Views
No views yet
1
2datasets:
3 - custom
4library_name: onmt
5model-index:
6 - name: Malayalam to Hindi Translation
7 results:
8 - task:
9 name: Translation
10 type: translation
11 dataset:
12 name: Custom Hindi- Malayalam Parallel Corpus
13 type: translation
14 metrics:
15 - name: BLEU
16 type: bleu
17 value: 35.5
18 - name : COMET
19 - type:comet
20 - value: 0.582
21
22
23🇮🇳 Malayalam to Hindi Translation Model (OpenNMT)
24
25This is a Neural Machine Translation (NMT) model trained to translate Malayalam (ml) to Hindi (hi) using the OpenNMT framework. It was trained on a custom curated low-resource parallel corpus.
26
27 Model Architecture
28
29- Framework: **OpenNMT (PyTorch)**
30- Architecture: **Transformer**
31- Type: **Sequence-to-sequence**
32- Layers: 6 encoder / 6 decoder
33- Embedding size: 512
34- FFN size: 2048
35- Attention heads: 8
36- Positional encoding: sinusoidal
37- Tokenizer: SentencePiece (trained jointly on hi-ml)
38- Vocabulary size: 32,000 (joint BPE)
39
40
41Evaluation
42
43The model was evaluated on a manually annotated Hindi-Malayalam test set consisting of 10,000 sentence pairs.
44
45| Metric | Score |
46|--------|-------|
47| BLEU | 35.5 |
48| COMET | 0.582 |
49
50Usage
51
52 IN CLI
53
54
55onmt_translate \
56 -model model.tm_best_checkpoint.pt \
57 -src input.txt \
58 -output output.txt \
59 -replace_unk \
60 -verbose \
61 -gpu -1 \
62 -min_length 1
63
64Dataset
65
66This model was trained on a custom dataset compiled from:
67
68* (https://github.com/AI4Bharat/IndicTrans)
69* Manually aligned Malayalam-Hindi sentences from news and educational data