vietnamese address datasets. It achieves the following results on the evaluation set:

• Loss: 0.1725
• Precision: 0.9363
• Recall: 0.9453
• F1: 0.9407

Model description

More information needed

Intended uses & limitations

More information needed

Training and evaluation data

More information needed

Training procedure

Training hyperparameters

The following hyperparameters were used during training:

• learning_rate: 5e-05
• train_batch_size: 16
• eval_batch_size: 16
• seed: 42
• optimizer: Use OptimizerNames.ADAMW_TORCH with betas=(0.9,0.999) and epsilon=1e-08 and optimizer_args=No additional optimizer arguments
• lr_scheduler_type: linear
• num_epochs: 3

Usage

from transformers import AutoModelForTokenClassification, AutoTokenizer
import torch

model = AutoModelForTokenClassification.from_pretrained('kiendt/phobert-ner-address')
tokenizer = AutoTokenizer.from_pretrained('kiendt/phobert-ner-address')

label_list = ['B_PRO', 'B_CITY', 'NUMBER_TYPE', 'B_DIST', 'TO_TYPE', 'B_STREET', 'I_PRO', 'I_DIST', 'PRO_TYPE', 'OTHER', 'I_STREET', 'B_WARD', 'STREET_TYPE', 'I_CITY', 'CITY_TYPE', 'O', 'NUMBER', 'WARD_TYPE', 'I_WARD', 'DIST_TYPE', 'TO']

id2label = {i: label for i, label in enumerate(label_list)}

def predict_entities(text):
    inputs = tokenizer(text, return_tensors="pt", padding=True, truncation=True, max_length=128)
    with torch.no_grad():
        outputs = model(**inputs)
        logits = outputs.logits

    predictions = torch.argmax(logits, dim=-1)
    predicted_labels = [id2label[label.item()] for label in predictions[0]]

    tokens = tokenizer.convert_ids_to_tokens(inputs['input_ids'][0])

    print("\nTokens and Predicted Labels:")
    print(f"{'Token':<15} {'Predicted Label'}")
    print("-" * 40)
    for token, label in zip(tokens, predicted_labels):
        print(f"{token:<15} {label}")

predict_entities("Km 1 đường Nguyễn Văn Linh, PHƯỜNG PHÚC ĐỒNG, QUẬN LONG BIÊN, THÀNH PHỐ HÀ NỘI")

Training results

Framework versions

• Transformers 4.55.2
• Pytorch 2.6.0+cu124
• Datasets 4.0.0
• Tokenizers 0.21.4