Mt5 Small Amharic Text Summaization

🚀 mt5-small-amharic-text-summaization

This model is a fine - tuned version of google/mt5-small, which can be used for text summarization, providing a practical solution for quickly extracting key information from text.

🚀 Quick Start

This model is a fine - tuned version of google/mt5-small on the None dataset. It achieves the following results on the evaluation set:

• eval_loss: 1.9166
• eval_rouge1: 17.2202
• eval_rouge2: 9.6515
• eval_rougeL: 16.8757
• eval_rougeLsum: 16.8802
• eval_runtime: 189.6659
• eval_samples_per_second: 14.178
• eval_steps_per_second: 1.777
• epoch: 10.0
• step: 13450

✨ Features

• Fine - tuned Model: Based on the pre - trained google/mt5-small, it is fine - tuned for specific tasks.
• Multiple Metrics Evaluation: Evaluated using multiple ROUGE metrics, providing comprehensive performance assessment.

🔧 Technical Details

Model description

More information needed

Intended uses & limitations

Trained on news articles and headlines.

Training procedure

Training hyperparameters

The following hyperparameters were used during training:

• learning_rate: 5.6e - 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
• num_epochs: 12

Framework versions

• Transformers 4.35.2
• Pytorch 2.1.0+cu121
• Datasets 2.16.1
• Tokenizers 0.15.0

📄 License

This model is released under the Apache 2.0 license.