Update README.md
Browse files
README.md
CHANGED
|
@@ -70,7 +70,9 @@ Due to time constraints, accuracy measures were calculated on a subset of this t
|
|
| 70 |
#### Perplexity
|
| 71 |
The perplexity of the original DictaBERT on the full test set is 22.87.
|
| 72 |
The perplexity of Knesset-DictaBERT on the full test set is 6.60.
|
|
|
|
| 73 |
#### Accuracy
|
|
|
|
| 74 |
- **1-accuracy results**
|
| 75 |
|
| 76 |
Knesset-DictaBERT identified the correct token in the top-1 prediction in 52.55% of the cases.
|
|
@@ -84,6 +86,7 @@ The original Dicta model achieved a top-2 accuracy of 58.60%.
|
|
| 84 |
|
| 85 |
|
| 86 |
- **5-accuracy results**
|
|
|
|
| 87 |
Knesset-DictaBERT identified the correct token within the top-5 predictions in 73.59% of the cases.
|
| 88 |
The original Dicta model achieved a top-5 accuracy of 68.98%.
|
| 89 |
|
|
|
|
| 70 |
#### Perplexity
|
| 71 |
The perplexity of the original DictaBERT on the full test set is 22.87.
|
| 72 |
The perplexity of Knesset-DictaBERT on the full test set is 6.60.
|
| 73 |
+
|
| 74 |
#### Accuracy
|
| 75 |
+
|
| 76 |
- **1-accuracy results**
|
| 77 |
|
| 78 |
Knesset-DictaBERT identified the correct token in the top-1 prediction in 52.55% of the cases.
|
|
|
|
| 86 |
|
| 87 |
|
| 88 |
- **5-accuracy results**
|
| 89 |
+
-
|
| 90 |
Knesset-DictaBERT identified the correct token within the top-5 predictions in 73.59% of the cases.
|
| 91 |
The original Dicta model achieved a top-5 accuracy of 68.98%.
|
| 92 |
|