1-800-BAD-CODE
commited on
Commit
•
dc7df76
1
Parent(s):
0ad55a4
Update README.md
Browse files
README.md
CHANGED
@@ -89,7 +89,7 @@ Applying all these predictions to the input text, we can punctuate, true-case, a
|
|
89 |
|
90 |
## Tokenizer
|
91 |
|
92 |
-
Instead of the hacky wrapper used by FairSeq and strangely ported (not fixed) by HuggingFace, the xlm-roberta SentencePiece model was adjusted to correctly encode
|
93 |
the text. Per HF's comments,
|
94 |
|
95 |
```python
|
@@ -128,6 +128,8 @@ with open("/path/to/new/sp.model", "wb") as f:
|
|
128 |
f.write(m.SerializeToString())
|
129 |
```
|
130 |
|
|
|
|
|
131 |
|
132 |
## Post-Punctuation Tokens
|
133 |
This model predicts the following set of punctuation tokens after each subtoken:
|
@@ -165,8 +167,7 @@ This model predicts the following set of punctuation tokens before each subword:
|
|
165 |
# Training Details
|
166 |
This model was trained in the NeMo framework.
|
167 |
|
168 |
-
This model was trained on an A100 for
|
169 |
-
For validation and train metrics, see the [Tensorboard Logs](https://tensorboard.dev/experiment/xxnULI1aTeK37vUDL4ejiw/).
|
170 |
|
171 |
## Training Data
|
172 |
This model was trained with News Crawl data from WMT.
|
@@ -176,17 +177,19 @@ This model was trained with News Crawl data from WMT.
|
|
176 |
Languages were chosen based on whether the News Crawl corpus contained enough reliable-quality data as judged by the author.
|
177 |
|
178 |
# Limitations
|
179 |
-
This model was trained on news data, and may not perform well on conversational or informal data.
|
|
|
|
|
180 |
|
181 |
Further, this model is unlikely to be of production quality.
|
182 |
It was trained with "only" 1M lines per language, and the dev sets may have been noisy due to the nature of web-scraped news data.
|
183 |
|
184 |
This model over-predicts Spanish question marks, especially the inverted question mark `¿` (see metrics below).
|
185 |
-
Since `¿` is a rare token, especially in the
|
186 |
-
|
187 |
-
|
188 |
-
|
189 |
-
|
190 |
|
191 |
|
192 |
# Evaluation
|
@@ -211,11 +214,17 @@ Each test example was generated using the following procedure:
|
|
211 |
2. Lower-case the concatenated sentence
|
212 |
3. Remove all punctuation
|
213 |
|
|
|
|
|
214 |
The data is a held-out portion of News Crawl, which has been deduplicated.
|
215 |
3,000 lines of data per language was used, generating 3,000 unique examples of 11 sentences each.
|
216 |
We generate 3,000 examples, where example `i` begins with sentence `i` and is followed by 10 random
|
217 |
sentences selected from the 3,000 sentence test set.
|
218 |
|
|
|
|
|
|
|
|
|
219 |
## Selected Language Evaluation Reports
|
220 |
For now, metrics for a few selected languages are shown below.
|
221 |
Given the amount of work required to collect and pretty-print metrics in 47 languages, I'll add more eventually.
|
@@ -229,11 +238,11 @@ Expand any of the following tabs to see metrics for that language.
|
|
229 |
```text
|
230 |
punct_post test report:
|
231 |
label precision recall f1 support
|
232 |
-
<NULL> (label_id: 0) 99.
|
233 |
-
<ACRONYM> (label_id: 1)
|
234 |
-
. (label_id: 2) 90.
|
235 |
-
, (label_id: 3)
|
236 |
-
? (label_id: 4)
|
237 |
? (label_id: 5) 0.00 0.00 0.00 0
|
238 |
, (label_id: 6) 0.00 0.00 0.00 0
|
239 |
。 (label_id: 7) 0.00 0.00 0.00 0
|
@@ -247,31 +256,31 @@ punct_post test report:
|
|
247 |
፣ (label_id: 15) 0.00 0.00 0.00 0
|
248 |
፧ (label_id: 16) 0.00 0.00 0.00 0
|
249 |
-------------------
|
250 |
-
micro avg 97.
|
251 |
-
macro avg 81.
|
252 |
-
weighted avg 97.
|
253 |
```
|
254 |
|
255 |
-
```
|
256 |
cap test report:
|
257 |
label precision recall f1 support
|
258 |
-
LOWER (label_id: 0) 99.
|
259 |
-
UPPER (label_id: 1) 96.
|
260 |
-------------------
|
261 |
-
micro avg 99.
|
262 |
-
macro avg 98.
|
263 |
-
weighted avg 99.
|
264 |
```
|
265 |
|
266 |
-
```
|
267 |
seg test report:
|
268 |
label precision recall f1 support
|
269 |
-
NOSTOP (label_id: 0) 99.
|
270 |
-
FULLSTOP (label_id: 1) 99.
|
271 |
-------------------
|
272 |
-
micro avg 99.
|
273 |
-
macro avg 99.
|
274 |
-
weighted avg 99.
|
275 |
```
|
276 |
|
277 |
</details>
|
@@ -284,22 +293,22 @@ seg test report:
|
|
284 |
```text
|
285 |
punct_pre test report:
|
286 |
label precision recall f1 support
|
287 |
-
<NULL> (label_id: 0) 99.
|
288 |
-
¿ (label_id: 1)
|
289 |
-------------------
|
290 |
-
micro avg 99.
|
291 |
-
macro avg
|
292 |
-
weighted avg 99.
|
|
|
|
|
293 |
```
|
294 |
-
|
295 |
-
```text
|
296 |
punct_post test report:
|
297 |
label precision recall f1 support
|
298 |
-
<NULL> (label_id: 0) 99.
|
299 |
-
<ACRONYM> (label_id: 1)
|
300 |
-
. (label_id: 2) 91.92
|
301 |
-
, (label_id: 3)
|
302 |
-
? (label_id: 4)
|
303 |
? (label_id: 5) 0.00 0.00 0.00 0
|
304 |
, (label_id: 6) 0.00 0.00 0.00 0
|
305 |
。 (label_id: 7) 0.00 0.00 0.00 0
|
@@ -313,31 +322,31 @@ punct_post test report:
|
|
313 |
፣ (label_id: 15) 0.00 0.00 0.00 0
|
314 |
፧ (label_id: 16) 0.00 0.00 0.00 0
|
315 |
-------------------
|
316 |
-
micro avg 97.
|
317 |
-
macro avg
|
318 |
-
weighted avg 97.
|
319 |
```
|
320 |
|
321 |
-
```
|
322 |
cap test report:
|
323 |
label precision recall f1 support
|
324 |
-
LOWER (label_id: 0) 99.82 99.86 99.84
|
325 |
-
UPPER (label_id: 1) 95.
|
326 |
-------------------
|
327 |
-
micro avg 99.69 99.69 99.69
|
328 |
-
macro avg 97.
|
329 |
-
weighted avg 99.69 99.69 99.69
|
330 |
```
|
331 |
|
332 |
-
```
|
333 |
seg test report:
|
334 |
label precision recall f1 support
|
335 |
-
NOSTOP (label_id: 0) 99.99 99.
|
336 |
-
FULLSTOP (label_id: 1) 99.
|
337 |
-------------------
|
338 |
-
micro avg 99.96 99.96 99.96
|
339 |
-
macro avg 99.
|
340 |
-
weighted avg 99.96 99.96 99.96
|
341 |
```
|
342 |
|
343 |
</details>
|
@@ -349,7 +358,7 @@ seg test report:
|
|
349 |
```text
|
350 |
punct_post test report:
|
351 |
label precision recall f1 support
|
352 |
-
<NULL> (label_id: 0) 99.
|
353 |
<ACRONYM> (label_id: 1) 0.00 0.00 0.00 0
|
354 |
. (label_id: 2) 0.00 0.00 0.00 0
|
355 |
, (label_id: 3) 0.00 0.00 0.00 0
|
@@ -363,35 +372,35 @@ punct_post test report:
|
|
363 |
؟ (label_id: 11) 0.00 0.00 0.00 0
|
364 |
، (label_id: 12) 0.00 0.00 0.00 0
|
365 |
; (label_id: 13) 0.00 0.00 0.00 0
|
366 |
-
። (label_id: 14) 91.
|
367 |
-
፣ (label_id: 15)
|
368 |
-
፧ (label_id: 16)
|
369 |
-------------------
|
370 |
-
micro avg 99.
|
371 |
-
macro avg
|
372 |
-
weighted avg 99.
|
373 |
```
|
374 |
|
375 |
-
```
|
376 |
cap test report:
|
377 |
label precision recall f1 support
|
378 |
-
LOWER (label_id: 0) 98.
|
379 |
-
UPPER (label_id: 1)
|
380 |
-------------------
|
381 |
-
micro avg
|
382 |
-
macro avg
|
383 |
-
weighted avg 96.
|
384 |
```
|
385 |
|
386 |
-
```
|
387 |
seg test report:
|
388 |
label precision recall f1 support
|
389 |
-
NOSTOP (label_id: 0) 99.
|
390 |
-
FULLSTOP (label_id: 1)
|
391 |
-------------------
|
392 |
-
micro avg 99.
|
393 |
-
macro avg
|
394 |
-
weighted avg 99.
|
395 |
```
|
396 |
|
397 |
</details>
|
@@ -403,14 +412,14 @@ seg test report:
|
|
403 |
```text
|
404 |
punct_post test report:
|
405 |
label precision recall f1 support
|
406 |
-
<NULL> (label_id: 0) 99.
|
407 |
<ACRONYM> (label_id: 1) 0.00 0.00 0.00 0
|
408 |
. (label_id: 2) 0.00 0.00 0.00 0
|
409 |
, (label_id: 3) 0.00 0.00 0.00 0
|
410 |
? (label_id: 4) 0.00 0.00 0.00 0
|
411 |
-
? (label_id: 5) 81.
|
412 |
-
, (label_id: 6) 74.93
|
413 |
-
。 (label_id: 7) 96.
|
414 |
、 (label_id: 8) 0.00 0.00 0.00 0
|
415 |
・ (label_id: 9) 0.00 0.00 0.00 0
|
416 |
। (label_id: 10) 0.00 0.00 0.00 0
|
@@ -421,31 +430,31 @@ punct_post test report:
|
|
421 |
፣ (label_id: 15) 0.00 0.00 0.00 0
|
422 |
፧ (label_id: 16) 0.00 0.00 0.00 0
|
423 |
-------------------
|
424 |
-
micro avg 97.
|
425 |
-
macro avg
|
426 |
-
weighted avg 97.
|
427 |
```
|
428 |
|
429 |
-
```
|
430 |
cap test report:
|
431 |
label precision recall f1 support
|
432 |
-
LOWER (label_id: 0) 94.
|
433 |
-
UPPER (label_id: 1)
|
434 |
-------------------
|
435 |
-
micro avg
|
436 |
-
macro avg
|
437 |
-
weighted avg
|
438 |
```
|
439 |
|
440 |
-
```
|
441 |
seg test report:
|
442 |
label precision recall f1 support
|
443 |
-
NOSTOP (label_id: 0) 99.99 99.
|
444 |
-
FULLSTOP (label_id: 1) 99.
|
445 |
-------------------
|
446 |
-
micro avg 99.
|
447 |
-
macro avg 99.
|
448 |
-
weighted avg 99.
|
449 |
```
|
450 |
|
451 |
</details>
|
@@ -457,16 +466,16 @@ seg test report:
|
|
457 |
```text
|
458 |
punct_post test report:
|
459 |
label precision recall f1 support
|
460 |
-
<NULL> (label_id: 0) 99.
|
461 |
<ACRONYM> (label_id: 1) 0.00 0.00 0.00 0
|
462 |
. (label_id: 2) 0.00 0.00 0.00 0
|
463 |
, (label_id: 3) 0.00 0.00 0.00 0
|
464 |
? (label_id: 4) 0.00 0.00 0.00 0
|
465 |
-
? (label_id: 5)
|
466 |
, (label_id: 6) 0.00 0.00 0.00 0
|
467 |
-
。 (label_id: 7)
|
468 |
-
、 (label_id: 8)
|
469 |
-
・ (label_id: 9)
|
470 |
। (label_id: 10) 0.00 0.00 0.00 0
|
471 |
؟ (label_id: 11) 0.00 0.00 0.00 0
|
472 |
، (label_id: 12) 0.00 0.00 0.00 0
|
@@ -475,31 +484,31 @@ punct_post test report:
|
|
475 |
፣ (label_id: 15) 0.00 0.00 0.00 0
|
476 |
፧ (label_id: 16) 0.00 0.00 0.00 0
|
477 |
-------------------
|
478 |
-
micro avg 95.
|
479 |
-
macro avg
|
480 |
-
weighted avg 96.
|
|
|
|
|
481 |
```
|
482 |
-
|
483 |
-
```text
|
484 |
cap test report:
|
485 |
label precision recall f1 support
|
486 |
-
LOWER (label_id: 0) 92.
|
487 |
-
UPPER (label_id: 1)
|
488 |
-------------------
|
489 |
-
micro avg
|
490 |
-
macro avg
|
491 |
-
weighted avg 89.
|
492 |
```
|
493 |
|
494 |
-
```
|
495 |
seg test report:
|
496 |
label precision recall f1 support
|
497 |
-
NOSTOP (label_id: 0) 99.
|
498 |
-
FULLSTOP (label_id: 1) 99.
|
499 |
-------------------
|
500 |
-
micro avg 99.
|
501 |
-
macro avg 99.
|
502 |
-
weighted avg 99.
|
503 |
```
|
504 |
|
505 |
</details>
|
@@ -511,17 +520,17 @@ seg test report:
|
|
511 |
```text
|
512 |
punct_post test report:
|
513 |
label precision recall f1 support
|
514 |
-
<NULL> (label_id: 0) 99.
|
515 |
<ACRONYM> (label_id: 1) 0.00 0.00 0.00 0
|
516 |
. (label_id: 2) 0.00 0.00 0.00 0
|
517 |
-
, (label_id: 3)
|
518 |
-
? (label_id: 4)
|
519 |
? (label_id: 5) 0.00 0.00 0.00 0
|
520 |
, (label_id: 6) 0.00 0.00 0.00 0
|
521 |
。 (label_id: 7) 0.00 0.00 0.00 0
|
522 |
、 (label_id: 8) 0.00 0.00 0.00 0
|
523 |
・ (label_id: 9) 0.00 0.00 0.00 0
|
524 |
-
। (label_id: 10) 96.
|
525 |
؟ (label_id: 11) 0.00 0.00 0.00 0
|
526 |
، (label_id: 12) 0.00 0.00 0.00 0
|
527 |
; (label_id: 13) 0.00 0.00 0.00 0
|
@@ -529,31 +538,85 @@ punct_post test report:
|
|
529 |
፣ (label_id: 15) 0.00 0.00 0.00 0
|
530 |
፧ (label_id: 16) 0.00 0.00 0.00 0
|
531 |
-------------------
|
532 |
-
micro avg 99.11 99.11 99.11
|
533 |
-
macro avg
|
534 |
-
weighted avg 99.
|
|
|
|
|
535 |
```
|
536 |
-
|
537 |
-
```text
|
538 |
cap test report:
|
539 |
label precision recall f1 support
|
540 |
-
LOWER (label_id: 0) 97.
|
541 |
-
UPPER (label_id: 1) 89.
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
542 |
-------------------
|
543 |
-
micro avg
|
544 |
-
macro avg
|
545 |
-
weighted avg
|
546 |
```
|
|
|
|
|
|
|
|
|
|
|
|
|
547 |
|
548 |
```text
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
549 |
seg test report:
|
550 |
label precision recall f1 support
|
551 |
-
NOSTOP (label_id: 0)
|
552 |
-
FULLSTOP (label_id: 1) 99.
|
553 |
-------------------
|
554 |
-
micro avg 99.
|
555 |
-
macro avg 99.
|
556 |
-
weighted avg 99.
|
557 |
```
|
558 |
|
559 |
</details>
|
|
|
89 |
|
90 |
## Tokenizer
|
91 |
|
92 |
+
Instead of the hacky wrapper used by FairSeq and strangely ported (not fixed) by HuggingFace, the `xlm-roberta` SentencePiece model was adjusted to correctly encode
|
93 |
the text. Per HF's comments,
|
94 |
|
95 |
```python
|
|
|
128 |
f.write(m.SerializeToString())
|
129 |
```
|
130 |
|
131 |
+
Now we can use just the SP model without a wrapper.
|
132 |
+
|
133 |
|
134 |
## Post-Punctuation Tokens
|
135 |
This model predicts the following set of punctuation tokens after each subtoken:
|
|
|
167 |
# Training Details
|
168 |
This model was trained in the NeMo framework.
|
169 |
|
170 |
+
This model was trained on an A100 for approximately 9 hours.
|
|
|
171 |
|
172 |
## Training Data
|
173 |
This model was trained with News Crawl data from WMT.
|
|
|
177 |
Languages were chosen based on whether the News Crawl corpus contained enough reliable-quality data as judged by the author.
|
178 |
|
179 |
# Limitations
|
180 |
+
This model was trained on news data, and may not perform well on conversational or informal data. Notably,
|
181 |
+
when presented with many short sentences, the model misses obvious sentence boundaries since the model was
|
182 |
+
trained on relatively-long sentences.
|
183 |
|
184 |
Further, this model is unlikely to be of production quality.
|
185 |
It was trained with "only" 1M lines per language, and the dev sets may have been noisy due to the nature of web-scraped news data.
|
186 |
|
187 |
This model over-predicts Spanish question marks, especially the inverted question mark `¿` (see metrics below).
|
188 |
+
Since `¿` is a rare token, especially in the context of a 47-language model, Spanish questions were over-sampled
|
189 |
+
by selecting more of these sentences from additional training data that was not used. However, this seems to have
|
190 |
+
"over-corrected" the problem and a lot of Spanish question marks are predicted.
|
191 |
+
|
192 |
+
The model may also over-predict commas.
|
193 |
|
194 |
|
195 |
# Evaluation
|
|
|
214 |
2. Lower-case the concatenated sentence
|
215 |
3. Remove all punctuation
|
216 |
|
217 |
+
Targets are generated as we lower-case letters and remove punctuation.
|
218 |
+
|
219 |
The data is a held-out portion of News Crawl, which has been deduplicated.
|
220 |
3,000 lines of data per language was used, generating 3,000 unique examples of 11 sentences each.
|
221 |
We generate 3,000 examples, where example `i` begins with sentence `i` and is followed by 10 random
|
222 |
sentences selected from the 3,000 sentence test set.
|
223 |
|
224 |
+
For measuring true-casing and sentence boundary detection, reference punctuation tokens were used for
|
225 |
+
conditioning (see graph above). If we use predicted punctuation instead, then incorrect punctuation will
|
226 |
+
result in true-casing and SBD targets not aligning correctly and these metrics will be artificially low.
|
227 |
+
|
228 |
## Selected Language Evaluation Reports
|
229 |
For now, metrics for a few selected languages are shown below.
|
230 |
Given the amount of work required to collect and pretty-print metrics in 47 languages, I'll add more eventually.
|
|
|
238 |
```text
|
239 |
punct_post test report:
|
240 |
label precision recall f1 support
|
241 |
+
<NULL> (label_id: 0) 99.25 98.43 98.84 564908
|
242 |
+
<ACRONYM> (label_id: 1) 63.14 84.67 72.33 613
|
243 |
+
. (label_id: 2) 90.97 93.91 92.42 32040
|
244 |
+
, (label_id: 3) 73.95 84.32 78.79 24271
|
245 |
+
? (label_id: 4) 79.05 81.94 80.47 1041
|
246 |
? (label_id: 5) 0.00 0.00 0.00 0
|
247 |
, (label_id: 6) 0.00 0.00 0.00 0
|
248 |
。 (label_id: 7) 0.00 0.00 0.00 0
|
|
|
256 |
፣ (label_id: 15) 0.00 0.00 0.00 0
|
257 |
፧ (label_id: 16) 0.00 0.00 0.00 0
|
258 |
-------------------
|
259 |
+
micro avg 97.60 97.60 97.60 622873
|
260 |
+
macro avg 81.27 88.65 84.57 622873
|
261 |
+
weighted avg 97.77 97.60 97.67 622873
|
262 |
```
|
263 |
|
264 |
+
```
|
265 |
cap test report:
|
266 |
label precision recall f1 support
|
267 |
+
LOWER (label_id: 0) 99.72 99.85 99.78 2134956
|
268 |
+
UPPER (label_id: 1) 96.33 93.52 94.91 91996
|
269 |
-------------------
|
270 |
+
micro avg 99.59 99.59 99.59 2226952
|
271 |
+
macro avg 98.03 96.68 97.34 2226952
|
272 |
+
weighted avg 99.58 99.59 99.58 2226952
|
273 |
```
|
274 |
|
275 |
+
```
|
276 |
seg test report:
|
277 |
label precision recall f1 support
|
278 |
+
NOSTOP (label_id: 0) 99.99 99.98 99.99 591540
|
279 |
+
FULLSTOP (label_id: 1) 99.61 99.89 99.75 34333
|
280 |
-------------------
|
281 |
+
micro avg 99.97 99.97 99.97 625873
|
282 |
+
macro avg 99.80 99.93 99.87 625873
|
283 |
+
weighted avg 99.97 99.97 99.97 625873
|
284 |
```
|
285 |
|
286 |
</details>
|
|
|
293 |
```text
|
294 |
punct_pre test report:
|
295 |
label precision recall f1 support
|
296 |
+
<NULL> (label_id: 0) 99.94 99.89 99.92 636941
|
297 |
+
¿ (label_id: 1) 56.73 71.35 63.20 1288
|
298 |
-------------------
|
299 |
+
micro avg 99.83 99.83 99.83 638229
|
300 |
+
macro avg 78.34 85.62 81.56 638229
|
301 |
+
weighted avg 99.85 99.83 99.84 638229
|
302 |
+
```
|
303 |
+
|
304 |
```
|
|
|
|
|
305 |
punct_post test report:
|
306 |
label precision recall f1 support
|
307 |
+
<NULL> (label_id: 0) 99.19 98.41 98.80 578271
|
308 |
+
<ACRONYM> (label_id: 1) 30.10 56.36 39.24 55
|
309 |
+
. (label_id: 2) 91.92 93.12 92.52 30856
|
310 |
+
, (label_id: 3) 72.98 82.44 77.42 27761
|
311 |
+
? (label_id: 4) 52.77 71.85 60.85 1286
|
312 |
? (label_id: 5) 0.00 0.00 0.00 0
|
313 |
, (label_id: 6) 0.00 0.00 0.00 0
|
314 |
。 (label_id: 7) 0.00 0.00 0.00 0
|
|
|
322 |
፣ (label_id: 15) 0.00 0.00 0.00 0
|
323 |
፧ (label_id: 16) 0.00 0.00 0.00 0
|
324 |
-------------------
|
325 |
+
micro avg 97.40 97.40 97.40 638229
|
326 |
+
macro avg 69.39 80.44 73.77 638229
|
327 |
+
weighted avg 97.60 97.40 97.48 638229
|
328 |
```
|
329 |
|
330 |
+
```
|
331 |
cap test report:
|
332 |
label precision recall f1 support
|
333 |
+
LOWER (label_id: 0) 99.82 99.86 99.84 2324724
|
334 |
+
UPPER (label_id: 1) 95.92 94.70 95.30 79266
|
335 |
-------------------
|
336 |
+
micro avg 99.69 99.69 99.69 2403990
|
337 |
+
macro avg 97.87 97.28 97.57 2403990
|
338 |
+
weighted avg 99.69 99.69 99.69 2403990
|
339 |
```
|
340 |
|
341 |
+
```
|
342 |
seg test report:
|
343 |
label precision recall f1 support
|
344 |
+
NOSTOP (label_id: 0) 99.99 99.96 99.98 607057
|
345 |
+
FULLSTOP (label_id: 1) 99.31 99.88 99.60 34172
|
346 |
-------------------
|
347 |
+
micro avg 99.96 99.96 99.96 641229
|
348 |
+
macro avg 99.65 99.92 99.79 641229
|
349 |
+
weighted avg 99.96 99.96 99.96 641229
|
350 |
```
|
351 |
|
352 |
</details>
|
|
|
358 |
```text
|
359 |
punct_post test report:
|
360 |
label precision recall f1 support
|
361 |
+
<NULL> (label_id: 0) 99.83 99.28 99.56 729664
|
362 |
<ACRONYM> (label_id: 1) 0.00 0.00 0.00 0
|
363 |
. (label_id: 2) 0.00 0.00 0.00 0
|
364 |
, (label_id: 3) 0.00 0.00 0.00 0
|
|
|
372 |
؟ (label_id: 11) 0.00 0.00 0.00 0
|
373 |
، (label_id: 12) 0.00 0.00 0.00 0
|
374 |
; (label_id: 13) 0.00 0.00 0.00 0
|
375 |
+
። (label_id: 14) 91.27 97.90 94.47 25341
|
376 |
+
፣ (label_id: 15) 61.93 82.11 70.60 5818
|
377 |
+
፧ (label_id: 16) 67.41 81.73 73.89 1177
|
378 |
-------------------
|
379 |
+
micro avg 99.08 99.08 99.08 762000
|
380 |
+
macro avg 80.11 90.26 84.63 762000
|
381 |
+
weighted avg 99.21 99.08 99.13 762000
|
382 |
```
|
383 |
|
384 |
+
```
|
385 |
cap test report:
|
386 |
label precision recall f1 support
|
387 |
+
LOWER (label_id: 0) 98.40 98.03 98.21 1064
|
388 |
+
UPPER (label_id: 1) 71.23 75.36 73.24 69
|
389 |
-------------------
|
390 |
+
micro avg 96.65 96.65 96.65 1133
|
391 |
+
macro avg 84.81 86.69 85.73 1133
|
392 |
+
weighted avg 96.74 96.65 96.69 1133
|
393 |
```
|
394 |
|
395 |
+
```
|
396 |
seg test report:
|
397 |
label precision recall f1 support
|
398 |
+
NOSTOP (label_id: 0) 99.99 99.85 99.92 743158
|
399 |
+
FULLSTOP (label_id: 1) 95.20 99.62 97.36 21842
|
400 |
-------------------
|
401 |
+
micro avg 99.85 99.85 99.85 765000
|
402 |
+
macro avg 97.59 99.74 98.64 765000
|
403 |
+
weighted avg 99.85 99.85 99.85 765000
|
404 |
```
|
405 |
|
406 |
</details>
|
|
|
412 |
```text
|
413 |
punct_post test report:
|
414 |
label precision recall f1 support
|
415 |
+
<NULL> (label_id: 0) 99.53 97.31 98.41 435611
|
416 |
<ACRONYM> (label_id: 1) 0.00 0.00 0.00 0
|
417 |
. (label_id: 2) 0.00 0.00 0.00 0
|
418 |
, (label_id: 3) 0.00 0.00 0.00 0
|
419 |
? (label_id: 4) 0.00 0.00 0.00 0
|
420 |
+
? (label_id: 5) 81.85 87.31 84.49 1513
|
421 |
+
, (label_id: 6) 74.08 93.67 82.73 35921
|
422 |
+
。 (label_id: 7) 96.51 96.93 96.72 32097
|
423 |
、 (label_id: 8) 0.00 0.00 0.00 0
|
424 |
・ (label_id: 9) 0.00 0.00 0.00 0
|
425 |
। (label_id: 10) 0.00 0.00 0.00 0
|
|
|
430 |
፣ (label_id: 15) 0.00 0.00 0.00 0
|
431 |
፧ (label_id: 16) 0.00 0.00 0.00 0
|
432 |
-------------------
|
433 |
+
micro avg 97.00 97.00 97.00 505142
|
434 |
+
macro avg 87.99 93.81 90.59 505142
|
435 |
+
weighted avg 97.48 97.00 97.15 505142
|
436 |
```
|
437 |
|
438 |
+
```
|
439 |
cap test report:
|
440 |
label precision recall f1 support
|
441 |
+
LOWER (label_id: 0) 94.89 94.98 94.94 2951
|
442 |
+
UPPER (label_id: 1) 81.34 81.03 81.18 796
|
443 |
-------------------
|
444 |
+
micro avg 92.02 92.02 92.02 3747
|
445 |
+
macro avg 88.11 88.01 88.06 3747
|
446 |
+
weighted avg 92.01 92.02 92.01 3747
|
447 |
```
|
448 |
|
449 |
+
```
|
450 |
seg test report:
|
451 |
label precision recall f1 support
|
452 |
+
NOSTOP (label_id: 0) 99.99 99.97 99.98 473642
|
453 |
+
FULLSTOP (label_id: 1) 99.55 99.90 99.72 34500
|
454 |
-------------------
|
455 |
+
micro avg 99.96 99.96 99.96 508142
|
456 |
+
macro avg 99.77 99.93 99.85 508142
|
457 |
+
weighted avg 99.96 99.96 99.96 508142
|
458 |
```
|
459 |
|
460 |
</details>
|
|
|
466 |
```text
|
467 |
punct_post test report:
|
468 |
label precision recall f1 support
|
469 |
+
<NULL> (label_id: 0) 99.34 95.90 97.59 406341
|
470 |
<ACRONYM> (label_id: 1) 0.00 0.00 0.00 0
|
471 |
. (label_id: 2) 0.00 0.00 0.00 0
|
472 |
, (label_id: 3) 0.00 0.00 0.00 0
|
473 |
? (label_id: 4) 0.00 0.00 0.00 0
|
474 |
+
? (label_id: 5) 70.55 73.56 72.02 1456
|
475 |
, (label_id: 6) 0.00 0.00 0.00 0
|
476 |
+
。 (label_id: 7) 94.38 96.95 95.65 32537
|
477 |
+
、 (label_id: 8) 54.28 87.62 67.03 18610
|
478 |
+
・ (label_id: 9) 28.18 71.64 40.45 1100
|
479 |
। (label_id: 10) 0.00 0.00 0.00 0
|
480 |
؟ (label_id: 11) 0.00 0.00 0.00 0
|
481 |
، (label_id: 12) 0.00 0.00 0.00 0
|
|
|
484 |
፣ (label_id: 15) 0.00 0.00 0.00 0
|
485 |
፧ (label_id: 16) 0.00 0.00 0.00 0
|
486 |
-------------------
|
487 |
+
micro avg 95.51 95.51 95.51 460044
|
488 |
+
macro avg 69.35 85.13 74.55 460044
|
489 |
+
weighted avg 96.91 95.51 96.00 460044
|
490 |
+
```
|
491 |
+
|
492 |
```
|
|
|
|
|
493 |
cap test report:
|
494 |
label precision recall f1 support
|
495 |
+
LOWER (label_id: 0) 92.33 94.03 93.18 4174
|
496 |
+
UPPER (label_id: 1) 83.51 79.46 81.43 1587
|
497 |
-------------------
|
498 |
+
micro avg 90.02 90.02 90.02 5761
|
499 |
+
macro avg 87.92 86.75 87.30 5761
|
500 |
+
weighted avg 89.90 90.02 89.94 5761
|
501 |
```
|
502 |
|
503 |
+
```
|
504 |
seg test report:
|
505 |
label precision recall f1 support
|
506 |
+
NOSTOP (label_id: 0) 99.99 99.92 99.96 428544
|
507 |
+
FULLSTOP (label_id: 1) 99.07 99.87 99.47 34500
|
508 |
-------------------
|
509 |
+
micro avg 99.92 99.92 99.92 463044
|
510 |
+
macro avg 99.53 99.90 99.71 463044
|
511 |
+
weighted avg 99.92 99.92 99.92 463044
|
512 |
```
|
513 |
|
514 |
</details>
|
|
|
520 |
```text
|
521 |
punct_post test report:
|
522 |
label precision recall f1 support
|
523 |
+
<NULL> (label_id: 0) 99.75 99.44 99.59 560358
|
524 |
<ACRONYM> (label_id: 1) 0.00 0.00 0.00 0
|
525 |
. (label_id: 2) 0.00 0.00 0.00 0
|
526 |
+
, (label_id: 3) 69.55 78.48 73.75 8084
|
527 |
+
? (label_id: 4) 63.30 87.07 73.31 317
|
528 |
? (label_id: 5) 0.00 0.00 0.00 0
|
529 |
, (label_id: 6) 0.00 0.00 0.00 0
|
530 |
。 (label_id: 7) 0.00 0.00 0.00 0
|
531 |
、 (label_id: 8) 0.00 0.00 0.00 0
|
532 |
・ (label_id: 9) 0.00 0.00 0.00 0
|
533 |
+
। (label_id: 10) 96.92 98.66 97.78 32118
|
534 |
؟ (label_id: 11) 0.00 0.00 0.00 0
|
535 |
، (label_id: 12) 0.00 0.00 0.00 0
|
536 |
; (label_id: 13) 0.00 0.00 0.00 0
|
|
|
538 |
፣ (label_id: 15) 0.00 0.00 0.00 0
|
539 |
፧ (label_id: 16) 0.00 0.00 0.00 0
|
540 |
-------------------
|
541 |
+
micro avg 99.11 99.11 99.11 600877
|
542 |
+
macro avg 82.38 90.91 86.11 600877
|
543 |
+
weighted avg 99.17 99.11 99.13 600877
|
544 |
+
```
|
545 |
+
|
546 |
```
|
|
|
|
|
547 |
cap test report:
|
548 |
label precision recall f1 support
|
549 |
+
LOWER (label_id: 0) 97.19 96.72 96.95 2466
|
550 |
+
UPPER (label_id: 1) 89.14 90.60 89.86 734
|
551 |
+
-------------------
|
552 |
+
micro avg 95.31 95.31 95.31 3200
|
553 |
+
macro avg 93.17 93.66 93.41 3200
|
554 |
+
weighted avg 95.34 95.31 95.33 3200
|
555 |
+
```
|
556 |
+
|
557 |
+
```
|
558 |
+
seg test report:
|
559 |
+
label precision recall f1 support
|
560 |
+
NOSTOP (label_id: 0) 100.00 99.99 99.99 569472
|
561 |
+
FULLSTOP (label_id: 1) 99.82 99.99 99.91 34405
|
562 |
-------------------
|
563 |
+
micro avg 99.99 99.99 99.99 603877
|
564 |
+
macro avg 99.91 99.99 99.95 603877
|
565 |
+
weighted avg 99.99 99.99 99.99 603877
|
566 |
```
|
567 |
+
|
568 |
+
</details>
|
569 |
+
|
570 |
+
|
571 |
+
<details>
|
572 |
+
<summary>Arabic</summary>
|
573 |
|
574 |
```text
|
575 |
+
punct_post test report:
|
576 |
+
label precision recall f1 support
|
577 |
+
<NULL> (label_id: 0) 99.30 96.94 98.10 688043
|
578 |
+
<ACRONYM> (label_id: 1) 93.33 77.78 84.85 18
|
579 |
+
. (label_id: 2) 93.31 93.78 93.54 28175
|
580 |
+
, (label_id: 3) 0.00 0.00 0.00 0
|
581 |
+
? (label_id: 4) 0.00 0.00 0.00 0
|
582 |
+
? (label_id: 5) 0.00 0.00 0.00 0
|
583 |
+
, (label_id: 6) 0.00 0.00 0.00 0
|
584 |
+
。 (label_id: 7) 0.00 0.00 0.00 0
|
585 |
+
、 (label_id: 8) 0.00 0.00 0.00 0
|
586 |
+
・ (label_id: 9) 0.00 0.00 0.00 0
|
587 |
+
। (label_id: 10) 0.00 0.00 0.00 0
|
588 |
+
؟ (label_id: 11) 65.93 82.79 73.40 860
|
589 |
+
، (label_id: 12) 44.89 79.20 57.30 20941
|
590 |
+
; (label_id: 13) 0.00 0.00 0.00 0
|
591 |
+
። (label_id: 14) 0.00 0.00 0.00 0
|
592 |
+
፣ (label_id: 15) 0.00 0.00 0.00 0
|
593 |
+
፧ (label_id: 16) 0.00 0.00 0.00 0
|
594 |
+
-------------------
|
595 |
+
micro avg 96.29 96.29 96.29 738037
|
596 |
+
macro avg 79.35 86.10 81.44 738037
|
597 |
+
weighted avg 97.49 96.29 96.74 738037
|
598 |
+
```
|
599 |
+
|
600 |
+
```
|
601 |
+
cap test report:
|
602 |
+
label precision recall f1 support
|
603 |
+
LOWER (label_id: 0) 97.10 99.49 98.28 4137
|
604 |
+
UPPER (label_id: 1) 98.71 92.89 95.71 1729
|
605 |
+
-------------------
|
606 |
+
micro avg 97.55 97.55 97.55 5866
|
607 |
+
macro avg 97.90 96.19 96.99 5866
|
608 |
+
weighted avg 97.57 97.55 97.52 5866
|
609 |
+
```
|
610 |
+
|
611 |
+
```
|
612 |
seg test report:
|
613 |
label precision recall f1 support
|
614 |
+
NOSTOP (label_id: 0) 99.99 99.97 99.98 710456
|
615 |
+
FULLSTOP (label_id: 1) 99.39 99.85 99.62 30581
|
616 |
-------------------
|
617 |
+
micro avg 99.97 99.97 99.97 741037
|
618 |
+
macro avg 99.69 99.91 99.80 741037
|
619 |
+
weighted avg 99.97 99.97 99.97 741037
|
620 |
```
|
621 |
|
622 |
</details>
|