clovaai / deep-text-recognition-benchmark

Text recognition (optical character recognition) with deep learning methods, ICCV 2019
Apache License 2.0
3.77k stars 1.11k forks source link

Which parameters should I change for the Korean language? #315

Closed lvforce closed 3 years ago

lvforce commented 3 years ago

I have this character: 0123456789!"#$%&'()*+,-./:;<=>?@[]^_`{|}~abcdefghijklmnopqrstuvwxyzABCDEFGHIJKLMNOPQRSTUVWXYZㆍ가각간갇갈갉감갑값갓갔강갖같갚개객갤갭갯갱갸걀거걱건걷걸검겁것겉게겐겔겟겠겨격겪견결겸겹겼경곁근글긁금급긋긍기긴길김깁깃깊까깍깎깐깔깜깝깡깥깨깬꺼꺾껄껌껍껏껐껑께껴꼈꼬꼭꼴꼼꼽꽁꽂꽃꽈꽉꽝꽤꾀꾸꾼꿀꿇꿈꿔꿨꿰뀌뀐뀔끄끈끊끌끓끔끗끝끼낀낄낌나낙낚난날낡남납낫났낭낮낯낱낳내낸낼냄냅냇냈냉냐냥너넉넌널델뎀뎅뎌뎬도독돈돋돌돔돕돗동돛돼됐되된될됨됩두둑둔둘둠둡둥둬뒀뒤뒷듀듈듐드득든듣들듬듭듯등디딕딘딛딜딥딧딩딪따딱딴딸땀땃땄땅때땐땡떠떡떤떨떻떼또똑똔똘똥뚜뚝뚫뚱뛰뛴뛸뜀뜨뜩뜬뜯뜰뜸뜻띄띈띠띤라락란랄람랍만많맏말맑맘맙맛망맞맡맣매맥맨맬맴맵맷맹맺먀머먹먼멀멈멋멍메멕멘멜멤멧멩며멱면멸몄명몇모목몫몬몰몸몹못몽뫼묀묘무묵묶문묻물뭄뭇뭉뭐뭔뭘뮌뮐뮤뮬므믈믐미믹민믿밀밈밋밍및밑바박밖반받발밝밟밤밥밧방밭배백밴밸삼삽삿샀상새색샌샐샘샛생샤샨샬샴샵샷샹섀서석섞선섣설섬섭섯섰성세섹센셀셈셉셋셍셔션셜셤셨셰셴셸소속손솔솜솟송솽쇄쇠쇼숀숄숍숏숑수숙순술숨숫숭숱숲숴쉐쉬쉴쉼쉽슈슐슘슛슝스슨슬슭슴습슷승시식신싣실싫심십싯싱연열엷염엽엿였영옅옆예옌옐옙옛오옥온올옭옮옳옴옵옷옹옻와왁완왈왓왔왕왜외왼요욕욘욜용우욱운울움웁웃웅워웍원월웜웠웨웬웰웸웹위윅윈윌윔윗윙유육윤율윳융으은을음읍응의이익인일읽잃임입잇있잉잊잎자작잔잖잘잠잡찐찔찢차착찬찮찰참찹찻창찾채책챈챌챔챗챙챠처척천철첨첩첫청체첸첼쳇쳐쳤초촉촌촐촘촛총촨촬최쵸추축춘출춤춥춧충춰췄췌취츄츠측츨츰츳층치칙친칠침칩칭카칸칼캄캅캇캉캐캔캘캠캡캣캥캬커컨컫컬컴컵컷컸컹케켁켄켈켐텡톈토톡톤톨톰톱통퇴투툭툰툴툼퉁튀튕튜튠튬트특튼튿틀틈티틱틴틸팀팁팅파팍팎판팔팜팝팟팡팥패팩팬팰팹팻팽퍼퍽펀펄펌펑페펙펜펠펨펩펫펴편펼폄폈평폐포폭폰폴폼퐁푀표푸푹푼풀품풋풍퓌퓨퓰프픈플픔피픽핀필핌핍핏핑

Should I change the batch_max_length and data_filtering_off?

kimlia545 commented 3 years ago

First, you have to change this part. train.py parser.add_argument('--character', type=str, default='0123456789abcdefghijklmnopqrstuvwxyz', help='character label') Please refer to this place. #85