Downloads · 30 days
8
11% of all-time downloads
Ijaj7/Xetu-assamese-to-english-translator
Xetu-assamese-to-english-translator is a machine learning model from Ijaj7. Use it for the machine learning task on the model card, and read the license before you ship it in a product. The card lists the license as mit.
Model Name: xetu-assamese-to-english-translator A high-quality Assamese → English neural machine translation model for the Xetu Chat App.
Downloads · 30 days
8
11% of all-time downloads
All-time downloads
74
Public
Parameters
77.1M
310 MB on disk
Likes
0
Public
Click a slice to open those files.
.safetensors308 MB · 99%
From the Hugging Face model README
Model Name: xetu-assamese-to-english-translator
A high-quality Assamese → English neural machine translation model for the Xetu Chat App.
This repository contains a fine-tuned OPUS-MT multilingual model (opus-mt-mul-en) optimized for Assamese → English translation.
The model is trained for real-time bilingual messaging inside the Xetu Chat App, supporting auto-translate and seamless bilingual communication.
Base Model: Helsinki-NLP/opus-mt-mul-en (MarianMT)
Task: Machine Translation (Assamese → English)
Architecture: Seq2Seq Transformer (MarianMT)
Use Case:
This fine-tuned version shows strong translation quality, especially for conversational and semi-formal Assamese text.
| Split | Size |
|---|---|
| Train | 50,000 pairs |
| Validation | 500 pairs |
| Test | 2,000 pairs |
{
"source_text": "Assamese sentence",
"target_text": "English sentence"
}
This model does not require language prefix tokens (e.g., >>eng<<).
You can directly pass Assamese input.
AutoTokenizer)max_source_length = 256max_target_length = 256-100 for loss maskingExample preprocessing:
tokenizer("অসমীয়া বাক্য")
| Parameter | Value |
|---|---|
| Epochs | 5 |
| Learning rate | 3e-5 |
| Warmup steps | 800 |
| Weight decay | default |
| Batch size/device | 4 |
| Gradient accumulation | 4 |
| Effective batch size | 16 |
| Precision | fp16 = True |
| Optimizer | AdamW |
50,000 / 16 = 3,12515,625Training was stable with no gradient explosions.
| Epoch | Eval Loss | BLEU | chrF |
|---|---|---|---|
| 1 | 1.5293 | 27.75 | 54.47 |
| 2 | 1.4229 | 29.78 | 56.69 |
| 3 | 1.3669 | 31.55 | 58.27 |
| 4 | 1.3440 | 32.29 | 59.13 |
| 5 | 1.3405 | 32.71 | 59.43 |
Consistent improvements every epoch — no overfitting observed.
| Metric | Value |
|---|---|
| Test BLEU | 34.624 |
| Test chrF | 60.1136 |
| Test Loss | 1.3238 |
Test BLEU is higher than validation BLEU, indicating good generalization.
from transformers import AutoTokenizer, AutoModelForSeq2SeqLM
import torch
model_name = "Ijaj7/xetu-assamese-to-english-translator"
tokenizer = AutoTokenizer.from_pretrained(model_name)
model = AutoModelForSeq2SeqLM.from_pretrained(model_name)
text = "মই আজ বিদ্যালয়লৈ গ’লোঁ।"
inputs = tokenizer(text, return_tensors="pt")
with torch.no_grad():
outputs = model.generate(**inputs)
print(tokenizer.decode(outputs[0], skip_special_tokens=True))
curl -X POST \
-H "Authorization: Bearer YOUR_HF_TOKEN" \
-H "Content-Type: application/json" \
-d '{"inputs": "মই কলেজলৈ গৈছো।"}' \
https://api-inference.huggingface.co/models/Ijaj7/xetu-assamese-to-English-translator
Optimized for:
This model pairs with your English → Assamese model for full bilingual support.
| Assamese | English |
|---|---|
| তোমাৰ লগত কথা পাতি ভাল লাগে | It feels good to talk to you |
| মই আজি ঘৰত থাকিম। | I'll be at home today. |
| মোৰ কেইটামান কাম আছে। | I have a few things to do. |
| তুমি অলপ অপেক্ষা কৰিব পাৰিবা নে? | Could you wait a while? |
This model is released under the MIT License.
@misc{Ijaj7_Xetu_as_en_2025,
title = {Xetu: Assamese-to-English Translation Model},
author = {Ijaj Ahmed},
year = {2025},
howpublished = {\url{https://huggingface.co/Ijaj7/xetu-assamese-to-english-translator}},
note = {Accessed: 2025-11-28}
}