CAP Coded US Congressional Bills

🚀 CAP_coded_US_Congressional_bills

This model predicts the issue category of US Congressional bills, trained on a large dataset of bills from 1950 - 2015.

🚀 Quick Start

This model is designed to predict the issue category of US Congressional bills. It has been trained on approximately 250k US Congressional bills spanning from 1950 to 2015.

✨ Features

• Predictive Power: Accurately predicts the issue category of US Congressional bills.
• Large - scale Training: Trained on a substantial dataset of ~250k bills from 1950 - 2015.
• Case - sensitive: The model is cased, meaning it takes into account letter case during prediction.

📦 Installation

No installation steps were provided in the original document, so this section is skipped.

💻 Usage Examples

Basic Usage

The model can be used to predict the issue category of bills. Here are some example inputs:

• Example 1: "A bill to prohibt discrimination in employment because of race, color, religion, national origin, or ancestry"
• Example 2: "A bill to require the promulgation of regulations to improve aviation safety in adverse weather conditions, and for other purposes."

📚 Documentation

Model Details

• Model Type: A model for predicting issue categories of US Congressional bills.
• Training Data: Approximately 250k US Congressional bills from 1950 - 2015.
• Issue Coding Scheme: Follows the Comparative Agenda Project: https://www.comparativeagendas.net/pages/master - codebook

Training Metrics

• Train Loss: 0.1318
• Train Sparse Categorical Accuracy: 0.9268
• Validation Loss: 0.2439
• Validation Sparse Categorical Accuracy: 0.9161

Training Hyperparameters

• optimizer: {'name': 'Adam', 'learning_rate': 5e - 05, 'decay': 0.0, 'beta_1': 0.9, 'beta_2': 0.999, 'epsilon': 1e - 07, 'amsgrad': False}
• training_precision: float32

Framework Versions

• Transformers 4.19.3
• TensorFlow 2.8.2
• Tokenizers 0.12.1

🔧 Technical Details

The model follows the issue coding scheme of the Comparative Agenda Project. It has been trained with specific hyperparameters and shows good performance on both training and validation datasets.

📄 License

No license information was provided in the original document, so this section is skipped.