Pre-trained models

cs

language code: cs, language name: czech, training sample size: 229k

model file: cs-sentence-detector.onlpm, evaluation report: cs-sentence-detector.txt, training algorithm: MAXENT, evaluation score: 85.47%
model file: cs-tokenizer.onlpm, evaluation report: cs-tokenizer.txt, training algorithm: MAXENT_QN, evaluation score: 99.95%
model file: cs-pos-tagger.onlpm, evaluation report: cs-pos-tagger.txt, training algorithm: MAXENT, evaluation score: 97.94%
model file: cs-lemmatizer.onlpm, evaluation report: cs-lemmatizer.txt, training algorithm: MAXENT, evaluation score: 93.83%

da

language code: da, language name: danish, training sample size: 8k

model file: da-sentence-detector.onlpm, evaluation report: da-sentence-detector.txt, training algorithm: MAXENT_QN, evaluation score: 87.73%
model file: da-tokenizer.onlpm, evaluation report: da-tokenizer.txt, training algorithm: MAXENT_QN, evaluation score: 99.89%
model file: da-pos-tagger.onlpm, evaluation report: da-pos-tagger.txt, training algorithm: PERCEPTRON, evaluation score: 96.24%
model file: da-lemmatizer.onlpm, evaluation report: da-lemmatizer.txt, training algorithm: MAXENT, evaluation score: 94.29%

de

language code: de, language name: german, training sample size: 124k

model file: de-sentence-detector.onlpm, evaluation report: de-sentence-detector.txt, training algorithm: MAXENT_QN, evaluation score: 72.88%
model file: de-tokenizer.onlpm, evaluation report: de-tokenizer.txt, training algorithm: MAXENT_QN, evaluation score: 99.93%
model file: de-pos-tagger.onlpm, evaluation report: de-pos-tagger.txt, training algorithm: MAXENT, evaluation score: 97.20%
model file: de-lemmatizer.onlpm, evaluation report: de-lemmatizer.txt, training algorithm: MAXENT, evaluation score: 85.02%

el

language code: el, language name: greek, training sample size: 7k

model file: el-sentence-detector.onlpm, evaluation report: el-sentence-detector.txt, training algorithm: MAXENT_QN, evaluation score: 87.08%
model file: el-tokenizer.onlpm, evaluation report: el-tokenizer.txt, training algorithm: MAXENT_QN, evaluation score: 99.97%
model file: el-pos-tagger.onlpm, evaluation report: el-pos-tagger.txt, training algorithm: MAXENT, evaluation score: 97.82%
model file: el-lemmatizer.onlpm, evaluation report: el-lemmatizer.txt, training algorithm: MAXENT, evaluation score: 96.83%

en

language code: en, language name: english, training sample size: 75k

model file: en-sentence-detector.onlpm, evaluation report: en-sentence-detector.txt, training algorithm: MAXENT_QN, evaluation score: 75.93%
model file: en-tokenizer.onlpm, evaluation report: en-tokenizer.txt, training algorithm: MAXENT_QN, evaluation score: 99.64%
model file: en-pos-tagger.onlpm, evaluation report: en-pos-tagger.txt, training algorithm: MAXENT, evaluation score: 96.72%
model file: en-lemmatizer.onlpm, evaluation report: en-lemmatizer.txt, training algorithm: MAXENT, evaluation score: 93.94%

es

language code: es, language name: spanish, training sample size: 61k

model file: es-sentence-detector.onlpm, evaluation report: es-sentence-detector.txt, training algorithm: MAXENT_QN, evaluation score: 97.27%
model file: es-tokenizer.onlpm, evaluation report: es-tokenizer.txt, training algorithm: MAXENT_QN, evaluation score: 99.92%
model file: es-pos-tagger.onlpm, evaluation report: es-pos-tagger.txt, training algorithm: MAXENT, evaluation score: 95.94%
model file: es-lemmatizer.onlpm, evaluation report: es-lemmatizer.txt, training algorithm: MAXENT, evaluation score: 95.86%

fi

language code: fi, language name: finnish, training sample size: 59k

model file: fi-sentence-detector.onlpm, evaluation report: fi-sentence-detector.txt, training algorithm: MAXENT_QN, evaluation score: 81.95%
model file: fi-tokenizer.onlpm, evaluation report: fi-tokenizer.txt, training algorithm: MAXENT_QN, evaluation score: 99.76%
model file: fi-pos-tagger.onlpm, evaluation report: fi-pos-tagger.txt, training algorithm: MAXENT, evaluation score: 93.68%
model file: fi-lemmatizer.onlpm, evaluation report: fi-lemmatizer.txt, training algorithm: MAXENT, evaluation score: 93.76%

fr

language code: fr, language name: french, training sample size: 42k

model file: fr-sentence-detector.onlpm, evaluation report: fr-sentence-detector.txt, training algorithm: MAXENT_QN, evaluation score: 92.62%
model file: fr-tokenizer.onlpm, evaluation report: fr-tokenizer.txt, training algorithm: MAXENT_QN, evaluation score: 99.94%
model file: fr-pos-tagger.onlpm, evaluation report: fr-pos-tagger.txt, training algorithm: MAXENT, evaluation score: 96.90%
model file: fr-lemmatizer.onlpm, evaluation report: fr-lemmatizer.txt, training algorithm: MAXENT, evaluation score: 95.10%

he

language code: he, language name: hebrew, training sample size: 11k

model file: he-sentence-detector.onlpm, evaluation report: he-sentence-detector.txt, training algorithm: MAXENT_QN, evaluation score: 95.74%
model file: he-tokenizer.onlpm, evaluation report: he-tokenizer.txt, training algorithm: MAXENT_QN, evaluation score: 92.59%
model file: he-pos-tagger.onlpm, evaluation report: he-pos-tagger.txt, training algorithm: MAXENT, evaluation score: 94.68%
model file: he-lemmatizer.onlpm, evaluation report: he-lemmatizer.txt, training algorithm: MAXENT, evaluation score: 96.93%

it

language code: it, language name: italian, training sample size: 72k

model file: it-sentence-detector.onlpm, evaluation report: it-sentence-detector.txt, training algorithm: MAXENT, evaluation score: 74.14%
model file: it-tokenizer.onlpm, evaluation report: it-tokenizer.txt, training algorithm: MAXENT_QN, evaluation score: 99.79%
model file: it-pos-tagger.onlpm, evaluation report: it-pos-tagger.txt, training algorithm: MAXENT, evaluation score: 96.80%
model file: it-lemmatizer.onlpm, evaluation report: it-lemmatizer.txt, training algorithm: MAXENT, evaluation score: 96.07%

ja

language code: ja, language name: japanese, training sample size: 18k

model file: ja-sentence-detector.onlpm, evaluation report: ja-sentence-detector.txt, training algorithm: MAXENT, evaluation score: 99.25%
model file: ja-tokenizer.onlpm, evaluation report: ja-tokenizer.txt, training algorithm: MAXENT_QN, evaluation score: 81.43%
model file: ja-pos-tagger.onlpm, evaluation report: ja-pos-tagger.txt, training algorithm: PERCEPTRON, evaluation score: 97.86%
model file: ja-lemmatizer.onlpm, evaluation report: ja-lemmatizer.txt, training algorithm: MAXENT, evaluation score: 97.65%

ko

language code: ko, language name: korean, training sample size: 37k

model file: ko-sentence-detector.onlpm, evaluation report: ko-sentence-detector.txt, training algorithm: MAXENT_QN, evaluation score: 95.14%
model file: ko-tokenizer.onlpm, evaluation report: ko-tokenizer.txt, training algorithm: MAXENT_QN, evaluation score: 99.40%
model file: ko-pos-tagger.onlpm, evaluation report: ko-pos-tagger.txt, training algorithm: MAXENT, evaluation score: 89.85%
model file: ko-lemmatizer.onlpm, evaluation report: ko-lemmatizer.txt, training algorithm: MAXENT, evaluation score: 87.83%

no

language code: no, language name: norwegian, training sample size: 61k

model file: no-sentence-detector.onlpm, evaluation report: no-sentence-detector.txt, training algorithm: MAXENT_QN, evaluation score: 84.07%
model file: no-tokenizer.onlpm, evaluation report: no-tokenizer.txt, training algorithm: MAXENT_QN, evaluation score: 99.93%
model file: no-pos-tagger.onlpm, evaluation report: no-pos-tagger.txt, training algorithm: MAXENT, evaluation score: 97.05%
model file: no-lemmatizer.onlpm, evaluation report: no-lemmatizer.txt, training algorithm: MAXENT, evaluation score: 94.25%

pl

language code: pl, language name: polish, training sample size: 72k

model file: pl-sentence-detector.onlpm, evaluation report: pl-sentence-detector.txt, training algorithm: MAXENT_QN, evaluation score: 95.99%
model file: pl-tokenizer.onlpm, evaluation report: pl-tokenizer.txt, training algorithm: MAXENT_QN, evaluation score: 99.86%
model file: pl-pos-tagger.onlpm, evaluation report: pl-pos-tagger.txt, training algorithm: MAXENT, evaluation score: 96.76%
model file: pl-lemmatizer.onlpm, evaluation report: pl-lemmatizer.txt, training algorithm: MAXENT, evaluation score: 91.82%

pt

language code: pt, language name: portuguese, training sample size: 113k

model file: pt-sentence-detector.onlpm, evaluation report: pt-sentence-detector.txt, training algorithm: MAXENT_QN, evaluation score: 74.60%
model file: pt-tokenizer.onlpm, evaluation report: pt-tokenizer.txt, training algorithm: MAXENT_QN, evaluation score: 99.61%
model file: pt-pos-tagger.onlpm, evaluation report: pt-pos-tagger.txt, training algorithm: MAXENT, evaluation score: 96.36%
model file: pt-lemmatizer.onlpm, evaluation report: pt-lemmatizer.txt, training algorithm: MAXENT, evaluation score: 94.25%

ru

language code: ru, language name: russian, training sample size: 186k

model file: ru-sentence-detector.onlpm, evaluation report: ru-sentence-detector.txt, training algorithm: MAXENT_QN, evaluation score: 92.73%
model file: ru-tokenizer.onlpm, evaluation report: ru-tokenizer.txt, training algorithm: MAXENT_QN, evaluation score: 99.87%
model file: ru-pos-tagger.onlpm, evaluation report: ru-pos-tagger.txt, training algorithm: MAXENT, evaluation score: 98.20%
model file: ru-lemmatizer.onlpm, evaluation report: ru-lemmatizer.txt, training algorithm: MAXENT, evaluation score: 97.11%

sv

language code: sv, language name: swedish, training sample size: 20k

model file: sv-sentence-detector.onlpm, evaluation report: sv-sentence-detector.txt, training algorithm: MAXENT, evaluation score: 86.66%
model file: sv-tokenizer.onlpm, evaluation report: sv-tokenizer.txt, training algorithm: MAXENT_QN, evaluation score: 99.97%
model file: sv-pos-tagger.onlpm, evaluation report: sv-pos-tagger.txt, training algorithm: PERCEPTRON, evaluation score: 96.37%
model file: sv-lemmatizer.onlpm, evaluation report: sv-lemmatizer.txt, training algorithm: MAXENT, evaluation score: 96.42%

uk

language code: uk, language name: ukrainian, training sample size: 17k

model file: uk-sentence-detector.onlpm, evaluation report: uk-sentence-detector.txt, training algorithm: MAXENT_QN, evaluation score: 92.87%
model file: uk-tokenizer.onlpm, evaluation report: uk-tokenizer.txt, training algorithm: MAXENT_QN, evaluation score: 99.92%
model file: uk-pos-tagger.onlpm, evaluation report: uk-pos-tagger.txt, training algorithm: MAXENT, evaluation score: 98.09%
model file: uk-lemmatizer.onlpm, evaluation report: uk-lemmatizer.txt, training algorithm: MAXENT, evaluation score: 96.97%

zh

language code: zh, language name: chinese, training sample size: 23k

model file: zh-sentence-detector.onlpm, evaluation report: zh-sentence-detector.txt, training algorithm: MAXENT, evaluation score: 97.79%
model file: zh-tokenizer.onlpm, evaluation report: zh-tokenizer.txt, training algorithm: MAXENT_QN, evaluation score: 95.77%
model file: zh-pos-tagger.onlpm, evaluation report: zh-pos-tagger.txt, training algorithm: MAXENT, evaluation score: 95.91%
model file: zh-lemmatizer.onlpm, evaluation report: zh-lemmatizer.txt, training algorithm: MAXENT, evaluation score: 99.50%