File size: 843 Bytes
b371be6
9763521
 
 
b371be6
a8ab035
 
204da07
b371be6
8c8d451
204da07
 
 
 
 
 
8c8d451
204da07
8c8d451
 
 
 
9763521
204da07
8c8d451
 
 
9763521
 
204da07
8c8d451
 
 
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
---

language:
- en
- ml
license: mit
tags:
- fastai
- translation
---


# Fine Tune En-ML translation

* source group: English
* target group: Chinese 

This is a ML translation model created for fun for english to Malayalam translation fine-tuned on KDE-Dataset.

[Tweet](https://twitter.com/kurianbenoy2/status/1503082136009465857?s=20&t=7Hn-KUqHZRY6VJ16-i1qdA)

# Model card

## Model description

Used a fine tuned model on top of MarianMT models created by Helsinki-NLP group. The [training code is described here](https://kurianbenoy.com/ml-blog/fastai/huggingface/translation/fine%20tuning/malayalam/2022/03/12/_03_13_huggingace_translation_models.html).

## Intended uses & limitations

Intended to use just for fun, and for sake of learning

Limitations: Returns really bad predictions occasionally