Commit Graph

218 Commits

Author SHA1 Message Date
Joerg Tiedemann
4b0d49ddbb small fixes to system specific environments 2021-11-04 22:30:41 +02:00
Joerg Tiedemann
0e7b3e173a Merge branch 'master' of github.com:Helsinki-NLP/OPUS-MT-train 2021-11-04 10:52:52 +02:00
Joerg Tiedemann
3453256b3e add fbgemm 2021-11-04 10:52:29 +02:00
Joerg Tiedemann
08ed366914 allas storage commands 2021-11-04 09:57:48 +02:00
Joerg Tiedemann
52f0bea9a9 cleanup environment definitions 2021-11-03 16:14:29 +02:00
Joerg Tiedemann
dbcdab4d6b Merge branch 'master' of github.com:Helsinki-NLP/OPUS-MT-train 2021-11-02 18:48:19 +02:00
Joerg Tiedemann
c5cd41ab4d simplified vocab recipes 2021-11-02 18:47:24 +02:00
Joerg Tiedemann
71174062d6 fix vocab yaml script added 2021-11-02 18:38:28 +02:00
Joerg Tiedemann
ae9637b09c avoid setting TATOEBA_DATASET recursively 2021-10-10 20:20:31 +03:00
Joerg Tiedemann
07cae1b0a5 pivotlang option added to tatoeba langgroup models and removed raw-langcodes in dist packages 2021-10-10 00:57:03 +03:00
Joerg Tiedemann
f397e16097 merged from puhti and mahti 2021-10-06 19:37:48 +03:00
Joerg Tiedemann
2289fe2f1c latest changes on puhti 2021-10-06 19:34:29 +03:00
Joerg Tiedemann
68b935c773 setup for mahti 2021-10-06 09:57:17 +03:00
Joerg Tiedemann
f4ffae653b create valid yaml files from vocab 2021-10-05 17:43:46 +03:00
Joerg Tiedemann
378eff0710 Merge branch 'master' of github.com:Helsinki-NLP/OPUS-MT-train 2021-10-05 14:44:04 +03:00
Joerg Tiedemann
4d2e4a0c23 bt tatoeba on mahti 2021-10-05 14:43:52 +03:00
tiedemann
b7e378e6ad
Update README.md 2021-10-05 09:33:17 +03:00
Joerg Tiedemann
6db5b3b716 fixed a problem with langlabel files 2021-09-13 00:07:51 +03:00
Joerg Tiedemann
72e1bcb7ec fixed multithreading issues with data recipe 2021-08-09 22:19:05 +03:00
Joerg Tiedemann
fc8c2b33c0 Merge branch 'master' of github.com:Helsinki-NLP/OPUS-MT-train 2021-08-09 14:09:46 +03:00
Joerg Tiedemann
3da042c3cf latest changes to makefiles 2021-08-09 14:06:10 +03:00
tiedemann
4aa9459471
Merge pull request #57 from raphaelmerx/master
Marian compilation: max 8 jobs to avoid memory error
2021-07-18 11:06:24 +03:00
Raphael Merx
79fa4c8fb5 Marian compilation: max 8 jobs 2021-06-15 14:54:52 +08:00
Joerg Tiedemann
200662863e added recipes for tatoeba models other than English 2021-05-04 08:49:16 +03:00
Joerg Tiedemann
f84944faa9 updated to latest wiki release and individual languages 2021-04-12 22:12:49 +03:00
Joerg Tiedemann
fa746419ab updated to latest wiki release and individual languages 2021-04-12 22:09:09 +03:00
Joerg Tiedemann
cde8f0d0af balance dev data in multiligual models and a bug fixed in preprocess script 2021-03-30 00:00:28 +03:00
Joerg Tiedemann
3cd0bd3f75 create vocabulary files from spm models) 2021-03-14 22:05:21 +02:00
Joerg Tiedemann
8170bced38 added recipe for refreshing release info 2021-03-13 00:34:27 +02:00
Joerg Tiedemann
bb39c060c0 added recipe for refreshing release info 2021-03-13 00:29:23 +02:00
Joerg Tiedemann
07f021e0bb added recipes for storing and fetching working data 2021-03-06 15:19:17 +02:00
Joerg Tiedemann
2067577021 adjustments for mahti and tatoeba back translations 2021-03-02 15:39:47 +02:00
Joerg Tiedemann
e4f76608d3 adjustments for mahti and tatoeba back translations 2021-03-02 09:53:47 +02:00
Joerg Tiedemann
1f77b44651 adjust to mahti 2021-02-25 21:19:08 +02:00
Joerg Tiedemann
6537fdea13 backtranslation for Tatoeba data 2021-02-25 17:17:21 +02:00
Joerg Tiedemann
f81a2ad638 fixed language label problem in tatoeba model training recipes 2021-02-20 01:09:43 +02:00
Joerg Tiedemann
098509d257 fixed language label problem in tatoeba model training recipes 2021-02-19 23:56:21 +02:00
Joerg Tiedemann
d880d8b917 memad project recipes added 2021-02-19 19:41:25 +02:00
Joerg Tiedemann
ca2a249845 bugfixing in tatoeba MT model recipes 2021-02-18 13:49:16 +02:00
Joerg Tiedemann
53c5680268 fixed tatoeba group recipes 2021-02-16 20:36:00 +02:00
Joerg Tiedemann
f2eb32239b memad tatoeba models 2021-02-15 20:36:26 +02:00
Joerg Tiedemann
3c6793045b moved results table generation for tatoeba models 2021-02-15 20:35:29 +02:00
Joerg Tiedemann
c46eb49c26 memad models with tatoeba data and some cleanup in tatoeba langgroup language expansion 2021-02-03 20:49:14 +02:00
Joerg Tiedemann
4305ad01b9 tatoeba model result lists added 2021-01-18 14:53:54 +02:00
Joerg Tiedemann
b5fbc4a52a Merge branch 'master' of github.com:Helsinki-NLP/OPUS-MT-train 2021-01-14 23:12:46 +02:00
Joerg Tiedemann
385e8298b2 more fixes with evaluation recipes of multilingual tatoeba models 2021-01-14 23:07:12 +02:00
tiedemann
f74531468e
Update README.md 2021-01-14 22:31:09 +02:00
Joerg Tiedemann
81ce0bf8c4 fixed a problem with fine-tuning Tatoeba multilingual models for specific language pairs 2021-01-09 00:29:04 +02:00
Joerg Tiedemann
71d49406eb tutorial links added 2021-01-07 23:50:33 +02:00
Joerg Tiedemann
527ab54caa subset result tables for Tatoeba now also with reverse translation direction 2021-01-07 23:19:24 +02:00