Commit
•
e74c5af
1
Parent(s):
789002b
Update README.md
Browse files
README.md
CHANGED
|
@@ -250,6 +250,17 @@ The data is a held-out portion of News Crawl, which has been deduplicated.
|
|
| 250 |
<summary>True-casing Report</summary>
|
| 251 |
|
| 252 |
```text
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 253 |
label precision recall f1 support
|
| 254 |
LOWER (label_id: 0) 99.83 99.81 99.82 2020678
|
| 255 |
UPPER (label_id: 1) 95.51 95.90 95.71 83873
|
|
@@ -264,9 +275,20 @@ The data is a held-out portion of News Crawl, which has been deduplicated.
|
|
| 264 |
|
| 265 |
<details open>
|
| 266 |
|
| 267 |
-
<summary>
|
| 268 |
|
| 269 |
```text
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 270 |
label precision recall f1 support
|
| 271 |
NOSTOP (label_id: 0) 100.00 99.97 99.98 471608
|
| 272 |
FULLSTOP (label_id: 1) 99.63 99.93 99.78 32923
|
|
|
|
| 250 |
<summary>True-casing Report</summary>
|
| 251 |
|
| 252 |
```text
|
| 253 |
+
# With predicted punctuation (not aligned with targets)
|
| 254 |
+
label precision recall f1 support
|
| 255 |
+
LOWER (label_id: 0) 99.76 99.72 99.74 2020678
|
| 256 |
+
UPPER (label_id: 1) 93.32 94.20 93.76 83873
|
| 257 |
+
-------------------
|
| 258 |
+
micro avg 99.50 99.50 99.50 2104551
|
| 259 |
+
macro avg 96.54 96.96 96.75 2104551
|
| 260 |
+
weighted avg 99.50 99.50 99.50 2104551
|
| 261 |
+
|
| 262 |
+
|
| 263 |
+
# With reference punctuation (punctuation matches targets)
|
| 264 |
label precision recall f1 support
|
| 265 |
LOWER (label_id: 0) 99.83 99.81 99.82 2020678
|
| 266 |
UPPER (label_id: 1) 95.51 95.90 95.71 83873
|
|
|
|
| 275 |
|
| 276 |
<details open>
|
| 277 |
|
| 278 |
+
<summary>Sentence Boundary Detection report</summary>
|
| 279 |
|
| 280 |
```text
|
| 281 |
+
# With predicted punctuation (not aligned with targets)
|
| 282 |
+
label precision recall f1 support
|
| 283 |
+
NOSTOP (label_id: 0) 99.59 99.45 99.52 471608
|
| 284 |
+
FULLSTOP (label_id: 1) 91.47 93.53 92.49 29930
|
| 285 |
+
-------------------
|
| 286 |
+
micro avg 99.09 99.09 99.09 501538
|
| 287 |
+
macro avg 95.53 96.49 96.00 501538
|
| 288 |
+
weighted avg 99.10 99.09 99.10 501538
|
| 289 |
+
|
| 290 |
+
|
| 291 |
+
# With reference punctuation (punctuation matches targets)
|
| 292 |
label precision recall f1 support
|
| 293 |
NOSTOP (label_id: 0) 100.00 99.97 99.98 471608
|
| 294 |
FULLSTOP (label_id: 1) 99.63 99.93 99.78 32923
|