Ars-Linguistica / mlconjug3

A Python library to conjugate verbs in French, English, Spanish, Italian, Portuguese and Romanian (more soon) using Machine Learning techniques.
MIT License
71 stars 10 forks source link

Addition of Dutch #331

Closed JeroenSteen closed 1 year ago

JeroenSteen commented 1 year ago

Is it possible to add Dutch as a language to conjugate verbs aswell with ML? What kind of sources are needed to add Dutch to mlconjug3? What should the structure be like of the training data, and is JSON or XML prefered? Here some tense names:

"Voltooid deelwoord",
"Onvoltooid tegenwoordige tijd (ott)",
"Tegenwoordige tijd, bijzinsvolgorde",
"Voltooid tegenwoordige tijd (vtt)",
"Onvoltooid verleden tijd (ovt)",
"Verleden tijd, bijzinsvolgorde",
"Voltooid verleden tijd (vvt)",
"Onvoltooid tegenwoordige toekomende tijd (ottt)",
"Voltooid tegenwoordige toekomende tijd (vttt)",
"Onvoltooid verleden toekomende tijd (ovtt)",
"Voltooid verleden toekomende tijd (vvtt)",
"Gebiedende wijs",
"Aanvoegende wijs"

"Past participle",
"present simple tense (ott)",
"present tense, subordinate clause order",
"present perfect tense (vtt)",
"past simple (past)",
"past tense, subordinate order",
"past perfect (vvt)",
"present future perfect tense (ottt)",
"present future perfect (vttt)",
"simple past future tense (novtt)",
"past future perfect (vvtt)",
"Imperative",
"Subjunctive"

Maybe you need these keys:

infinitive
past_participle

present_simple_tense_ott_1s
present_simple_tense_ott_2s
present_simple_tense_ott_3s
present_simple_tense_ott_1p
present_simple_tense_ott_2p
present_simple_tense_ott_3p

present_tense_subordinate_clause_order_1s
present_tense_subordinate_clause_order_2s
present_tense_subordinate_clause_order_3s
present_tense_subordinate_clause_order_1p
present_tense_subordinate_clause_order_2p
present_tense_subordinate_clause_order_3p

present_perfect_tense_vtt_1s
present_perfect_tense_vtt_2s
present_perfect_tense_vtt_3s
present_perfect_tense_vtt_1p
present_perfect_tense_vtt_2p
present_perfect_tense_vtt_3p

past_simple_ovt_1s
past_simple_ovt_2s
past_simple_ovt_3s
past_simple_ovt_1p
past_simple_ovt_2p
past_simple_ovt_3p

past_tense_subordinate_order_1s
past_tense_subordinate_order_2s
past_tense_subordinate_order_3s
past_tense_subordinate_order_1p
past_tense_subordinate_order_2p
past_tense_subordinate_order_3p

past_perfect_vvt_1s
past_perfect_vvt_2s
past_perfect_vvt_3s
past_perfect_vvt_1p
past_perfect_vvt_2p
past_perfect_vvt_3p

present_future_perfect_tense_ottt_1s
present_future_perfect_tense_ottt_2s
present_future_perfect_tense_ottt_3s
present_future_perfect_tense_ottt_1p
present_future_perfect_tense_ottt_2p
present_future_perfect_tense_ottt_3p

present_future_perfect_vttt_1s
present_future_perfect_vttt_2s
present_future_perfect_vttt_3s
present_future_perfect_vttt_1p
present_future_perfect_vttt_2p
present_future_perfect_vttt_3p

simple_past_future_tense_ovtt_1s
simple_past_future_tense_ovtt_2s
simple_past_future_tense_ovtt_3s
simple_past_future_tense_ovtt_1p
simple_past_future_tense_ovtt_2p
simple_past_future_tense_ovtt_3p

past_future_perfect_vvtt_1s
past_future_perfect_vvtt_2s
past_future_perfect_vvtt_3s
past_future_perfect_vvtt_1p
past_future_perfect_vvtt_2p
past_future_perfect_vvtt_3p

imperative
subjunctive

Also I saw this: "More information on Verbiste at https://perso.b2b2c.ca/~sarrazip/dev/conjug_manager.html", but that page is not found.

SekouDiaoNlp commented 1 year ago

Hi @JeroenSteen , mlconjug3 version 4.0 will be released later this month with support for German and Dutch. I will release first version 3.9 which makes it easier for users to train their own models and add new languages.

pdzong commented 1 year ago

Hi @SekouDiaoNlp. Is support for the German language still to be expected soon?

addisonklinke commented 1 week ago

@SekouDiaoNlp I'm also interested in German and see v4.0 hasn't been released yet. I work professionally as an ML engineer, so happy to help you prepare a dataset and/or run the training if needed!