Update README.md
Browse files
README.md
CHANGED
|
@@ -6,21 +6,23 @@ model-index:
|
|
| 6 |
- name: rttl-ai/SentyBert
|
| 7 |
results:
|
| 8 |
- task:
|
| 9 |
-
type: task-classification
|
| 10 |
-
name: Text Classification
|
| 11 |
dataset:
|
| 12 |
-
type: sst2
|
| 13 |
-
name: sst2
|
| 14 |
config: default
|
| 15 |
-
split: validation
|
| 16 |
metrics:
|
| 17 |
-
|
| 18 |
-
|
| 19 |
-
|
| 20 |
-
|
| 21 |
-
|
| 22 |
-
|
| 23 |
-
|
|
|
|
|
|
|
| 24 |
---
|
| 25 |
|
| 26 |
# rttl-ai/SentyBert
|
|
@@ -35,4 +37,4 @@ This model reaches an accuracy of 99.92 on the dev set.
|
|
| 35 |
- **Resources for more information:**
|
| 36 |
- The model was pre-trained with task-adaptive pre-training [TAPT](https://arxiv.org/pdf/2004.10964.pdf) with an increased masking rate, no corruption strategy, and using WWM, following [this paper](https://aclanthology.org/2023.eacl-main.217.pdf)
|
| 37 |
- fine-tuned on sst with subtrees
|
| 38 |
-
- fine-tuned on sst2
|
|
|
|
| 6 |
- name: rttl-ai/SentyBert
|
| 7 |
results:
|
| 8 |
- task:
|
| 9 |
+
type: task-classification
|
| 10 |
+
name: Text Classification
|
| 11 |
dataset:
|
| 12 |
+
type: sst2
|
| 13 |
+
name: sst2
|
| 14 |
config: default
|
| 15 |
+
split: validation
|
| 16 |
metrics:
|
| 17 |
+
- type: f1
|
| 18 |
+
value: 0.9992
|
| 19 |
+
name: F1 Macro
|
| 20 |
+
- type: accuracy
|
| 21 |
+
value: 0.9992
|
| 22 |
+
name: Accuracy
|
| 23 |
+
datasets:
|
| 24 |
+
- sst2
|
| 25 |
+
- sst
|
| 26 |
---
|
| 27 |
|
| 28 |
# rttl-ai/SentyBert
|
|
|
|
| 37 |
- **Resources for more information:**
|
| 38 |
- The model was pre-trained with task-adaptive pre-training [TAPT](https://arxiv.org/pdf/2004.10964.pdf) with an increased masking rate, no corruption strategy, and using WWM, following [this paper](https://aclanthology.org/2023.eacl-main.217.pdf)
|
| 39 |
- fine-tuned on sst with subtrees
|
| 40 |
+
- fine-tuned on sst2
|