@@ -183,7 +183,30 @@ def __getattribute__(self, name: str) -> str:
183183        indices = ["А" , "Б" , "В" , "Г" , "Д" , "Е" ],
184184    ),
185185    Language .BEMBA : TranslationLiterals (language = Language .BEMBA ),
186-     Language .BENGALI : TranslationLiterals (language = Language .BENGALI , question_word = "প্রশ্ন" ),
186+     Language .BENGALI : TranslationLiterals (
187+         language = Language .BENGALI ,
188+         question_word = "প্রশ্ন" ,
189+         answer = "উত্তর" ,
190+         confirmation_word = "তাই না" ,
191+         yes = "হ্যাঁ" ,
192+         no = "না" ,
193+         also = "সাথে" ,
194+         cause_word = "কারণ" ,
195+         effect_word = "অতএব" ,
196+         true = "সত্য" ,
197+         false = "মিথ্যা" ,
198+         neither = "ন তাই, ন তাই না" ,
199+         full_stop = "।" ,
200+         comma = "," ,
201+         question_mark = "?" ,
202+         exclamation_mark = "!" ,
203+         word_space = " " ,
204+         sentence_space = " " ,
205+         colon = ":" ,
206+         indices = ["ক" , "খ" , "গ" , "ঘ" , "ঙ" , "চ" ],
207+         or_word = "বা" ,
208+         and_word = "এবং" ,
209+     ),
187210    Language .BHOJPURI : TranslationLiterals (language = Language .BHOJPURI ),
188211    Language .BIHARI : TranslationLiterals (language = Language .BIHARI ),  # Deprecated 
189212    Language .BOSNIAN : TranslationLiterals (language = Language .BOSNIAN ),
@@ -544,7 +567,30 @@ def __getattribute__(self, name: str) -> str:
544567        semicolon = "·" ,
545568    ),
546569    Language .GUARANI : TranslationLiterals (language = Language .GUARANI ),
547-     Language .GUJARATI : TranslationLiterals (language = Language .GUJARATI ),
570+     Language .GUJARATI : TranslationLiterals (
571+         language = Language .GUJARATI ,
572+         question_word = "પ્રશ્ન" ,
573+         answer = "જવાબ" ,
574+         confirmation_word = "ખરું ને" ,
575+         yes = "હા" ,
576+         no = "ના" ,
577+         also = "અને" ,
578+         cause_word = "કારણ કે" ,
579+         effect_word = "તેથી" ,
580+         true = "સાચું" ,
581+         false = "ખોટું" ,
582+         neither = "ન તો આ, ન તે" ,
583+         or_word = "અથવા" ,
584+         and_word = "અને" ,
585+         full_stop = "।" ,
586+         comma = "," ,
587+         question_mark = "?" ,
588+         exclamation_mark = "!" ,
589+         word_space = " " ,
590+         sentence_space = " " ,
591+         colon = ":" ,
592+         indices = ["અ" , "આ" , "ઇ" , "ઈ" , "ઉ" , "ઊ" ],
593+     ),
548594    Language .HAITIAN : TranslationLiterals (
549595        # From https://github.com/EleutherAI/lm-evaluation-harness/blob/0845b588303f1f59af98dd1c5bdbd78a9e75a1e2/lm_eval/tasks/xcopa/utils.py 
550596        language = Language .HAITIAN ,
@@ -859,7 +905,29 @@ def __getattribute__(self, name: str) -> str:
859905        colon = ":" ,
860906        semicolon = ";" ,
861907    ),
862-     Language .PUNJABI : TranslationLiterals (language = Language .PUNJABI ),
908+     Language .PUNJABI : TranslationLiterals (
909+         language = Language .PUNJABI ,
910+         question_word = "ਸਵਾਲ" ,
911+         answer = "ਜਵਾਬ" ,
912+         confirmation_word = "ਹਾਂ ਨਾ" ,
913+         yes = "ਹਾਂ" ,
914+         no = "ਨਹੀਂ" ,
915+         also = "ਨਾਲ ਹੀ" ,
916+         cause_word = "ਕਿਉਂਕਿ" ,
917+         effect_word = "ਇਸ ਲਈ" ,
918+         true = "ਸੱਚ" ,
919+         false = "ਝੂਠ" ,
920+         neither = "ਨਾ ਤਾਂ, ਨਾ" ,
921+         full_stop = "।" ,
922+         comma = "," ,
923+         question_mark = "?" ,
924+         exclamation_mark = "!" ,
925+         word_space = " " ,
926+         sentence_space = " " ,
927+         colon = ":" ,
928+         or_word = "ਜਾਂ" ,
929+         and_word = "ਅਤੇ" ,
930+     ),
863931    Language .QUECHUA : TranslationLiterals (
864932        # From https://github.com/EleutherAI/lm-evaluation-harness/blob/0845b588303f1f59af98dd1c5bdbd78a9e75a1e2/lm_eval/tasks/xcopa/utils.py 
865933        language = Language .QUECHUA ,
@@ -1102,10 +1170,27 @@ def __getattribute__(self, name: str) -> str:
11021170    Language .TAJIK : TranslationLiterals (language = Language .TAJIK ),
11031171    Language .TAMASHEQ : TranslationLiterals (language = Language .TAMASHEQ ),
11041172    Language .TAMIL : TranslationLiterals (
1105-         # From https://github.com/EleutherAI/lm-evaluation-harness/blob/0845b588303f1f59af98dd1c5bdbd78a9e75a1e2/lm_eval/tasks/xcopa/utils.py 
11061173        language = Language .TAMIL ,
1107-         cause_word = "காரணமாக" ,
1174+         question_word = "கேள்வி" ,
1175+         answer = "பதில்" ,
1176+         confirmation_word = "இல்லையா" ,
1177+         yes = "ஆம்" ,
1178+         no = "இல்லை" ,
1179+         also = "மேலும்" ,
1180+         cause_word = "ஏனெனில்" ,
11081181        effect_word = "எனவே" ,
1182+         true = "உண்மை" ,
1183+         false = "பொய்" ,
1184+         neither = "இல்லை, இல்லை" ,
1185+         full_stop = "।" ,
1186+         comma = "," ,
1187+         question_mark = "?" ,
1188+         exclamation_mark = "!" ,
1189+         word_space = " " ,
1190+         sentence_space = " " ,
1191+         colon = ":" ,
1192+         or_word = "அல்லது" ,
1193+         and_word = "மற்றும்" ,
11091194    ),
11101195    Language .TATAR : TranslationLiterals (
11111196        language = Language .TATAR ,
0 commit comments