Upload README.md
Browse files
README.md
ADDED
@@ -0,0 +1,32 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
1 |
+
Hugging Face's logo
|
2 |
+
---
|
3 |
+
language:
|
4 |
+
- yo
|
5 |
+
- en
|
6 |
+
datasets:
|
7 |
+
- JW300 + [Menyo-20k](https://huggingface.co/datasets/menyo20k_mt)
|
8 |
+
---
|
9 |
+
# m2m100_418M-eng-yor-mt
|
10 |
+
## Model description
|
11 |
+
**m2m100_418M-eng-yor-mt** is a **machine translation** model from English language to Yorùbá language based on a fine-tuned facebook/m2m100_418M model. It establishes a **strong baseline** for automatically translating texts from English to Yorùbá.
|
12 |
+
|
13 |
+
Specifically, this model is a *facebook/m2m100_418M* model that was fine-tuned on JW300 Yorùbá corpus and [Menyo-20k](https://huggingface.co/datasets/menyo20k_mt).
|
14 |
+
|
15 |
+
#### Limitations and bias
|
16 |
+
This model is limited by its training dataset. This may not generalize well for all use cases in different domains.
|
17 |
+
## Training data
|
18 |
+
This model was fine-tuned on on JW300 corpus and [Menyo-20k](https://huggingface.co/datasets/menyo20k_mt) dataset
|
19 |
+
|
20 |
+
## Training procedure
|
21 |
+
This model was trained on NVIDIA V100 GPU
|
22 |
+
|
23 |
+
## Eval results on Test set (BLEU score)
|
24 |
+
Fine-tuning mbarr50-large achieves ** BLEU** on [Menyo-20k test set](https://arxiv.org/abs/2103.08647) while mt5-base achieves 9.82
|
25 |
+
|
26 |
+
### BibTeX entry and citation info
|
27 |
+
By David Adelani
|
28 |
+
```
|
29 |
+
|
30 |
+
```
|
31 |
+
|
32 |
+
|