Commit
•
38ea57a
1
Parent(s):
2f391ad
Update README.md
Browse files
README.md
CHANGED
|
@@ -172,3 +172,53 @@ This is also a base-sized model with many languages and many tasks, so capacity
|
|
| 172 |
|
| 173 |
|
| 174 |
# Evaluation
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 172 |
|
| 173 |
|
| 174 |
# Evaluation
|
| 175 |
+
In these metrics, keep in mind that
|
| 176 |
+
1. That data is noisy
|
| 177 |
+
2. Sentence boundaries and true-casing is conditioned on predicted punctuation
|
| 178 |
+
|
| 179 |
+
<details>
|
| 180 |
+
<summary>English</summary>
|
| 181 |
+
|
| 182 |
+
```
|
| 183 |
+
punct_post test report:
|
| 184 |
+
label precision recall f1 support
|
| 185 |
+
<NULL> (label_id: 0) 98.71 98.69 98.70 107750
|
| 186 |
+
. (label_id: 1) 87.82 88.89 88.36 6005
|
| 187 |
+
, (label_id: 2) 67.90 67.24 67.57 3571
|
| 188 |
+
? (label_id: 3) 80.51 78.19 79.33 486
|
| 189 |
+
？ (label_id: 4) 0.00 0.00 0.00 0
|
| 190 |
+
， (label_id: 5) 0.00 0.00 0.00 0
|
| 191 |
+
。 (label_id: 6) 0.00 0.00 0.00 0
|
| 192 |
+
、 (label_id: 7) 0.00 0.00 0.00 0
|
| 193 |
+
・ (label_id: 8) 0.00 0.00 0.00 0
|
| 194 |
+
। (label_id: 9) 0.00 0.00 0.00 0
|
| 195 |
+
؟ (label_id: 10) 0.00 0.00 0.00 0
|
| 196 |
+
، (label_id: 11) 0.00 0.00 0.00 0
|
| 197 |
+
; (label_id: 12) 0.00 0.00 0.00 0
|
| 198 |
+
። (label_id: 13) 0.00 0.00 0.00 0
|
| 199 |
+
፣ (label_id: 14) 0.00 0.00 0.00 0
|
| 200 |
+
፧ (label_id: 15) 0.00 0.00 0.00 0
|
| 201 |
+
-------------------
|
| 202 |
+
micro avg 97.15 97.15 97.15 117812
|
| 203 |
+
macro avg 83.74 83.25 83.49 117812
|
| 204 |
+
weighted avg 97.15 97.15 97.15 117812
|
| 205 |
+
|
| 206 |
+
cap test report:
|
| 207 |
+
label precision recall f1 support
|
| 208 |
+
LOWER (label_id: 0) 99.62 99.49 99.56 362399
|
| 209 |
+
UPPER (label_id: 1) 89.11 91.75 90.41 16506
|
| 210 |
+
-------------------
|
| 211 |
+
micro avg 99.15 99.15 99.15 378905
|
| 212 |
+
macro avg 94.37 95.62 94.98 378905
|
| 213 |
+
weighted avg 99.17 99.15 99.16 378905
|
| 214 |
+
|
| 215 |
+
seg test report:
|
| 216 |
+
label precision recall f1 support
|
| 217 |
+
NOSTOP (label_id: 0) 99.29 99.43 99.36 111466
|
| 218 |
+
FULLSTOP (label_id: 1) 89.69 87.49 88.58 6346
|
| 219 |
+
-------------------
|
| 220 |
+
micro avg 98.78 98.78 98.78 117812
|
| 221 |
+
macro avg 94.49 93.46 93.97 117812
|
| 222 |
+
weighted avg 98.77 98.78 98.78 117812
|
| 223 |
+
```
|
| 224 |
+
</details>
|