Update README.md
Browse files
README.md
CHANGED
@@ -24,7 +24,7 @@ aristoBERTo is provided by the [Diogenet project](https://diogenet.ucsd.edu) of
|
|
24 |
|
25 |
## Intended uses
|
26 |
|
27 |
-
This model was created for fine-tuning with spaCy and the ancient Greek Universal Dependency datasets as well as a NER corpus produced by the [Diogenet project](https://diogenet.ucsd.edu).
|
28 |
|
29 |
|
30 |
It achieves the following results on the evaluation set:
|
@@ -48,25 +48,6 @@ The following hyperparameters were used during training:
|
|
48 |
|
49 |
| Training Loss | Epoch | Step | Validation Loss |
|
50 |
|:-------------:|:-----:|:-------:|:---------------:|
|
51 |
-
| 2.2017 | 1.0 | 170711 | 2.4411 |
|
52 |
-
| 2.0302 | 2.0 | 341422 | 2.2853 |
|
53 |
-
| 1.9459 | 3.0 | 512133 | 2.1911 |
|
54 |
-
| 1.8847 | 4.0 | 682844 | 2.1187 |
|
55 |
-
| 1.85 | 5.0 | 853555 | 2.0841 |
|
56 |
-
| 1.834 | 6.0 | 1024266 | 2.0267 |
|
57 |
-
| 1.7599 | 7.0 | 1194977 | 1.9874 |
|
58 |
-
| 1.7147 | 8.0 | 1365688 | 1.9495 |
|
59 |
-
| 1.6882 | 9.0 | 1536399 | 1.9030 |
|
60 |
-
| 1.6433 | 10.0 | 1707110 | 1.8845 |
|
61 |
-
| 1.6289 | 11.0 | 1877821 | 1.8485 |
|
62 |
-
| 1.5758 | 12.0 | 2048532 | 1.8097 |
|
63 |
-
| 1.5439 | 13.0 | 2219243 | 1.7916 |
|
64 |
-
| 1.5358 | 14.0 | 2389954 | 1.7680 |
|
65 |
-
| 1.4946 | 15.0 | 2560665 | 1.7360 |
|
66 |
-
| 1.4753 | 16.0 | 2731376 | 1.7052 |
|
67 |
-
| 1.4628 | 17.0 | 2902087 | 1.6804 |
|
68 |
-
| 1.4223 | 18.0 | 3072798 | 1.6720 |
|
69 |
-
| 1.3909 | 19.0 | 3243509 | 1.6521 |
|
70 |
| 1.377 | 20.0 | 3414220 | 1.6314 |
|
71 |
|
72 |
|
|
|
24 |
|
25 |
## Intended uses
|
26 |
|
27 |
+
This model was created for fine-tuning with spaCy and the ancient Greek Universal Dependency datasets as well as a NER corpus produced by the [Diogenet project](https://diogenet.ucsd.edu). As a fill-mask model, AristoBERTo can also be used in the restoration of damaged Greek papyri, inscriptions, and manuscripts.
|
28 |
|
29 |
|
30 |
It achieves the following results on the evaluation set:
|
|
|
48 |
|
49 |
| Training Loss | Epoch | Step | Validation Loss |
|
50 |
|:-------------:|:-----:|:-------:|:---------------:|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
51 |
| 1.377 | 20.0 | 3414220 | 1.6314 |
|
52 |
|
53 |
|