Roberta Base Topic Classification Nyt News

🚀 roberta-base_topic_classification_nyt_news

This model is a fine - tuned version of [roberta - base](https://huggingface.co/roberta - base) on the NYT News dataset, which contains 256,000 news titles from articles published from 2000 to the present (https://www.kaggle.com/datasets/aryansingh0909/nyt - articles - 21m - 2000 - present). It can effectively classify news topics, providing high - accuracy classification results for news data.

🚀 Quick Start

Using the Model with HuggingFace

from transformers import AutoTokenizer, AutoModelForSequenceClassification
from transformers import pipeline

tokenizer = AutoTokenizer.from_pretrained("dstefa/roberta-base_topic_classification_nyt_news")
model = AutoModelForSequenceClassification.from_pretrained("dstefa/roberta-base_topic_classification_nyt_news")
pipe = pipeline("text-classification", model=model, tokenizer=tokenizer)

text = "Kederis proclaims innocence Olympic champion Kostas Kederis today left hospital ahead of his date with IOC inquisitors claiming his innocence and vowing."
pipe(text)

[{'label': 'Sports', 'score': 0.9989326596260071}]

✨ Features

• High - accuracy Classification: Achieves high scores in accuracy, F1, precision, and recall on the test set.
• Multiple Topic Coverage: Can classify news into multiple topics such as sports, arts, business, health, etc.

📦 Installation

This section does not have specific installation commands in the original text, so it is skipped.

💻 Usage Examples

Basic Usage

from transformers import AutoTokenizer, AutoModelForSequenceClassification
from transformers import pipeline

tokenizer = AutoTokenizer.from_pretrained("dstefa/roberta-base_topic_classification_nyt_news")
model = AutoModelForSequenceClassification.from_pretrained("dstefa/roberta-base_topic_classification_nyt_news")
pipe = pipeline("text-classification", model=model, tokenizer=tokenizer)

text = "Kederis proclaims innocence Olympic champion Kostas Kederis today left hospital ahead of his date with IOC inquisitors claiming his innocence and vowing."
result = pipe(text)
print(result)

Advanced Usage

# Classify multiple texts at once
texts = [
    "Olympic champion Kostas Kederis today left hospital ahead of his date with IOC inquisitors claiming his innocence and vowing.",
    "Although many individuals are doing fever checks to screen for Covid - 19, many Covid - 19 patients never have a fever.",
    "Twelve myths about Russia's War in Ukraine exposed"
]
results = pipe(texts)
for res in results:
    print(res)

📚 Documentation

Model Information

Training Data Classification

Training Procedure

Training Hyperparameters

The following hyperparameters were used during training:

• learning_rate: 5e - 05
• train_batch_size: 8
• eval_batch_size: 8
• seed: 42
• optimizer: Adam with betas=(0.9,0.999) and epsilon = 1e - 08
• lr_scheduler_type: linear
• lr_scheduler_warmup_steps: 500
• num_epochs: 5

Training Results

Model Performance

Framework Versions

• Transformers 4.32.1
• Pytorch 2.1.0+cu121
• Datasets 2.12.0
• Tokenizers 0.13.2

🔧 Technical Details

The model is fine - tuned based on the roberta - base model. During the training process, appropriate hyperparameters are selected, and the Adam optimizer is used for optimization. The linear learning rate scheduler is used to adjust the learning rate, which helps the model converge better and achieve high - accuracy classification results on the news dataset.

📄 License

This project is licensed under the MIT license.