Downloads · 30 days
8
5% of all-time downloads
opus-mt-sw-en-finetuned
marlonbino/opus-mt-sw-en-finetuned
opus-mt-sw-en-finetuned is a translation model from marlonbino. Use it when you need text moved from one language to another. The card lists the license as apache-2.0.
This model performs translation trained using MLflow and deployed on Hugging Face.
All-time downloads
171
Public
Parameters
77.1M
1.5 GB on disk
Likes
0
Public
Repo makeup
Click a slice to open those files.
.safetensors308 MB · 99%
At a glance
- Task
- Translation
- License
- apache-2.0
- Model type
- marian
- Access
- Public
- Created
- Oct 8, 2025
- Updated
- Nov 6, 2025
- SHA
- e804b4f3
Datasets
Model card
From the Hugging Face model README
opus-mt-sw-en-finetuned
This model performs translation trained using MLflow and deployed on Hugging Face.
Model Details
- Model Name: opus-mt-sw-en-finetuned
- Version: 7
- Task: Translation
- Languages: sw,en
- Framework: pytorch
- License: apache-2.0
Intended Uses & Limitations
Intended Uses
- Translation tasks
- Research and development
- Child helpline services support
Limitations
- Performance may vary on out-of-distribution data
- Should be evaluated on your specific use case before production deployment
- Designed for child helpline contexts, may need adaptation for other domains
Training Data
- Dataset: synthetic.jsonl
- Size: Not specified
- Languages: sw,en
Training Configuration
| Parameter | Value |
|---|---|
| Dataset Config Auto Segment Long Sequences | True |
| Dataset Config Max Length Ratio | 2.5 |
| Dataset Config Max Samples | None |
| Dataset Config Min Length | 3 |
| Dataset Config Primary Dataset | custom |
| Dataset Config Segment Max Tokens | 400 |
| Dataset Config Segment Overlap Tokens | 100 |
| Dataset Config Validation Split | 0.1 |
| Evaluation Config Inference Test Frequency | every_eval |
| Evaluation Config Run Inference Validation | True |
| Evaluation Config Test Size | 500 |
| Language Name | Swahili |
| Language Pair | sw-en |
| Max Length | 512 |
| Model Name | Helsinki-NLP/opus-mt-mul-en |
| Total Parameters | 77518848 |
| Trainable Parameters | 76994560 |
| Training Config Batch Size | 16 |
| Training Config Dataloader Num Workers | 4 |
| Training Config Early Stopping Patience | 3 |
| Training Config Early Stopping Threshold | 0.001 |
| Training Config Eval Steps | 500 |
| Training Config Eval Strategy | steps |
| Training Config Generation Length Penalty | 0.85 |
| Training Config Generation Max Length | 128 |
| Training Config Generation No Repeat Ngram Size | 3 |
| Training Config Generation Num Beams | 4 |
| Training Config Generation Repetition Penalty | 1.0 |
| Training Config Generation Top K | 50 |
| Training Config Gradient Accumulation Steps | 2 |
| Training Config Learning Rate | 2e-05 |
| Training Config Logging Steps | 50 |
| Training Config Lr Scheduler | cosine_with_restarts |
| Training Config Max Length | 512 |
| Training Config Mixed Precision | fp16 |
| Training Config Num Epochs | 10 |
| Training Config Pin Memory | True |
| Training Config Save Strategy | epoch |
| Training Config Warmup Steps | 1000 |
| Training Config Weight Decay | 0.01 |
| Vocab Size | 64172 |
Performance Metrics
Evaluation Results
| Metric | Value |
|---|---|
| Baseline Bleu | 0.0145 |
| Baseline Chrf | 16.4877 |
| Baseline Hallucination Rate | 0.0050 |
| Baseline Keyword Preservation | 0.0983 |
| Baseline Urgency Preservation | 0.3019 |
| Bleu Improvement | 0.4410 |
| Bleu Improvement Percent | 3051.3105 |
| Chrf Improvement | 42.8156 |
| Chrf Improvement Percent | 259.6828 |
| Epoch | 10.0000 |
| Eval Bertscore F1 | 0.9432 |
| Eval Bleu | 0.4555 |
| Eval Chrf | 59.3032 |
| Eval Loss | 0.1448 |
| Eval Meteor | 0.5747 |
| Eval Runtime | 106.5231 |
| Eval Samples Per Second | 11.0210 |
| Eval Steps Per Second | 0.6950 |
| Final Epoch | 10.0000 |
| Final Eval Bertscore F1 | 0.9432 |
| Final Eval Bleu | 0.4555 |
| Final Eval Chrf | 59.3032 |
| Final Eval Loss | 0.1448 |
| Final Eval Meteor | 0.5747 |
| Final Eval Runtime | 106.5231 |
| Final Eval Samples Per Second | 11.0210 |
| Final Eval Steps Per Second | 0.6950 |
| Grad Norm | 0.8218 |
| Inference Test/Abuse Reporting Success | 1.0000 |
| Inference Test/Code Switching Success | 1.0000 |
| Inference Test/Emergency Success | 1.0000 |
| Inference Test/Emotional Distress Success | 1.0000 |
| Inference Test/Empty Input Success | 1.0000 |
| Inference Test/Fragmented Trauma Success | 1.0000 |
| Inference Test/Help Request Success | 1.0000 |
| Inference Test/Incomplete Success | 1.0000 |
| Inference Test/Location Info Success | 1.0000 |
| Inference Test/Numbers Preservation Success | 0.0000 |
| Inference Test/Simple Greeting Success | 1.0000 |
| Inference Test/Whitespace Only Success | 1.0000 |
| Inference Validation Pass Rate | 0.9167 |
| Learning Rate | 0.0000 |
| Loss | 0.1418 |
| Total Flos | 11212302632878080.0000 |
| Total Samples | 11732.0000 |
| Train Loss | 0.3851 |
| Train Runtime | 1770.8962 |
| Train Samples | 10558.0000 |
| Train Samples Per Second | 59.6200 |
| Train Steps Per Second | 1.8630 |
| Validation Samples | 1174.0000 |
Usage
Installation
pip install transformers torch
Translation Example
from transformers import pipeline
translator = pipeline("translation", model="marlonbino/opus-mt-sw-en-finetuned")
result = translator("Your text here")
print(result[0]["translation_text"])
MLflow Tracking
- Experiment: translation-sw-en
- Run ID:
618fb8990ea248d18259b61c18b9017a - Training Date: 2025-10-31 21:38:14
- Tracking URI: http://192.168.10.6:5000
Training Metrics Visualization
View detailed training metrics and TensorBoard logs in the Training metrics tab.
Citation
@misc{opus_mt_sw_en_finetuned,
title={opus-mt-sw-en-finetuned},
author={OpenCHS Team},
year={2025},
publisher={Hugging Face},
url={https://huggingface.co/marlonbino/opus-mt-sw-en-finetuned}
}
Contact
Model card auto-generated from MLflow
- Task
- Translation
- Type
- marian
- License
- apache-2.0
- Languages
- sw, en
- Created
- Oct 8, 2025
- Updated
- Nov 6, 2025