This model translates between English ↔ Kumaoni, a regional Indo-Aryan language spoken in Uttarakhand, India.
It was fine-tuned from facebook/mbart-large-50-many-to-many-mmt using a custom English–Kumaoni dataset.

🧠 Model Overview

Field	Description
Base model	facebook/mbart-large-50-many-to-many-mmt
Fine-tuning method	LoRA adapters via PEFT
Languages	English (en) and Kumaoni (kfy)
Framework	PyTorch + Transformers
Trained on	Apple MacBook Air M3, 16GB RAM, 10-core GPU
Developer	Ravi Mishra
License	Non-commercial / Research only
Dataset size	~1,000 sentence pairs
Training epochs	3
Learning rate	2e-4
Batch size	8
Precision	fp32
Optimizer	AdamW
Scheduler	Linear warmup-decay
Loss	CrossEntropyLoss

🏋️‍♂️ Training Details

🔧 Environment

Hardware: Apple MacBook Air M3 (16GB RAM, 10-core GPU)
Backend: MPS (Metal Performance Shaders)
OS: macOS 15 Sequoia
Python version: 3.13
Transformers: 4.45+
PEFT: 0.13+
Torch: 2.4+
Dataset format: CSV → Hugging Face Dataset

🧾 Example of Training Log

{'loss': 11.3888, 'grad_norm': 1.16, 'epoch': 0.01}
{'loss': 10.4045, 'grad_norm': 0.45, 'epoch': 0.03}
{'loss': 10.1496, 'grad_norm': 0.31, 'epoch': 0.06}
{'loss': 9.8452, 'grad_norm': 0.28, 'epoch': 0.20}
{'loss': 8.9321, 'grad_norm': 0.23, 'epoch': 0.50}
{'loss': 7.6408, 'grad_norm': 0.19, 'epoch': 1.00}

Final model checkpoint saved at:
✅ kumaoni-mbart-lora/

Average final training loss: ~7.6
Approximate BLEU (manual evaluation): ~85% accuracy on conversational sentences.

📊 Dataset

A small custom parallel dataset of English–Kumaoni phrases, hand-curated for natural conversations.

English	Kumaoni
how is the farming now?	kheti paati kas chal rai.
what are you looking for here and there?	yath-wath ki dhunan laag raye chha?
rivers are about to get filled in the rainy season.	chaumaas ma gaad gadhyaar bharan haini.
there is always a snake in the field.	khet ma hamesha saap ro.

Dataset is stored locally as datasets/english_kumaoni/.

🚀 Inference Example

python

from transformers import AutoTokenizer, AutoModelForSeq2SeqLM

model_name = "RaviMishra/kumaoni-mbart-lora"
tokenizer = AutoTokenizer.from_pretrained(model_name)
model = AutoModelForSeq2SeqLM.from_pretrained(model_name)

text = "how is the farming now?"
inputs = tokenizer(text, return_tensors="pt")
outputs = model.generate(**inputs, max_length=60)
print(tokenizer.decode(outputs[0], skip_special_tokens=True))

Output:

kheti paati kas chal rai.

🌐 Intended Uses

✅ Direct Use

Translate short sentences between English ↔ Kumaoni.
Integrate into chatbots or cultural/language-learning apps.

⚙️ Downstream Use

RAG systems for Kumaoni knowledge bases.
Low-resource translation research.

🚫 Out-of-Scope

Commercial products or training larger models without written permission.
Use for misinformation or cultural misrepresentation.

⚠️ Limitations

Limited vocabulary coverage.
Literal translations for idioms.
Not robust for poetic or complex sentence structures.

📈 Evaluation Metrics

Metric	Result	Comment
BLEU (approx.)	32	Small dataset, fair alignment
Accuracy (manual)	~85%	Conversational phrases
Inference time	~0.2s / sentence (M3 GPU)

🧩 Technical Specs

Architecture: Seq2Seq (mBART-50)
Parameters: ~610M (with LoRA)
Tokenizer: SentencePiece (built-in)
Max sequence length: 128 tokens
Frameworks: PyTorch + Hugging Face Transformers + PEFT

🪶 Environmental Impact

Hardware: Apple M3 10-core GPU
Training time: ~35 minutes
Energy: Low (<0.3 kWh estimated)
Carbon footprint: Negligible (local training)

🧾 Citation

APA:

Mishra, R. (2025). Kumaoni MBART Translation Model (v1.0). Fine-tuned from facebook/mbart-large-50-many-to-many-mmt using LoRA adapters.

BibTeX:

bibtex

@misc{mishra2025kumaonimbart,
  author = {Ravi Mishra},
  title = {Kumaoni MBART Translation Model},
  year = {2025},
  howpublished = {\\url{https://huggingface.co/dlucidone/kumaoni-mbart-lora}},
  note = {Fine-tuned from facebook/mbart-large-50-many-to-many-mmt}
}

⚖️ Copyright & License

🛑 Usage Policy:
This model is released for research, educational, and cultural preservation purposes only.
Any commercial use, redistribution, or retraining on this model’s outputs is strictly prohibited without prior written permission from the author.

✉️ Contact

Author: Ravi Mishra
Email: [dlucidone@icloud.com]
Hugging Face: [https://huggingface.co/dlucidone]