Update README.md
Browse files
README.md
CHANGED
@@ -16,4 +16,14 @@ model-index:
|
|
16 |
metrics:
|
17 |
- type: f1
|
18 |
value: 0.9992
|
19 |
-
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
16 |
metrics:
|
17 |
- type: f1
|
18 |
value: 0.9992
|
19 |
+
|
20 |
+
|
21 |
+
Model Description: This model is a fine-tuned checkpoint of bert-large-uncased, fine-tuned on SST-2. This model reaches F1 of 99.92 on the dev set.
|
22 |
+
Developed by: rttl labs
|
23 |
+
Model Type: Text Classification
|
24 |
+
Language(s): English
|
25 |
+
License: Apache-2.0
|
26 |
+
Resources for more information:
|
27 |
+
The model was pre-trained with task-adaptive pre-training (TAPT), with an increased masking rate, no corruption strategy, and WWM, following this paper.
|
28 |
+
then fine-tuned on sst with subtrees
|
29 |
+
then fine-tuned on sst2
|