Create README.md
Browse files
README.md
ADDED
|
@@ -0,0 +1,31 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
---
|
| 2 |
+
language: african-languages
|
| 3 |
+
tags:
|
| 4 |
+
- african-languages
|
| 5 |
+
- machine-translation
|
| 6 |
+
- text
|
| 7 |
+
license: apache-2.0
|
| 8 |
+
model-index:
|
| 9 |
+
- name: Maakhane Benchmark ModelsFon XLSR Wav2Vec2 Large 53
|
| 10 |
+
results:
|
| 11 |
+
- task:
|
| 12 |
+
name: Machine Translation
|
| 13 |
+
type: machine-translation
|
| 14 |
+
dataset:
|
| 15 |
+
name: masakhane benchmarks
|
| 16 |
+
args: african-languages
|
| 17 |
+
|
| 18 |
+
---
|
| 19 |
+
# Masakhane Benchmark Models
|
| 20 |
+
|
| 21 |
+
I created a demo for very easy interaction with the [benchmark models on Masakhane](https://github.com/masakhane-io/masakhane-mt/tree/master/benchmarks) which were trained with JoeyNMT
|
| 22 |
+
|
| 23 |
+
To access the space click [here](https://huggingface.co/spaces/chrisjay/masakhane-benchmarks)
|
| 24 |
+
|
| 25 |
+
To include your language, all you need to do is:
|
| 26 |
+
1. Create a folder in the format *src-tgt/main* for your language pair, if it does not exist.
|
| 27 |
+
2. Inside the *main* folder put the following files:
|
| 28 |
+
a. model checkpoint. Rename it to `best.ckpt`.
|
| 29 |
+
b. `config.yaml` file. This is the JoeyNMT config file which loads the model an pre-processing parameters.
|
| 30 |
+
c. `src_vocab.txt` file.
|
| 31 |
+
d. `trg_vocab.txt` file.
|