Update README.md
Browse files
README.md
CHANGED
|
@@ -13,7 +13,11 @@ configs:
|
|
| 13 |
|
| 14 |
# akhooli/ar-mmarco-sample
|
| 15 |
This repo has samples from the Arabic (machine translation) version of the mMARCO dataset, together with mined rankings (English, but should apply
|
| 16 |
-
as translations are aligned).
|
| 17 |
The purpose is to train (using free compute, so not fully trained) an Arabic ColBERT V2 model.
|
| 18 |
-
The original dataset has a little over 800K queries. I filtered out ones with English words, leaving around 700K, then sampled 250K along with their
|
| 19 |
-
ranking examples (for 250K, the file size is a little over 8GB).
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 13 |
|
| 14 |
# akhooli/ar-mmarco-sample
|
| 15 |
This repo has samples from the Arabic (machine translation) version of the mMARCO dataset, together with mined rankings (English, but should apply
|
| 16 |
+
as translations are aligned across languages).
|
| 17 |
The purpose is to train (using free compute, so not fully trained) an Arabic ColBERT V2 model.
|
| 18 |
+
The original dataset has a little over 800K queries (training set). I filtered out ones with English words, leaving around 700K, then sampled 250K along with their
|
| 19 |
+
ranking examples (for 250K, the examples file size is a little over 8GB).
|
| 20 |
+
|
| 21 |
+
The source of this curated data is [unicamp-dl/mmarco](https://huggingface.co/datasets/unicamp-dl/mmarco) and the full examples json file (27 GB) is
|
| 22 |
+
linked in the [ColBERT V2 repo](https://github.com/stanford-futuredata/ColBERT?tab=readme-ov-file#advanced-training-colbertv2-style).
|
| 23 |
+
|