links
Browse files
README.md
CHANGED
|
@@ -2,6 +2,7 @@
|
|
| 2 |
language:
|
| 3 |
- 'no'
|
| 4 |
- nn
|
|
|
|
| 5 |
tags:
|
| 6 |
- sentence-transformers
|
| 7 |
- sentence-similarity
|
|
@@ -110,7 +111,8 @@ print(similarities)
|
|
| 110 |
To verify the utility of our models, we evaluated them on a selection of classification and clustering tasks for Norwegian from [MTEBv2](https://embeddings-benchmark.github.io/mteb/).
|
| 111 |
|
| 112 |
The heatmap below shows the results of evaluating five sentence-transformers on ten different tasks;
|
| 113 |
-
three of the sentence-transformer models we have fine-tuned
|
|
|
|
| 114 |
and the other two are relatively popular (and comparable) sentence similarity models ([FFI/SimCSE-NB-BERT-large](https://huggingface.co/FFI/SimCSE-NB-BERT-large) and [NbAiLab/nb-sbert-base](https://huggingface.co/NbAiLab/nb-sbert-base)).
|
| 115 |
|
| 116 |
|
|
@@ -121,10 +123,10 @@ We ranked the models using **Borda count** (which is used in MTEB), where each m
|
|
| 121 |
| Rank | Model | Borda Points |
|
| 122 |
|:----:|:--------------------------------|:-------------:|
|
| 123 |
| 1 | **Fremtind/norsbert4-large** | **44** |
|
| 124 |
-
| 2 | FFI/SimCSE-NB-BERT-large | 40 |
|
| 125 |
-
| 3 | Fremtind/norsbert4-base | 24 |
|
| 126 |
-
| 4 | NbAiLab/nb-sbert-base | 15 |
|
| 127 |
-
| 5 | Fremtind/mmBERT-base-norwegian | 7 |
|
| 128 |
|
| 129 |
|
| 130 |
## Training Details
|
|
|
|
| 2 |
language:
|
| 3 |
- 'no'
|
| 4 |
- nn
|
| 5 |
+
- nb
|
| 6 |
tags:
|
| 7 |
- sentence-transformers
|
| 8 |
- sentence-similarity
|
|
|
|
| 111 |
To verify the utility of our models, we evaluated them on a selection of classification and clustering tasks for Norwegian from [MTEBv2](https://embeddings-benchmark.github.io/mteb/).
|
| 112 |
|
| 113 |
The heatmap below shows the results of evaluating five sentence-transformers on ten different tasks;
|
| 114 |
+
three of the sentence-transformer models we have fine-tuned
|
| 115 |
+
([Fremtind/norsbert4-large](https://huggingface.co/Fremtind/norsbert4-large), [Fremtind/norsbert4-base](https://huggingface.co/Fremtind/norsbert4-base), [Fremtind/mmBERT-base-norwegian](https://huggingface.co/Fremtind/mmBERT-base-norwegian))
|
| 116 |
and the other two are relatively popular (and comparable) sentence similarity models ([FFI/SimCSE-NB-BERT-large](https://huggingface.co/FFI/SimCSE-NB-BERT-large) and [NbAiLab/nb-sbert-base](https://huggingface.co/NbAiLab/nb-sbert-base)).
|
| 117 |
|
| 118 |
|
|
|
|
| 123 |
| Rank | Model | Borda Points |
|
| 124 |
|:----:|:--------------------------------|:-------------:|
|
| 125 |
| 1 | **Fremtind/norsbert4-large** | **44** |
|
| 126 |
+
| 2 | [FFI/SimCSE-NB-BERT-large](https://huggingface.co/FFI/SimCSE-NB-BERT-large) | 40 |
|
| 127 |
+
| 3 | [Fremtind/norsbert4-base](https://huggingface.co/Murhaf/norsbert4-base) | 24 |
|
| 128 |
+
| 4 | [NbAiLab/nb-sbert-base](https://huggingface.co/NbAiLab/nb-sbert-base) | 15 |
|
| 129 |
+
| 5 | [Fremtind/mmBERT-base-norwegian](https://huggingface.co/Fremtind/mmBERT-base-norwegian) | 7 |
|
| 130 |
|
| 131 |
|
| 132 |
## Training Details
|