Update README.md
Browse files
README.md
CHANGED
|
@@ -80,3 +80,15 @@ print(sentence_embeddings)
|
|
| 80 |
|
| 81 |
The performance of the pretrained model was evaluated using [ScandEval](https://github.com/ScandEval/ScandEval).
|
| 82 |
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 80 |
|
| 81 |
The performance of the pretrained model was evaluated using [ScandEval](https://github.com/ScandEval/ScandEval).
|
| 82 |
|
| 83 |
+
| Task | Dataset | Score (±SE) |
|
| 84 |
+
|:-------------------------|:-------------|:--------------------------------|
|
| 85 |
+
| sentiment-classification | angry-tweets | mcc = 36.14 (±1.07) |
|
| 86 |
+
| | | macro_f1 = 56.57 (±0.84) |
|
| 87 |
+
| named-entity-recognition | dane | micro_f1 = 55.56 (±1.69) |
|
| 88 |
+
| | | micro_f1_no_misc = 57.44 (±1.9) |
|
| 89 |
+
| linguistic-acceptability | scala-da | mcc = 12.4 (±3.07) |
|
| 90 |
+
| | | macro_f1 = 53.54 (±2.15) |
|
| 91 |
+
| question-answering | scandiqa-da | em = 17.87 (±1.15) |
|
| 92 |
+
| | | f1 = 27.84 (±1.37) |
|
| 93 |
+
| speed | speed | speed = 18.59 (±0.05) |
|
| 94 |
+
|