Update README.md
Browse files
README.md
CHANGED
|
@@ -1 +1,31 @@
|
|
| 1 |
-
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
---
|
| 2 |
+
license: mit
|
| 3 |
+
datasets:
|
| 4 |
+
- unicamp-dl/mmarco
|
| 5 |
+
language:
|
| 6 |
+
- fr
|
| 7 |
+
tags:
|
| 8 |
+
- colbert
|
| 9 |
+
- ColBERT
|
| 10 |
+
---
|
| 11 |
+
|
| 12 |
+
## Training
|
| 13 |
+
|
| 14 |
+
#### Details
|
| 15 |
+
|
| 16 |
+
The model is initialized from the [ColBERTv1.0-bert-based-spanish-mmarcoES](antoinelouis/colbertv1-camembert-base-mmarcoFR) checkpoint and trained using the ColBERTv2 style of training.
|
| 17 |
+
It was trained on 1 Tesla L4 GPU with 24GBs of memory with 20k warmup steps warmup using a batch size of 64 and the AdamW optimizer with a constant learning rate of 1e-05.
|
| 18 |
+
Total training time was around 80 hours.
|
| 19 |
+
|
| 20 |
+
#### Data
|
| 21 |
+
|
| 22 |
+
The model is fine-tuned on the French version of the [mMARCO](https://huggingface.co/datasets/unicamp-dl/mmarco) dataset, a multi-lingual machine-translated version of the MS MARCO dataset.
|
| 23 |
+
|
| 24 |
+
|
| 25 |
+
## Evaluation
|
| 26 |
+
|
| 27 |
+
The model is evaluated on the smaller development set of mMARCO-fr, which consists of 6,980 queries for a corpus of 8.8M candidate passages. We report the mean reciprocal rank (MRR) and recall at various cut-offs (R@k).
|
| 28 |
+
|
| 29 |
+
| model | Vocab. | #Param. | Size | MRR@10 | R@50 | R@200 |
|
| 30 |
+
|:------------------------------------------------------------------------------------------------------------------------|:--------|--------:|------:|---------:|-------:|--------:|
|
| 31 |
+
| **ColBERTv2.0-mmarcoFR** | french | 110M | 440MB | **32.76** | **76.93** | **81.74** |
|