bert-base-uncased-agnews4-v01

Model Card for bert-based-uncased-agnews4

Model Description

AG News dataset and classifies texts into 4 categories:

• 0: [World News]
• 1: [Sports]
• 2: [Business]
• 3: [Sci/Tech]

The maximum token length used during training is set to 128.

Usage

from transformers import AutoTokenizer, AutoModelForSequenceClassification import torch

model_name = "Lucky0626/bert-based-uncased-agnews4-v01" tokenizer = AutoTokenizer.from_pretrained(model_name) model = AutoModelForSequenceClassification.from_pretrained(model_name)

text = "NASA launches a new satellite into space" inputs = tokenizer(text, return_tensors="pt", max_length=128, truncation=True)

with torch.no_grad(): outputs = model(**inputs) predictions = torch.argmax(outputs.logits, dim=-1)

print("Predicted label:", predictions.item())

text

Evaluation

• Dataset: AG News test set
• Metrics: Accuracy, F1-score (weighted)
• Example Results:
  • Accuracy: 0.85
  • F1-score (weighted): 0.85

Use Cases

• Automated news article classification
• Preprocessing for news recommendation systems
• Topic-based text analysis

Limitations

• Optimized for English text classification based on AG News dataset
• Performance may degrade on texts in other languages
• Does not include training on the latest news events

License

• The base model ([translate:bert-base-uncased]) is licensed under Apache 2.0
• The AG News dataset is an open dataset