Commit
•
f6a5fe8
1
Parent(s):
b498c8d
Update README.md
Browse files
README.md
CHANGED
|
@@ -1,3 +1,18 @@
|
|
| 1 |
---
|
| 2 |
license: apache-2.0
|
| 3 |
---
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
---
|
| 2 |
license: apache-2.0
|
| 3 |
---
|
| 4 |
+
|
| 5 |
+
RoBERTa-based sentence-level classification model. Similar to alex2awesome/quote-detection__roberta-base-sentence, except it's trained on a lot more data.
|
| 6 |
+
|
| 7 |
+
Scores this accuracy on the original gold-label training dataset mentioned in: https://arxiv.org/pdf/2305.14904.pdf
|
| 8 |
+
|
| 9 |
+
```{
|
| 10 |
+
"eval_samples": 86,
|
| 11 |
+
"test_f1": 0.8301562276049879,
|
| 12 |
+
"test_loss": 0.4262699034650141,
|
| 13 |
+
"test_runtime": 6.9868,
|
| 14 |
+
"test_samples_per_second": 12.309,
|
| 15 |
+
"test_steps_per_second": 12.309
|
| 16 |
+
}```
|
| 17 |
+
|
| 18 |
+
```{"DIRECT QUOTE_f1": 0.9350419796730004, "full_f1": 0.8307165286896945, "BACKGROUND_f1": 0.8335517693315858, "NO QUOTE_f1": 0.0, "INDIRECT QUOTE_f1": 0.8223510806536637, "PUBLISHED WORK_f1": 0.8823529411764706, "STATEMENT_f1": 0.9511400651465798, "PRESS REPORT_f1": 0.9538461538461539, "DECLINED COMMENT_f1": 1.0, "SOCIAL MEDIA POST_f1": 0.8695652173913044, "PROPOSAL/ORDER/LAW_f1": 0.8068181818181819, "PRICE SIGNAL_f1": 0.5, "NARRATIVE_f1": 0.9322709163346613, "DIRECT OBSERVATION_f1": 0.5316455696202532, "COMMUNICATION_f1": 0.975609756097561, "PUBLIC SPEECH_f1": 0.9473684210526316, "VOTE/POLL_f1": 0.8205128205128205, "COURT PROCEEDING_f1": 0.9491525423728813}```
|