--- library_name: transformers language: - abi - acd - ade - adj - ak - akp - ann - anv - atg - avn - bas - bav - bba - beh - bem - bfd - bfo - bim - biv - bkv - blh - bmq - bmv - bom - bov - box - bqj - bss - btt - bud - bwu - cce - cjk - cko - cme - csk - cwe - cwt - dag - de - dga - dgi - dig - dop - dug - dyi - dyo - ee - efi - en - es - fal - ff - fon - fr - gej - gkn - gng - gog - gud - gur - guw - gux - gwr - hag - hay - heh - hz - ife - ig - iri - izr - jbu - jmc - kam - kbp - kdc - kdl - kdn - ken - keu - kez - kg - ki - kia - kj - kki - kkj - kma - kmb - ksb - ktj - kub - kus - kyf - las - lee - lef - lem - lg - lia - lip - ln - lob - lon - lua - luy - maw - mcp - mcu - mda - mfq - mgo - mnf - mnh - mor - mos - muh - myk - myx - mzk - mzm - mzw - ncu - nd - ndz - nfr - ng - nhu - nim - nin - nmz - nnb - nnh - nnw - nr - nso - ntm - ntr - nuj - nwb - ny - nyf - nyn - nyo - nyy - nzi - oku - old - ozm - pai - pbl - pkb - pt - rim - rn - rw - seh - sg - sig - sil - sld - sn - snw - soy - spp - ss - st - suk - sw - sxb - tbz - tem - thk - tik - tlj - tn - toh - toi - tpm - ts - tsw - tum - tw - umb - vag - ve - vmw - vun - wmw - wo - wob - xh - xog - xon - xrb - xsm - xuo - yam - yaz - yo - zu tags: - translation - opus-mt-tc-bible license: apache-2.0 model-index: - name: opus-mt-tc-bible-big-alv-deu_eng_fra_por_spa results: - task: name: Translation bem-eng type: translation args: bem-eng dataset: name: flores200-devtest type: flores200-devtest args: bem-eng metrics: - name: BLEU type: bleu value: 13.1 - name: chr-F type: chrf value: 0.37071 - task: name: Translation ibo-eng type: translation args: ibo-eng dataset: name: flores200-devtest type: flores200-devtest args: ibo-eng metrics: - name: BLEU type: bleu value: 14.6 - name: chr-F type: chrf value: 0.38994 - task: name: Translation kin-eng type: translation args: kin-eng dataset: name: flores200-devtest type: flores200-devtest args: kin-eng metrics: - name: BLEU type: bleu value: 18.1 - name: chr-F type: chrf value: 0.41964 - task: name: Translation kin-fra type: translation args: kin-fra dataset: name: flores200-devtest type: flores200-devtest args: kin-fra metrics: - name: BLEU type: bleu value: 10.7 - name: chr-F type: chrf value: 0.34887 - task: name: Translation kon-eng type: translation args: kon-eng dataset: name: flores200-devtest type: flores200-devtest args: kon-eng metrics: - name: BLEU type: bleu value: 11.3 - name: chr-F type: chrf value: 0.34262 - task: name: Translation lin-eng type: translation args: lin-eng dataset: name: flores200-devtest type: flores200-devtest args: lin-eng metrics: - name: BLEU type: bleu value: 14.2 - name: chr-F type: chrf value: 0.37728 - task: name: Translation lin-fra type: translation args: lin-fra dataset: name: flores200-devtest type: flores200-devtest args: lin-fra metrics: - name: BLEU type: bleu value: 11.0 - name: chr-F type: chrf value: 0.35052 - task: name: Translation lug-eng type: translation args: lug-eng dataset: name: flores200-devtest type: flores200-devtest args: lug-eng metrics: - name: BLEU type: bleu value: 10.6 - name: chr-F type: chrf value: 0.31805 - task: name: Translation nso-eng type: translation args: nso-eng dataset: name: flores200-devtest type: flores200-devtest args: nso-eng metrics: - name: BLEU type: bleu value: 22.3 - name: chr-F type: chrf value: 0.45662 - task: name: Translation nso-fra type: translation args: nso-fra dataset: name: flores200-devtest type: flores200-devtest args: nso-fra metrics: - name: BLEU type: bleu value: 10.5 - name: chr-F type: chrf value: 0.33732 - task: name: Translation nya-eng type: translation args: nya-eng dataset: name: flores200-devtest type: flores200-devtest args: nya-eng metrics: - name: BLEU type: bleu value: 15.8 - name: chr-F type: chrf value: 0.39887 - task: name: Translation run-eng type: translation args: run-eng dataset: name: flores200-devtest type: flores200-devtest args: run-eng metrics: - name: BLEU type: bleu value: 15.5 - name: chr-F type: chrf value: 0.39846 - task: name: Translation run-fra type: translation args: run-fra dataset: name: flores200-devtest type: flores200-devtest args: run-fra metrics: - name: BLEU type: bleu value: 10.8 - name: chr-F type: chrf value: 0.34845 - task: name: Translation run-por type: translation args: run-por dataset: name: flores200-devtest type: flores200-devtest args: run-por metrics: - name: BLEU type: bleu value: 10.2 - name: chr-F type: chrf value: 0.33836 - task: name: Translation sna-eng type: translation args: sna-eng dataset: name: flores200-devtest type: flores200-devtest args: sna-eng metrics: - name: BLEU type: bleu value: 17.2 - name: chr-F type: chrf value: 0.41974 - task: name: Translation sna-fra type: translation args: sna-fra dataset: name: flores200-devtest type: flores200-devtest args: sna-fra metrics: - name: BLEU type: bleu value: 11.3 - name: chr-F type: chrf value: 0.36443 - task: name: Translation sna-por type: translation args: sna-por dataset: name: flores200-devtest type: flores200-devtest args: sna-por metrics: - name: BLEU type: bleu value: 10.2 - name: chr-F type: chrf value: 0.34260 - task: name: Translation sot-eng type: translation args: sot-eng dataset: name: flores200-devtest type: flores200-devtest args: sot-eng metrics: - name: BLEU type: bleu value: 20.7 - name: chr-F type: chrf value: 0.45415 - task: name: Translation sot-fra type: translation args: sot-fra dataset: name: flores200-devtest type: flores200-devtest args: sot-fra metrics: - name: BLEU type: bleu value: 10.7 - name: chr-F type: chrf value: 0.34608 - task: name: Translation ssw-eng type: translation args: ssw-eng dataset: name: flores200-devtest type: flores200-devtest args: ssw-eng metrics: - name: BLEU type: bleu value: 16.1 - name: chr-F type: chrf value: 0.39768 - task: name: Translation swh-deu type: translation args: swh-deu dataset: name: flores200-devtest type: flores200-devtest args: swh-deu metrics: - name: BLEU type: bleu value: 10.9 - name: chr-F type: chrf value: 0.38892 - task: name: Translation swh-eng type: translation args: swh-eng dataset: name: flores200-devtest type: flores200-devtest args: swh-eng metrics: - name: BLEU type: bleu value: 29.1 - name: chr-F type: chrf value: 0.54048 - task: name: Translation swh-fra type: translation args: swh-fra dataset: name: flores200-devtest type: flores200-devtest args: swh-fra metrics: - name: BLEU type: bleu value: 18.2 - name: chr-F type: chrf value: 0.44837 - task: name: Translation swh-por type: translation args: swh-por dataset: name: flores200-devtest type: flores200-devtest args: swh-por metrics: - name: BLEU type: bleu value: 17.6 - name: chr-F type: chrf value: 0.44062 - task: name: Translation swh-spa type: translation args: swh-spa dataset: name: flores200-devtest type: flores200-devtest args: swh-spa metrics: - name: BLEU type: bleu value: 11.6 - name: chr-F type: chrf value: 0.38855 - task: name: Translation tsn-eng type: translation args: tsn-eng dataset: name: flores200-devtest type: flores200-devtest args: tsn-eng metrics: - name: BLEU type: bleu value: 15.3 - name: chr-F type: chrf value: 0.40410 - task: name: Translation tsn-fra type: translation args: tsn-fra dataset: name: flores200-devtest type: flores200-devtest args: tsn-fra metrics: - name: BLEU type: bleu value: 10.3 - name: chr-F type: chrf value: 0.34284 - task: name: Translation tso-eng type: translation args: tso-eng dataset: name: flores200-devtest type: flores200-devtest args: tso-eng metrics: - name: BLEU type: bleu value: 17.6 - name: chr-F type: chrf value: 0.41504 - task: name: Translation tso-fra type: translation args: tso-fra dataset: name: flores200-devtest type: flores200-devtest args: tso-fra metrics: - name: BLEU type: bleu value: 10.1 - name: chr-F type: chrf value: 0.33502 - task: name: Translation xho-eng type: translation args: xho-eng dataset: name: flores200-devtest type: flores200-devtest args: xho-eng metrics: - name: BLEU type: bleu value: 23.7 - name: chr-F type: chrf value: 0.47667 - task: name: Translation xho-fra type: translation args: xho-fra dataset: name: flores200-devtest type: flores200-devtest args: xho-fra metrics: - name: BLEU type: bleu value: 14.1 - name: chr-F type: chrf value: 0.39392 - task: name: Translation xho-por type: translation args: xho-por dataset: name: flores200-devtest type: flores200-devtest args: xho-por metrics: - name: BLEU type: bleu value: 12.3 - name: chr-F type: chrf value: 0.37032 - task: name: Translation zul-eng type: translation args: zul-eng dataset: name: flores200-devtest type: flores200-devtest args: zul-eng metrics: - name: BLEU type: bleu value: 23.4 - name: chr-F type: chrf value: 0.47798 - task: name: Translation zul-fra type: translation args: zul-fra dataset: name: flores200-devtest type: flores200-devtest args: zul-fra metrics: - name: BLEU type: bleu value: 13.9 - name: chr-F type: chrf value: 0.39504 - task: name: Translation zul-por type: translation args: zul-por dataset: name: flores200-devtest type: flores200-devtest args: zul-por metrics: - name: BLEU type: bleu value: 12.3 - name: chr-F type: chrf value: 0.36947 - task: name: Translation ibo-eng type: translation args: ibo-eng dataset: name: flores101-devtest type: flores_101 args: ibo eng devtest metrics: - name: BLEU type: bleu value: 12.5 - name: chr-F type: chrf value: 0.36320 - task: name: Translation nya-eng type: translation args: nya-eng dataset: name: flores101-devtest type: flores_101 args: nya eng devtest metrics: - name: BLEU type: bleu value: 13.2 - name: chr-F type: chrf value: 0.36765 - task: name: Translation swh-por type: translation args: swh-por dataset: name: flores101-devtest type: flores_101 args: swh por devtest metrics: - name: BLEU type: bleu value: 14.7 - name: chr-F type: chrf value: 0.40847 - task: name: Translation xho-por type: translation args: xho-por dataset: name: flores101-devtest type: flores_101 args: xho por devtest metrics: - name: BLEU type: bleu value: 10.1 - name: chr-F type: chrf value: 0.33906 - task: name: Translation zul-fra type: translation args: zul-fra dataset: name: flores101-devtest type: flores_101 args: zul fra devtest metrics: - name: BLEU type: bleu value: 11.2 - name: chr-F type: chrf value: 0.35968 - task: name: Translation bem-eng type: translation args: bem-eng dataset: name: ntrex128 type: ntrex128 args: bem-eng metrics: - name: BLEU type: bleu value: 14.9 - name: chr-F type: chrf value: 0.38068 - task: name: Translation bem-spa type: translation args: bem-spa dataset: name: ntrex128 type: ntrex128 args: bem-spa metrics: - name: BLEU type: bleu value: 10.2 - name: chr-F type: chrf value: 0.33394 - task: name: Translation ibo-eng type: translation args: ibo-eng dataset: name: ntrex128 type: ntrex128 args: ibo-eng metrics: - name: BLEU type: bleu value: 17.4 - name: chr-F type: chrf value: 0.42002 - task: name: Translation ibo-fra type: translation args: ibo-fra dataset: name: ntrex128 type: ntrex128 args: ibo-fra metrics: - name: BLEU type: bleu value: 10.3 - name: chr-F type: chrf value: 0.34993 - task: name: Translation ibo-spa type: translation args: ibo-spa dataset: name: ntrex128 type: ntrex128 args: ibo-spa metrics: - name: BLEU type: bleu value: 11.7 - name: chr-F type: chrf value: 0.36108 - task: name: Translation kin-eng type: translation args: kin-eng dataset: name: ntrex128 type: ntrex128 args: kin-eng metrics: - name: BLEU type: bleu value: 16.9 - name: chr-F type: chrf value: 0.42892 - task: name: Translation kin-fra type: translation args: kin-fra dataset: name: ntrex128 type: ntrex128 args: kin-fra metrics: - name: BLEU type: bleu value: 10.7 - name: chr-F type: chrf value: 0.35842 - task: name: Translation kin-por type: translation args: kin-por dataset: name: ntrex128 type: ntrex128 args: kin-por metrics: - name: BLEU type: bleu value: 10.0 - name: chr-F type: chrf value: 0.34399 - task: name: Translation kin-spa type: translation args: kin-spa dataset: name: ntrex128 type: ntrex128 args: kin-spa metrics: - name: BLEU type: bleu value: 12.5 - name: chr-F type: chrf value: 0.37224 - task: name: Translation nde-eng type: translation args: nde-eng dataset: name: ntrex128 type: ntrex128 args: nde-eng metrics: - name: BLEU type: bleu value: 13.8 - name: chr-F type: chrf value: 0.39640 - task: name: Translation nde-spa type: translation args: nde-spa dataset: name: ntrex128 type: ntrex128 args: nde-spa metrics: - name: BLEU type: bleu value: 10.0 - name: chr-F type: chrf value: 0.34638 - task: name: Translation nso-eng type: translation args: nso-eng dataset: name: ntrex128 type: ntrex128 args: nso-eng metrics: - name: BLEU type: bleu value: 17.0 - name: chr-F type: chrf value: 0.42278 - task: name: Translation nso-spa type: translation args: nso-spa dataset: name: ntrex128 type: ntrex128 args: nso-spa metrics: - name: BLEU type: bleu value: 11.3 - name: chr-F type: chrf value: 0.35027 - task: name: Translation nya-eng type: translation args: nya-eng dataset: name: ntrex128 type: ntrex128 args: nya-eng metrics: - name: BLEU type: bleu value: 19.2 - name: chr-F type: chrf value: 0.42702 - task: name: Translation nya-fra type: translation args: nya-fra dataset: name: ntrex128 type: ntrex128 args: nya-fra metrics: - name: BLEU type: bleu value: 10.7 - name: chr-F type: chrf value: 0.35503 - task: name: Translation nya-por type: translation args: nya-por dataset: name: ntrex128 type: ntrex128 args: nya-por metrics: - name: BLEU type: bleu value: 10.4 - name: chr-F type: chrf value: 0.33912 - task: name: Translation nya-spa type: translation args: nya-spa dataset: name: ntrex128 type: ntrex128 args: nya-spa metrics: - name: BLEU type: bleu value: 12.5 - name: chr-F type: chrf value: 0.36355 - task: name: Translation ssw-eng type: translation args: ssw-eng dataset: name: ntrex128 type: ntrex128 args: ssw-eng metrics: - name: BLEU type: bleu value: 18.0 - name: chr-F type: chrf value: 0.43041 - task: name: Translation ssw-spa type: translation args: ssw-spa dataset: name: ntrex128 type: ntrex128 args: ssw-spa metrics: - name: BLEU type: bleu value: 11.4 - name: chr-F type: chrf value: 0.35392 - task: name: Translation swa-deu type: translation args: swa-deu dataset: name: ntrex128 type: ntrex128 args: swa-deu metrics: - name: BLEU type: bleu value: 11.2 - name: chr-F type: chrf value: 0.39475 - task: name: Translation swa-eng type: translation args: swa-eng dataset: name: ntrex128 type: ntrex128 args: swa-eng metrics: - name: BLEU type: bleu value: 30.4 - name: chr-F type: chrf value: 0.54492 - task: name: Translation swa-fra type: translation args: swa-fra dataset: name: ntrex128 type: ntrex128 args: swa-fra metrics: - name: BLEU type: bleu value: 15.6 - name: chr-F type: chrf value: 0.43008 - task: name: Translation swa-por type: translation args: swa-por dataset: name: ntrex128 type: ntrex128 args: swa-por metrics: - name: BLEU type: bleu value: 15.4 - name: chr-F type: chrf value: 0.42343 - task: name: Translation swa-spa type: translation args: swa-spa dataset: name: ntrex128 type: ntrex128 args: swa-spa metrics: - name: BLEU type: bleu value: 18.9 - name: chr-F type: chrf value: 0.44892 - task: name: Translation tsn-eng type: translation args: tsn-eng dataset: name: ntrex128 type: ntrex128 args: tsn-eng metrics: - name: BLEU type: bleu value: 20.1 - name: chr-F type: chrf value: 0.44944 - task: name: Translation tsn-fra type: translation args: tsn-fra dataset: name: ntrex128 type: ntrex128 args: tsn-fra metrics: - name: BLEU type: bleu value: 11.5 - name: chr-F type: chrf value: 0.36584 - task: name: Translation tsn-por type: translation args: tsn-por dataset: name: ntrex128 type: ntrex128 args: tsn-por metrics: - name: BLEU type: bleu value: 11.1 - name: chr-F type: chrf value: 0.35774 - task: name: Translation tsn-spa type: translation args: tsn-spa dataset: name: ntrex128 type: ntrex128 args: tsn-spa metrics: - name: BLEU type: bleu value: 13.8 - name: chr-F type: chrf value: 0.38149 - task: name: Translation ven-eng type: translation args: ven-eng dataset: name: ntrex128 type: ntrex128 args: ven-eng metrics: - name: BLEU type: bleu value: 12.7 - name: chr-F type: chrf value: 0.36848 - task: name: Translation xho-eng type: translation args: xho-eng dataset: name: ntrex128 type: ntrex128 args: xho-eng metrics: - name: BLEU type: bleu value: 21.8 - name: chr-F type: chrf value: 0.46636 - task: name: Translation xho-fra type: translation args: xho-fra dataset: name: ntrex128 type: ntrex128 args: xho-fra metrics: - name: BLEU type: bleu value: 11.6 - name: chr-F type: chrf value: 0.36905 - task: name: Translation xho-por type: translation args: xho-por dataset: name: ntrex128 type: ntrex128 args: xho-por metrics: - name: BLEU type: bleu value: 10.6 - name: chr-F type: chrf value: 0.35687 - task: name: Translation xho-spa type: translation args: xho-spa dataset: name: ntrex128 type: ntrex128 args: xho-spa metrics: - name: BLEU type: bleu value: 13.3 - name: chr-F type: chrf value: 0.38176 - task: name: Translation yor-eng type: translation args: yor-eng dataset: name: ntrex128 type: ntrex128 args: yor-eng metrics: - name: BLEU type: bleu value: 11.2 - name: chr-F type: chrf value: 0.34615 - task: name: Translation zul-eng type: translation args: zul-eng dataset: name: ntrex128 type: ntrex128 args: zul-eng metrics: - name: BLEU type: bleu value: 21.9 - name: chr-F type: chrf value: 0.45848 - task: name: Translation zul-fra type: translation args: zul-fra dataset: name: ntrex128 type: ntrex128 args: zul-fra metrics: - name: BLEU type: bleu value: 11.4 - name: chr-F type: chrf value: 0.36203 - task: name: Translation zul-por type: translation args: zul-por dataset: name: ntrex128 type: ntrex128 args: zul-por metrics: - name: BLEU type: bleu value: 10.8 - name: chr-F type: chrf value: 0.35081 - task: name: Translation zul-spa type: translation args: zul-spa dataset: name: ntrex128 type: ntrex128 args: zul-spa metrics: - name: BLEU type: bleu value: 13.0 - name: chr-F type: chrf value: 0.37270 - task: name: Translation multi-multi type: translation args: multi-multi dataset: name: tatoeba-test-v2020-07-28-v2023-09-26 type: tatoeba_mt args: multi-multi metrics: - name: BLEU type: bleu value: 27.3 - name: chr-F type: chrf value: 0.44751 - task: name: Translation run-eng type: translation args: run-eng dataset: name: tatoeba-test-v2021-08-07 type: tatoeba_mt args: run-eng metrics: - name: BLEU type: bleu value: 34.9 - name: chr-F type: chrf value: 0.49949 - task: name: Translation run-fra type: translation args: run-fra dataset: name: tatoeba-test-v2021-08-07 type: tatoeba_mt args: run-fra metrics: - name: BLEU type: bleu value: 22.4 - name: chr-F type: chrf value: 0.41431 - task: name: Translation swa-eng type: translation args: swa-eng dataset: name: tatoeba-test-v2021-08-07 type: tatoeba_mt args: swa-eng metrics: - name: BLEU type: bleu value: 41.5 - name: chr-F type: chrf value: 0.57031 - task: name: Translation xho-eng type: translation args: xho-eng dataset: name: tatoeba-test-v2021-03-30 type: tatoeba_mt args: xho-eng metrics: - name: BLEU type: bleu value: 36.4 - name: chr-F type: chrf value: 0.51250 - task: name: Translation zul-eng type: translation args: zul-eng dataset: name: tico19-test type: tico19-test args: zul-eng metrics: - name: BLEU type: bleu value: 25.2 - name: chr-F type: chrf value: 0.48762 - task: name: Translation zul-fra type: translation args: zul-fra dataset: name: tico19-test type: tico19-test args: zul-fra metrics: - name: BLEU type: bleu value: 13.7 - name: chr-F type: chrf value: 0.37823 - task: name: Translation zul-por type: translation args: zul-por dataset: name: tico19-test type: tico19-test args: zul-por metrics: - name: BLEU type: bleu value: 14.0 - name: chr-F type: chrf value: 0.38586 - task: name: Translation zul-spa type: translation args: zul-spa dataset: name: tico19-test type: tico19-test args: zul-spa metrics: - name: BLEU type: bleu value: 15.9 - name: chr-F type: chrf value: 0.40041 --- # opus-mt-tc-bible-big-alv-deu_eng_fra_por_spa ## Table of Contents - [Model Details](#model-details) - [Uses](#uses) - [Risks, Limitations and Biases](#risks-limitations-and-biases) - [How to Get Started With the Model](#how-to-get-started-with-the-model) - [Training](#training) - [Evaluation](#evaluation) - [Citation Information](#citation-information) - [Acknowledgements](#acknowledgements) ## Model Details Neural machine translation model for translating from Atlantic-Congo languages (alv) to unknown (deu+eng+fra+por+spa). This model is part of the [OPUS-MT project](https://github.com/Helsinki-NLP/Opus-MT), an effort to make neural machine translation models widely available and accessible for many languages in the world. All models are originally trained using the amazing framework of [Marian NMT](https://marian-nmt.github.io/), an efficient NMT implementation written in pure C++. The models have been converted to pyTorch using the transformers library by huggingface. Training data is taken from [OPUS](https://opus.nlpl.eu/) and training pipelines use the procedures of [OPUS-MT-train](https://github.com/Helsinki-NLP/Opus-MT-train). **Model Description:** - **Developed by:** Language Technology Research Group at the University of Helsinki - **Model Type:** Translation (transformer-big) - **Release**: 2024-05-30 - **License:** Apache-2.0 - **Language(s):** - Source Language(s): abi acd ade adj aka akp ann anv atg avn bas bav bba beh bem bfd bfo bim biv bkv blh bmq bmv bom bov box bqj bss btt bud bwu cce cjk cko cme csk cwe cwt dag dga dgi dig dop dug dyi dyo efi ewe fal fon fuc ful gej gkn gng gog gud gur guw gux gwr hag hay heh her ibo ife iri izr jbu jmc kam kbp kdc kdl kdn ken keu kez kia kik kin kki kkj kma kmb kon ksb ktj kua kub kus kyf las lee lef lem lia lin lip lob lon lua lug luy maw mcp mcu mda mfq mgo mnf mnh mor mos muh myk myx mzk mzm mzw nbl ncu nde ndo ndz nfr nhu nim nin nmz nnb nnh nnw nso ntm ntr nuj nwb nya nyf nyn nyo nyy nzi oku old ozm pai pbl pkb rim run sag seh sig sil sld sna snw sot soy spp ssw suk swa swc swh sxb tbz tem thk tik tlj toh toi tpm tsn tso tsw tum twi umb vag ven vmw vun wmw wob wol xho xog xon xrb xsm xuo yam yaz yor zul - Target Language(s): deu eng fra por spa - Valid Target Language Labels: >>deu<< >>eng<< >>fra<< >>por<< >>spa<< >>xxx<< - **Original Model**: [opusTCv20230926max50+bt+jhubc_transformer-big_2024-05-30.zip](https://object.pouta.csc.fi/Tatoeba-MT-models/alv-deu+eng+fra+por+spa/opusTCv20230926max50+bt+jhubc_transformer-big_2024-05-30.zip) - **Resources for more information:** - [OPUS-MT dashboard](https://opus.nlpl.eu/dashboard/index.php?pkg=opusmt&test=all&scoreslang=all&chart=standard&model=Tatoeba-MT-models/alv-deu%2Beng%2Bfra%2Bpor%2Bspa/opusTCv20230926max50%2Bbt%2Bjhubc_transformer-big_2024-05-30) - [OPUS-MT-train GitHub Repo](https://github.com/Helsinki-NLP/OPUS-MT-train) - [More information about MarianNMT models in the transformers library](https://huggingface.co./docs/transformers/model_doc/marian) - [Tatoeba Translation Challenge](https://github.com/Helsinki-NLP/Tatoeba-Challenge/) - [HPLT bilingual data v1 (as part of the Tatoeba Translation Challenge dataset)](https://hplt-project.org/datasets/v1) - [A massively parallel Bible corpus](https://aclanthology.org/L14-1215/) This is a multilingual translation model with multiple target languages. A sentence initial language token is required in the form of `>>id<<` (id = valid target language ID), e.g. `>>deu<<` ## Uses This model can be used for translation and text-to-text generation. ## Risks, Limitations and Biases **CONTENT WARNING: Readers should be aware that the model is trained on various public data sets that may contain content that is disturbing, offensive, and can propagate historical and current stereotypes.** Significant research has explored bias and fairness issues with language models (see, e.g., [Sheng et al. (2021)](https://aclanthology.org/2021.acl-long.330.pdf) and [Bender et al. (2021)](https://dl.acm.org/doi/pdf/10.1145/3442188.3445922)). ## How to Get Started With the Model A short example code: ```python from transformers import MarianMTModel, MarianTokenizer src_text = [ ">>deu<< Replace this with text in an accepted source language.", ">>spa<< This is the second sentence." ] model_name = "pytorch-models/opus-mt-tc-bible-big-alv-deu_eng_fra_por_spa" tokenizer = MarianTokenizer.from_pretrained(model_name) model = MarianMTModel.from_pretrained(model_name) translated = model.generate(**tokenizer(src_text, return_tensors="pt", padding=True)) for t in translated: print( tokenizer.decode(t, skip_special_tokens=True) ) ``` You can also use OPUS-MT models with the transformers pipelines, for example: ```python from transformers import pipeline pipe = pipeline("translation", model="Helsinki-NLP/opus-mt-tc-bible-big-alv-deu_eng_fra_por_spa") print(pipe(">>deu<< Replace this with text in an accepted source language.")) ``` ## Training - **Data**: opusTCv20230926max50+bt+jhubc ([source](https://github.com/Helsinki-NLP/Tatoeba-Challenge)) - **Pre-processing**: SentencePiece (spm32k,spm32k) - **Model Type:** transformer-big - **Original MarianNMT Model**: [opusTCv20230926max50+bt+jhubc_transformer-big_2024-05-30.zip](https://object.pouta.csc.fi/Tatoeba-MT-models/alv-deu+eng+fra+por+spa/opusTCv20230926max50+bt+jhubc_transformer-big_2024-05-30.zip) - **Training Scripts**: [GitHub Repo](https://github.com/Helsinki-NLP/OPUS-MT-train) ## Evaluation * [Model scores at the OPUS-MT dashboard](https://opus.nlpl.eu/dashboard/index.php?pkg=opusmt&test=all&scoreslang=all&chart=standard&model=Tatoeba-MT-models/alv-deu%2Beng%2Bfra%2Bpor%2Bspa/opusTCv20230926max50%2Bbt%2Bjhubc_transformer-big_2024-05-30) * test set translations: [opusTCv20230926max50+bt+jhubc_transformer-big_2024-05-29.test.txt](https://object.pouta.csc.fi/Tatoeba-MT-models/alv-deu+eng+fra+por+spa/opusTCv20230926max50+bt+jhubc_transformer-big_2024-05-29.test.txt) * test set scores: [opusTCv20230926max50+bt+jhubc_transformer-big_2024-05-29.eval.txt](https://object.pouta.csc.fi/Tatoeba-MT-models/alv-deu+eng+fra+por+spa/opusTCv20230926max50+bt+jhubc_transformer-big_2024-05-29.eval.txt) * benchmark results: [benchmark_results.txt](benchmark_results.txt) * benchmark output: [benchmark_translations.zip](benchmark_translations.zip) | langpair | testset | chr-F | BLEU | #sent | #words | |----------|---------|-------|-------|-------|--------| | run-eng | tatoeba-test-v2021-08-07 | 0.49949 | 34.9 | 1703 | 10041 | | run-fra | tatoeba-test-v2021-08-07 | 0.41431 | 22.4 | 1274 | 7479 | | swa-eng | tatoeba-test-v2021-08-07 | 0.57031 | 41.5 | 387 | 2508 | | swh-por | flores101-devtest | 0.40847 | 14.7 | 1012 | 26519 | | kin-eng | flores200-devtest | 0.41964 | 18.1 | 1012 | 24721 | | nso-eng | flores200-devtest | 0.45662 | 22.3 | 1012 | 24721 | | sna-eng | flores200-devtest | 0.41974 | 17.2 | 1012 | 24721 | | sot-eng | flores200-devtest | 0.45415 | 20.7 | 1012 | 24721 | | swh-eng | flores200-devtest | 0.54048 | 29.1 | 1012 | 24721 | | swh-fra | flores200-devtest | 0.44837 | 18.2 | 1012 | 28343 | | swh-por | flores200-devtest | 0.44062 | 17.6 | 1012 | 26519 | | tsn-eng | flores200-devtest | 0.40410 | 15.3 | 1012 | 24721 | | tso-eng | flores200-devtest | 0.41504 | 17.6 | 1012 | 24721 | | xho-eng | flores200-devtest | 0.47667 | 23.7 | 1012 | 24721 | | zul-eng | flores200-devtest | 0.47798 | 23.4 | 1012 | 24721 | | ibo-eng | ntrex128 | 0.42002 | 17.4 | 1997 | 47673 | | kin-eng | ntrex128 | 0.42892 | 16.9 | 1997 | 47673 | | nso-eng | ntrex128 | 0.42278 | 17.0 | 1997 | 47673 | | nya-eng | ntrex128 | 0.42702 | 19.2 | 1997 | 47673 | | ssw-eng | ntrex128 | 0.43041 | 18.0 | 1997 | 47673 | | swa-eng | ntrex128 | 0.54492 | 30.4 | 1997 | 47673 | | swa-fra | ntrex128 | 0.43008 | 15.6 | 1997 | 53481 | | swa-por | ntrex128 | 0.42343 | 15.4 | 1997 | 51631 | | swa-spa | ntrex128 | 0.44892 | 18.9 | 1997 | 54107 | | tsn-eng | ntrex128 | 0.44944 | 20.1 | 1997 | 47673 | | xho-eng | ntrex128 | 0.46636 | 21.8 | 1997 | 47673 | | zul-eng | ntrex128 | 0.45848 | 21.9 | 1997 | 47673 | | zul-eng | tico19-test | 0.48762 | 25.2 | 2100 | 56804 | | zul-spa | tico19-test | 0.40041 | 15.9 | 2100 | 66563 | ## Citation Information * Publications: [Democratizing neural machine translation with OPUS-MT](https://doi.org/10.1007/s10579-023-09704-w) and [OPUS-MT – Building open translation services for the World](https://aclanthology.org/2020.eamt-1.61/) and [The Tatoeba Translation Challenge – Realistic Data Sets for Low Resource and Multilingual MT](https://aclanthology.org/2020.wmt-1.139/) (Please, cite if you use this model.) ```bibtex @article{tiedemann2023democratizing, title={Democratizing neural machine translation with {OPUS-MT}}, author={Tiedemann, J{\"o}rg and Aulamo, Mikko and Bakshandaeva, Daria and Boggia, Michele and Gr{\"o}nroos, Stig-Arne and Nieminen, Tommi and Raganato, Alessandro and Scherrer, Yves and Vazquez, Raul and Virpioja, Sami}, journal={Language Resources and Evaluation}, number={58}, pages={713--755}, year={2023}, publisher={Springer Nature}, issn={1574-0218}, doi={10.1007/s10579-023-09704-w} } @inproceedings{tiedemann-thottingal-2020-opus, title = "{OPUS}-{MT} {--} Building open translation services for the World", author = {Tiedemann, J{\"o}rg and Thottingal, Santhosh}, booktitle = "Proceedings of the 22nd Annual Conference of the European Association for Machine Translation", month = nov, year = "2020", address = "Lisboa, Portugal", publisher = "European Association for Machine Translation", url = "https://aclanthology.org/2020.eamt-1.61", pages = "479--480", } @inproceedings{tiedemann-2020-tatoeba, title = "The Tatoeba Translation Challenge {--} Realistic Data Sets for Low Resource and Multilingual {MT}", author = {Tiedemann, J{\"o}rg}, booktitle = "Proceedings of the Fifth Conference on Machine Translation", month = nov, year = "2020", address = "Online", publisher = "Association for Computational Linguistics", url = "https://aclanthology.org/2020.wmt-1.139", pages = "1174--1182", } ``` ## Acknowledgements The work is supported by the [HPLT project](https://hplt-project.org/), funded by the European Union’s Horizon Europe research and innovation programme under grant agreement No 101070350. We are also grateful for the generous computational resources and IT infrastructure provided by [CSC -- IT Center for Science](https://www.csc.fi/), Finland, and the [EuroHPC supercomputer LUMI](https://www.lumi-supercomputer.eu/). ## Model conversion info * transformers version: 4.45.1 * OPUS-MT git hash: a0ea3b3 * port time: Mon Oct 7 17:13:22 EEST 2024 * port machine: LM0-400-22516.local