Token Classification
Transformers
ONNX
Safetensors
Transformers.js
modernbert
punctuation
punctuation-restoration
truecasing
capitalization
asr
speech-recognition
transcription
mmbert
Instructions to use Horizon-Labs/multilingual-punctuation-truecase-base with libraries, inference providers, notebooks, and local apps. Follow these links to get started.
- Libraries
- Transformers
How to use Horizon-Labs/multilingual-punctuation-truecase-base with Transformers:
# Use a pipeline as a high-level helper from transformers import pipeline pipe = pipeline("token-classification", model="Horizon-Labs/multilingual-punctuation-truecase-base")# pip install -U transformers accelerate # Load model directly from transformers import AutoTokenizer, AutoModelForTokenClassification tokenizer = AutoTokenizer.from_pretrained("Horizon-Labs/multilingual-punctuation-truecase-base") model = AutoModelForTokenClassification.from_pretrained("Horizon-Labs/multilingual-punctuation-truecase-base", device_map="auto") - Transformers.js
How to use Horizon-Labs/multilingual-punctuation-truecase-base with Transformers.js:
// npm i @huggingface/transformers import { pipeline } from '@huggingface/transformers'; // Allocate pipeline const pipe = await pipeline('token-classification', 'Horizon-Labs/multilingual-punctuation-truecase-base'); - Notebooks
- Google Colab
- Kaggle
Download eval/boundary_decoding.json from Horizon-Labs/multilingual-punctuation-truecase-base: direct link, hf CLI and curl.
- Browser
- Download file 12.9 kB
-
https://huggingface.co/Horizon-Labs/multilingual-punctuation-truecase-base/resolve/main/eval/boundary_decoding.json
- Command line
-
hf download hf://Horizon-Labs/multilingual-punctuation-truecase-base/eval/boundary_decoding.json
-
curl -L -o boundary_decoding.json https://huggingface.co/Horizon-Labs/multilingual-punctuation-truecase-base/resolve/main/eval/boundary_decoding.json
12.9 kB
| { | |
| "small": { | |
| "events_val": { | |
| "total": 4400, | |
| "gold_sentence_break": 223, | |
| "gold_comma_cap": 3764, | |
| "gold_lower": 357, | |
| "other": 56, | |
| "words": 1202193 | |
| }, | |
| "events_test": { | |
| "europarl": { | |
| "total": 664, | |
| "gold_sentence_break": 77, | |
| "gold_comma_cap": 413, | |
| "gold_lower": 173, | |
| "other": 1, | |
| "words": 284656 | |
| }, | |
| "ted": { | |
| "total": 613, | |
| "gold_sentence_break": 104, | |
| "gold_comma_cap": 423, | |
| "gold_lower": 82, | |
| "other": 4, | |
| "words": 411238 | |
| }, | |
| "flores": { | |
| "total": 7038, | |
| "gold_sentence_break": 424, | |
| "gold_comma_cap": 5752, | |
| "gold_lower": 842, | |
| "other": 20, | |
| "words": 4419380 | |
| } | |
| }, | |
| "val": { | |
| "1.0": { | |
| "val": { | |
| "punct_f1": 0.7307906190301977, | |
| "case_f1": 0.8294101701010678, | |
| "case_acc": 0.9701137068198062, | |
| "n_langs": 88 | |
| } | |
| }, | |
| "0.8": { | |
| "val": { | |
| "punct_f1": 0.7308791018892663, | |
| "case_f1": 0.8293353894271057, | |
| "case_acc": 0.970133805942292, | |
| "n_langs": 88 | |
| } | |
| }, | |
| "0.6": { | |
| "val": { | |
| "punct_f1": 0.7308620918823107, | |
| "case_f1": 0.8292544930893132, | |
| "case_acc": 0.970123186413515, | |
| "n_langs": 88 | |
| } | |
| }, | |
| "0.5": { | |
| "val": { | |
| "punct_f1": 0.7308040398536643, | |
| "case_f1": 0.8292059312094388, | |
| "case_acc": 0.9701066725875491, | |
| "n_langs": 88 | |
| } | |
| }, | |
| "0.4": { | |
| "val": { | |
| "punct_f1": 0.7306536410704461, | |
| "case_f1": 0.8290889028613576, | |
| "case_acc": 0.9700789277064517, | |
| "n_langs": 88 | |
| } | |
| }, | |
| "0.3": { | |
| "val": { | |
| "punct_f1": 0.7305379377054233, | |
| "case_f1": 0.8286873209137258, | |
| "case_acc": 0.9700531950049228, | |
| "n_langs": 88 | |
| } | |
| }, | |
| "0.2": { | |
| "val": { | |
| "punct_f1": 0.7304026799979512, | |
| "case_f1": 0.8284054240825921, | |
| "case_acc": 0.9700044119117845, | |
| "n_langs": 88 | |
| } | |
| }, | |
| "0.1": { | |
| "val": { | |
| "punct_f1": 0.7297176767296947, | |
| "case_f1": 0.8278816464247921, | |
| "case_acc": 0.9699029242024956, | |
| "n_langs": 88 | |
| } | |
| }, | |
| "0.0": { | |
| "val": { | |
| "punct_f1": 0.7196634079812206, | |
| "case_f1": 0.8220611289124872, | |
| "case_acc": 0.9686916237846855, | |
| "n_langs": 88 | |
| } | |
| } | |
| }, | |
| "test": { | |
| "1.0": { | |
| "europarl": { | |
| "punct_f1": 0.8261380273668834, | |
| "case_f1": 0.8868348243994514, | |
| "case_acc": 0.9782330689261637, | |
| "n_langs": 12 | |
| }, | |
| "flores": { | |
| "punct_f1": 0.5843370537610479, | |
| "case_f1": 0.7838620298007403, | |
| "case_acc": 0.9659089175031863, | |
| "n_langs": 199 | |
| }, | |
| "ted": { | |
| "punct_f1": 0.706248545676473, | |
| "case_f1": 0.8049908396743891, | |
| "case_acc": 0.9751145103698038, | |
| "n_langs": 25 | |
| } | |
| }, | |
| "0.8": { | |
| "europarl": { | |
| "punct_f1": 0.8258141840126738, | |
| "case_f1": 0.8871370399738834, | |
| "case_acc": 0.9783187024691161, | |
| "n_langs": 12 | |
| }, | |
| "flores": { | |
| "punct_f1": 0.5843568111945526, | |
| "case_f1": 0.7838850681906206, | |
| "case_acc": 0.9659211300815793, | |
| "n_langs": 199 | |
| }, | |
| "ted": { | |
| "punct_f1": 0.7061961638539231, | |
| "case_f1": 0.8046943251353734, | |
| "case_acc": 0.9750805574856645, | |
| "n_langs": 25 | |
| } | |
| }, | |
| "0.6": { | |
| "europarl": { | |
| "punct_f1": 0.8256825734989125, | |
| "case_f1": 0.887237083558912, | |
| "case_acc": 0.9783480501202577, | |
| "n_langs": 12 | |
| }, | |
| "flores": { | |
| "punct_f1": 0.5843598083838591, | |
| "case_f1": 0.7838936603668781, | |
| "case_acc": 0.9659318875688618, | |
| "n_langs": 199 | |
| }, | |
| "ted": { | |
| "punct_f1": 0.7062307083406358, | |
| "case_f1": 0.8045444335014676, | |
| "case_acc": 0.9750590037089226, | |
| "n_langs": 25 | |
| } | |
| }, | |
| "0.5": { | |
| "europarl": { | |
| "punct_f1": 0.8255786384826077, | |
| "case_f1": 0.8871950237089855, | |
| "case_acc": 0.9783500364302539, | |
| "n_langs": 12 | |
| }, | |
| "flores": { | |
| "punct_f1": 0.5843301206859476, | |
| "case_f1": 0.7839477721719622, | |
| "case_acc": 0.9659326771546675, | |
| "n_langs": 199 | |
| }, | |
| "ted": { | |
| "punct_f1": 0.7061640185685767, | |
| "case_f1": 0.8045495424337511, | |
| "case_acc": 0.9750656873327851, | |
| "n_langs": 25 | |
| } | |
| }, | |
| "0.4": { | |
| "europarl": { | |
| "punct_f1": 0.8255395996545533, | |
| "case_f1": 0.8873224055392178, | |
| "case_acc": 0.9783943274979, | |
| "n_langs": 12 | |
| }, | |
| "flores": { | |
| "punct_f1": 0.5842993350368568, | |
| "case_f1": 0.7839440474266325, | |
| "case_acc": 0.9659326171881042, | |
| "n_langs": 199 | |
| }, | |
| "ted": { | |
| "punct_f1": 0.7061392849590626, | |
| "case_f1": 0.8045373533500628, | |
| "case_acc": 0.9750655592794151, | |
| "n_langs": 25 | |
| } | |
| }, | |
| "0.3": { | |
| "europarl": { | |
| "punct_f1": 0.8253774115492617, | |
| "case_f1": 0.8872854549432246, | |
| "case_acc": 0.9783932054831085, | |
| "n_langs": 12 | |
| }, | |
| "flores": { | |
| "punct_f1": 0.5842476531168921, | |
| "case_f1": 0.7839607486079377, | |
| "case_acc": 0.9659249202346389, | |
| "n_langs": 199 | |
| }, | |
| "ted": { | |
| "punct_f1": 0.7060579793272004, | |
| "case_f1": 0.8044170340381045, | |
| "case_acc": 0.9750445122390226, | |
| "n_langs": 25 | |
| } | |
| }, | |
| "0.2": { | |
| "europarl": { | |
| "punct_f1": 0.8253888122952203, | |
| "case_f1": 0.8871991148168954, | |
| "case_acc": 0.978382955283062, | |
| "n_langs": 12 | |
| }, | |
| "flores": { | |
| "punct_f1": 0.5841020015222222, | |
| "case_f1": 0.7839006218964324, | |
| "case_acc": 0.9659034125200034, | |
| "n_langs": 199 | |
| }, | |
| "ted": { | |
| "punct_f1": 0.7058535813371581, | |
| "case_f1": 0.804284652813027, | |
| "case_acc": 0.9750194973501509, | |
| "n_langs": 25 | |
| } | |
| }, | |
| "0.1": { | |
| "europarl": { | |
| "punct_f1": 0.8247635923646622, | |
| "case_f1": 0.8869532750016352, | |
| "case_acc": 0.9783390700380635, | |
| "n_langs": 12 | |
| }, | |
| "flores": { | |
| "punct_f1": 0.5837730171644276, | |
| "case_f1": 0.7835788481691437, | |
| "case_acc": 0.9658492383411951, | |
| "n_langs": 199 | |
| }, | |
| "ted": { | |
| "punct_f1": 0.705551059997596, | |
| "case_f1": 0.8035873766125714, | |
| "case_acc": 0.9749633654098845, | |
| "n_langs": 25 | |
| } | |
| }, | |
| "0.0": { | |
| "europarl": { | |
| "punct_f1": 0.8207370341340817, | |
| "case_f1": 0.8845913781807311, | |
| "case_acc": 0.977830321503518, | |
| "n_langs": 12 | |
| }, | |
| "flores": { | |
| "punct_f1": 0.5800550275505717, | |
| "case_f1": 0.7800553476287355, | |
| "case_acc": 0.9652901359825775, | |
| "n_langs": 199 | |
| }, | |
| "ted": { | |
| "punct_f1": 0.7031424771140524, | |
| "case_f1": 0.801901972733539, | |
| "case_acc": 0.9746559663448485, | |
| "n_langs": 25 | |
| } | |
| } | |
| }, | |
| "best_beta_val": 0.8 | |
| }, | |
| "base": { | |
| "events_val": { | |
| "total": 4439, | |
| "gold_sentence_break": 243, | |
| "gold_comma_cap": 3723, | |
| "gold_lower": 412, | |
| "other": 61, | |
| "words": 1202193 | |
| }, | |
| "events_test": { | |
| "europarl": { | |
| "total": 711, | |
| "gold_sentence_break": 99, | |
| "gold_comma_cap": 398, | |
| "gold_lower": 212, | |
| "other": 2, | |
| "words": 284656 | |
| }, | |
| "ted": { | |
| "total": 707, | |
| "gold_sentence_break": 128, | |
| "gold_comma_cap": 441, | |
| "gold_lower": 132, | |
| "other": 6, | |
| "words": 411238 | |
| }, | |
| "flores": { | |
| "total": 7508, | |
| "gold_sentence_break": 529, | |
| "gold_comma_cap": 6018, | |
| "gold_lower": 946, | |
| "other": 15, | |
| "words": 4419380 | |
| } | |
| }, | |
| "val": { | |
| "1.0": { | |
| "val": { | |
| "punct_f1": 0.734981073040326, | |
| "case_f1": 0.8270321373775785, | |
| "case_acc": 0.9704420574182244, | |
| "n_langs": 88 | |
| } | |
| }, | |
| "0.8": { | |
| "val": { | |
| "punct_f1": 0.7348666312217191, | |
| "case_f1": 0.8267457448808136, | |
| "case_acc": 0.9704585962291421, | |
| "n_langs": 88 | |
| } | |
| }, | |
| "0.6": { | |
| "val": { | |
| "punct_f1": 0.7347106737402835, | |
| "case_f1": 0.8265297379124928, | |
| "case_acc": 0.9704554282536677, | |
| "n_langs": 88 | |
| } | |
| }, | |
| "0.5": { | |
| "val": { | |
| "punct_f1": 0.7346246467121375, | |
| "case_f1": 0.8263045026047617, | |
| "case_acc": 0.9704284057283917, | |
| "n_langs": 88 | |
| } | |
| }, | |
| "0.4": { | |
| "val": { | |
| "punct_f1": 0.7345426973080933, | |
| "case_f1": 0.8262052860225652, | |
| "case_acc": 0.970415196049132, | |
| "n_langs": 88 | |
| } | |
| }, | |
| "0.3": { | |
| "val": { | |
| "punct_f1": 0.7344502934444563, | |
| "case_f1": 0.8260142785136597, | |
| "case_acc": 0.970376844055259, | |
| "n_langs": 88 | |
| } | |
| }, | |
| "0.2": { | |
| "val": { | |
| "punct_f1": 0.7343913818645267, | |
| "case_f1": 0.8253554314280407, | |
| "case_acc": 0.9703197116293819, | |
| "n_langs": 88 | |
| } | |
| }, | |
| "0.1": { | |
| "val": { | |
| "punct_f1": 0.7335927979300151, | |
| "case_f1": 0.8246970384510802, | |
| "case_acc": 0.9701688970573676, | |
| "n_langs": 88 | |
| } | |
| }, | |
| "0.0": { | |
| "val": { | |
| "punct_f1": 0.7244328192302032, | |
| "case_f1": 0.8192754657610234, | |
| "case_acc": 0.9689914404283926, | |
| "n_langs": 88 | |
| } | |
| } | |
| }, | |
| "test": { | |
| "1.0": { | |
| "europarl": { | |
| "punct_f1": 0.8356902913059253, | |
| "case_f1": 0.8972502043630105, | |
| "case_acc": 0.9802417869552379, | |
| "n_langs": 12 | |
| }, | |
| "flores": { | |
| "punct_f1": 0.6017388694156922, | |
| "case_f1": 0.7956324531168975, | |
| "case_acc": 0.968557954295429, | |
| "n_langs": 199 | |
| }, | |
| "ted": { | |
| "punct_f1": 0.7233833225634991, | |
| "case_f1": 0.8161406092308486, | |
| "case_acc": 0.9766803448154714, | |
| "n_langs": 25 | |
| } | |
| }, | |
| "0.8": { | |
| "europarl": { | |
| "punct_f1": 0.8355590073510712, | |
| "case_f1": 0.8975453764127727, | |
| "case_acc": 0.980333230446175, | |
| "n_langs": 12 | |
| }, | |
| "flores": { | |
| "punct_f1": 0.6017946260654395, | |
| "case_f1": 0.7956525094132423, | |
| "case_acc": 0.9685799749478646, | |
| "n_langs": 199 | |
| }, | |
| "ted": { | |
| "punct_f1": 0.723420803444116, | |
| "case_f1": 0.8157411383616665, | |
| "case_acc": 0.9767111203268057, | |
| "n_langs": 25 | |
| } | |
| }, | |
| "0.6": { | |
| "europarl": { | |
| "punct_f1": 0.835660038747121, | |
| "case_f1": 0.8976584719195407, | |
| "case_acc": 0.9803770897220884, | |
| "n_langs": 12 | |
| }, | |
| "flores": { | |
| "punct_f1": 0.6017855881553794, | |
| "case_f1": 0.7956371738121963, | |
| "case_acc": 0.968583272794559, | |
| "n_langs": 199 | |
| }, | |
| "ted": { | |
| "punct_f1": 0.7234164468485581, | |
| "case_f1": 0.8155795294082455, | |
| "case_acc": 0.9766764383360653, | |
| "n_langs": 25 | |
| } | |
| }, | |
| "0.5": { | |
| "europarl": { | |
| "punct_f1": 0.8354054855518247, | |
| "case_f1": 0.8977266755821943, | |
| "case_acc": 0.9804040392940259, | |
| "n_langs": 12 | |
| }, | |
| "flores": { | |
| "punct_f1": 0.6017891079366826, | |
| "case_f1": 0.7956693394169906, | |
| "case_acc": 0.9685852441493816, | |
| "n_langs": 199 | |
| }, | |
| "ted": { | |
| "punct_f1": 0.7234462672576149, | |
| "case_f1": 0.8155163004506236, | |
| "case_acc": 0.9766731955454752, | |
| "n_langs": 25 | |
| } | |
| }, | |
| "0.4": { | |
| "europarl": { | |
| "punct_f1": 0.8354378021674701, | |
| "case_f1": 0.8977978726694159, | |
| "case_acc": 0.98042655471346, | |
| "n_langs": 12 | |
| }, | |
| "flores": { | |
| "punct_f1": 0.6017406899891955, | |
| "case_f1": 0.7956777925955816, | |
| "case_acc": 0.9685811177493032, | |
| "n_langs": 199 | |
| }, | |
| "ted": { | |
| "punct_f1": 0.7233887996468998, | |
| "case_f1": 0.8154746177961347, | |
| "case_acc": 0.9766681575598567, | |
| "n_langs": 25 | |
| } | |
| }, | |
| "0.3": { | |
| "europarl": { | |
| "punct_f1": 0.835163107232471, | |
| "case_f1": 0.897723039353909, | |
| "case_acc": 0.9804253800940185, | |
| "n_langs": 12 | |
| }, | |
| "flores": { | |
| "punct_f1": 0.6016212362387191, | |
| "case_f1": 0.7956116551634445, | |
| "case_acc": 0.9685685178074583, | |
| "n_langs": 199 | |
| }, | |
| "ted": { | |
| "punct_f1": 0.7232371368010861, | |
| "case_f1": 0.8153437742548227, | |
| "case_acc": 0.9766409054516443, | |
| "n_langs": 25 | |
| } | |
| }, | |
| "0.2": { | |
| "europarl": { | |
| "punct_f1": 0.8346520106457258, | |
| "case_f1": 0.8976163830920548, | |
| "case_acc": 0.9804076817739165, | |
| "n_langs": 12 | |
| }, | |
| "flores": { | |
| "punct_f1": 0.601488769752149, | |
| "case_f1": 0.7954701827995039, | |
| "case_acc": 0.9685452787349534, | |
| "n_langs": 199 | |
| }, | |
| "ted": { | |
| "punct_f1": 0.7230915320428797, | |
| "case_f1": 0.8151793877403215, | |
| "case_acc": 0.976604196774716, | |
| "n_langs": 25 | |
| } | |
| }, | |
| "0.1": { | |
| "europarl": { | |
| "punct_f1": 0.8340997066333364, | |
| "case_f1": 0.8972390895921495, | |
| "case_acc": 0.9803317671033537, | |
| "n_langs": 12 | |
| }, | |
| "flores": { | |
| "punct_f1": 0.6011514948042289, | |
| "case_f1": 0.7951917464498147, | |
| "case_acc": 0.9684931406121766, | |
| "n_langs": 199 | |
| }, | |
| "ted": { | |
| "punct_f1": 0.7227439821343179, | |
| "case_f1": 0.8145529445268722, | |
| "case_acc": 0.9765387513873018, | |
| "n_langs": 25 | |
| } | |
| }, | |
| "0.0": { | |
| "europarl": { | |
| "punct_f1": 0.8306300630299157, | |
| "case_f1": 0.8950221595849377, | |
| "case_acc": 0.9798176253769387, | |
| "n_langs": 12 | |
| }, | |
| "flores": { | |
| "punct_f1": 0.5965025018323125, | |
| "case_f1": 0.7921952002560139, | |
| "case_acc": 0.9679911229070182, | |
| "n_langs": 199 | |
| }, | |
| "ted": { | |
| "punct_f1": 0.7204567965506813, | |
| "case_f1": 0.8127051707077392, | |
| "case_acc": 0.9761902328619462, | |
| "n_langs": 25 | |
| } | |
| } | |
| }, | |
| "best_beta_val": 1.0 | |
| } | |
| } |