Version 1.0 – Initial Release
This release contains the source code for the research project Amharic Intention Detection Using AfroXLM-RoBERTa.
Contents
Data preprocessing scripts
Model training code
Model evaluation scripts
Inference/prediction code
Configuration files
Documentation (README.md)
Features
Fine-tuning of the AfroXLM-RoBERTa model for Amharic intention detection.
Support for preprocessing and tokenization of Amharic text.
Training, validation, and testing pipelines.
Performance evaluation using Accuracy, Precision, Recall, and F1-score.
Reproducible implementation for research purposes.
Notes
This release accompanies the research on Amharic intention detection. It is intended to support reproducibility and facilitate future research in Amharic Natural Language Processing (NLP). Future releases may include additional improvements, bug fixes, and enhanced documentation.
Version: v1.0