Initial Commit.

README.md

@@ -1,3 +1,64 @@
----
-license: mit
----
+---
+license: mit
+metrics:
+  - bleu
+base_model:
+  - facebook/nllb-200-distilled-600M
+pipeline_tag: translation
+tags:
+  - Efik
+  - English
+  - Translation
+  - Africa
+---
+# Efik ↔ English Translation Model
+
+This model provides **machine translation between English (eng_Latn) and Efik (efik_Latn)**. It was fine-tuned on **18k+ parallel sentences** using the **NLLB architecture** and can be used for both direct translation and integration into multilingual NLP pipelines.  
+
+### Uses
+
+- Translate text between English and Efik  
+- Assist in educational or localization projects involving Efik  
+- Support research in low-resource language NLP  
+
+### Limitations
+
+- Performance may decrease for **long, complex, or domain-specific text**  
+- Model trained on general domain data, may need additional fine-tuning for specialized applications  
+
+### How to Get Started
+
+```python
+from transformers import AutoModelForSeq2SeqLM, AutoTokenizer
+
+tokenizer = AutoTokenizer.from_pretrained("offiongbassey/efik-mt")
+model = AutoModelForSeq2SeqLM.from_pretrained("offiongbassey/efik-mt")
+
+# English → Efik
+text = "Jesus is God."
+inputs = tokenizer(f"eng_Latn {text}", return_tensors="pt")
+outputs = model.generate(**inputs, max_length=128)
+print(tokenizer.decode(outputs[0], skip_special_tokens=True))
+
+# Efik → English
+text = "Ka ke tietie."
+inputs = tokenizer(f"ibo_Latn {text}", return_tensors="pt")
+outputs = model.generate(**inputs, max_length=128)
+print(tokenizer.decode(outputs[0], skip_special_tokens=True))
+```
+
+### Training Details
+
+- Architecture: NLLB
+- Epochs trained: 3
+- BLEU Scores:
+  - EN → EF: 27.69
+  - EF → EN: 30.95
+
+### Citation
+@misc{offiongbassey2025efikmt,
+  title={Efik ↔ English Translation Model},
+  author={Offiong Bassey},
+  year={2025},
+  url={https://huggingface.co/offiongbassey/efik-mt}
+}