loic-dagnas-sinequa
commited on
Commit
·
da17af4
1
Parent(s):
8561a46
Update README.md
Browse files
@basilevc
@skirres
I have specified that by Chinese we meant simplified chinese as requested by Ariane here.
I have also reorder the language by the alphabetical order of the language codes
Just note that zs is not recognized by huggingface language tags.
README.md
CHANGED
@@ -8,7 +8,7 @@ language:
|
|
8 |
- ja
|
9 |
- nl
|
10 |
- pt
|
11 |
-
-
|
12 |
---
|
13 |
|
14 |
# Model Card for `passage-ranker.strawberry`
|
@@ -22,15 +22,15 @@ Model name: `passage-ranker.strawberry`
|
|
22 |
|
23 |
The model was trained and tested in the following languages:
|
24 |
|
25 |
-
-
|
26 |
-
- Dutch
|
27 |
- English
|
|
|
28 |
- French
|
29 |
-
- German
|
30 |
- Italian
|
31 |
- Japanese
|
|
|
32 |
- Portuguese
|
33 |
-
-
|
34 |
|
35 |
Besides the aforementioned languages, basic support can be expected for additional 91 languages that were used during
|
36 |
the pretraining of the base model (see Appendix A of [XLM-R paper](https://arxiv.org/abs/1911.02116)).
|
@@ -111,10 +111,10 @@ We evaluated the model on the datasets of the [MIRACL benchmark](https://github.
|
|
111 |
multilingual capacities. Note that not all training languages are part of the benchmark, so we only report the metrics
|
112 |
for the existing languages.
|
113 |
|
114 |
-
| Language
|
115 |
-
|
116 |
-
|
|
117 |
-
|
|
118 |
-
|
|
119 |
-
| Japanese
|
120 |
-
|
|
|
|
8 |
- ja
|
9 |
- nl
|
10 |
- pt
|
11 |
+
- zs
|
12 |
---
|
13 |
|
14 |
# Model Card for `passage-ranker.strawberry`
|
|
|
22 |
|
23 |
The model was trained and tested in the following languages:
|
24 |
|
25 |
+
- German
|
|
|
26 |
- English
|
27 |
+
- Spanish
|
28 |
- French
|
|
|
29 |
- Italian
|
30 |
- Japanese
|
31 |
+
- Dutch
|
32 |
- Portuguese
|
33 |
+
- Chinese
|
34 |
|
35 |
Besides the aforementioned languages, basic support can be expected for additional 91 languages that were used during
|
36 |
the pretraining of the base model (see Appendix A of [XLM-R paper](https://arxiv.org/abs/1911.02116)).
|
|
|
111 |
multilingual capacities. Note that not all training languages are part of the benchmark, so we only report the metrics
|
112 |
for the existing languages.
|
113 |
|
114 |
+
| Language | NDCG@10 |
|
115 |
+
|:--------------------|--------:|
|
116 |
+
| German | 0.320 |
|
117 |
+
| Spanish | 0.418 |
|
118 |
+
| French | 0.382 |
|
119 |
+
| Japanese | 0.479 |
|
120 |
+
| Simplified Chinese | 0.414 |
|