Skip to content
Stefan Weil edited this page Oct 12, 2022 · 19 revisions

Training

Examples of trainings for kraken

This sequence of examples was focused on getting a good kraken model for historic prints, especially for Fraktur script, but also with support for historic and modern Antiqua fonts.

frak2021

The training used the same data sets for training and evaluation as a previous training which was done with Tesseract.

It was done the latest kraken code plus additional fixes for a division by null and multithreading.

nohup time ketos train -t list.train -e list.eval -o frak2021_gpu -d cuda:0 | tee -a frak2021_gpu.log
nohup: Eingabe wird ignoriert und leite Standardfehlerausgabe auf Standardausgabe umgeleitet
Building training set
Building validation set
[52.2009] alphabet mismatch: chars in training set only: {'~', 'ὐ', 'ő', 'À', 'χ', '₆', 'Ò', 'ῳ', 'ï', 'ḡ', 'ώ', 'ἄ', 'μ', 'ϋ', '☛', 'Ἰ', 'φ', 'ō', 'ἰ', 'ξ', '͗', '№', '⅚', 'đ', 'ύ', 'ϖ', 'ὀ', 'Φ', 'Δ', '_', 'ϰ', 'ђ', 'ἀ', 'Τ', 'ᴐ', 'κ', 'Ó', 'Α', 'È', 'Ⅎ', 'ἐ', 'É', 'Ô', 'ὰ', '+', 'ϟ', '‣', 'Ž', 'η', '▲', 'Β', 'ί', '⅓', 'ž', 'Š', '⏓', 'ὁ', 'ό', '±', 'ά', 'ψ', '₈', 'î', '♂', '∙', 'θ', 'ⅎ', 'ℳ', 'Ç', 'є', 'ῆ', 'ϑ', '₄', 'ϱ', '✤', 'ὸ', '♀', '⅔', 'Ἀ', 'δ', 'ὴ', 'ἑ', 'ɔ', 'ζ', '᾽', 'ὶ', 'ì', 'ʞ', '⅙', '☚', '‛', '℔', 'ή'} (not included in accuracy test during training) 
[52.2010] alphabet mismatch: chars in validation set only: {'ϊ'} (not trained) 
Initializing model ✓
stage 1/∞
Accuracy report (1) 0.9612 1625450 63133
stage 2/∞
Accuracy report (2) 0.9604 1625450 64311
stage 3/∞
Accuracy report (3) 0.9578 1625450 68579
stage 4/∞
Accuracy report (4) 0.9519 1625450 78227
stage 5/∞
Accuracy report (5) 0.9428 1625450 92982
stage 6/∞
Accuracy report (6) 0.9421 1625450 94036
Moving best model frak2021_gpu_1.mlmodel (0.9611597061157227) to frak2021_gpu_best.mlmodel
154547.59user 1537.72system 29:15:33elapsed 148%CPU (0avgtext+0avgdata 5916304maxresident)k
385328inputs+453328outputs (68102177major+26511973minor)pagefaults 0swaps

Juristische Konsilien Tübingen

Get the data

The information here is no longer valid.

mkdir Juristische_Konsilien_Tuebingen
cd Juristische_Konsilien_Tuebingen
mkdir Training
cd Training
# Get image files.
wget --content-disposition https://files.transkribus.eu/Get?id=KWPGLJROIYFMCHXUKHROTBZI https://files.transkribus.eu/Get?id=FFZEEILUCMKWBMTOJNKXOQVO https://files.transkribus.eu/Get?id=DHTNZFGEQZVNFYSNNXSUGRZW https://files.transkribus.eu/Get?id=GLUHQPHRJZGIORAKNKSSHJHN https://files.transkribus.eu/Get?id=AXGGUKSGAASTMJLXHKLHTCMI https://files.transkribus.eu/Get?id=ZWNHITSTVYAYVRMXSACMYZSS https://files.transkribus.eu/Get?id=SJOIKRULJGZRUGOHANSBQCES https://files.transkribus.eu/Get?id=TXUOIQXJUFRBXPZXRIQKRGID https://files.transkribus.eu/Get?id=FBFDPQSANWKZHMEFPCAEBYGC https://files.transkribus.eu/Get?id=JJYAAMDIJMVXEMDQNQLFSNHP https://files.transkribus.eu/Get?id=CFHAJXQXKFTYWEMOXGMHIFTK https://files.transkribus.eu/Get?id=YJNIOPLPTVNXGLVKAYDWSZTJ https://files.transkribus.eu/Get?id=KSSNVVQMKCIPCOJNVRQIOIIH https://files.transkribus.eu/Get?id=OGZBGTHHLDZQRUFVQBDMDDPA https://files.transkribus.eu/Get?id=KYPDSMEVHCRVHUXKYNOXDTPT https://files.transkribus.eu/Get?id=UJUHLJMVEUIXOOVDPIQKPXFL https://files.transkribus.eu/Get?id=ZTTKWCFYHHJXPCJHVLGCFQCM https://files.transkribus.eu/Get?id=FOVUVMNPDJDZLKWIDYGJPLUX https://files.transkribus.eu/Get?id=DXRPIBLIMBBMIXVDXCSMOLDM https://files.transkribus.eu/Get?id=LZPWBKIXSJJIFAWPGUZSPVXT https://files.transkribus.eu/Get?id=HOHGKGGHOSQUUTAVACVQYWAL https://files.transkribus.eu/Get?id=GPQZHFWYWLZEXJCDTJKAALEE https://files.transkribus.eu/Get?id=TAAKRFOWGFBKCEFSHNQIVOIK https://files.transkribus.eu/Get?id=RSAIQAPCKZZTNAKLEHTWBPGM https://files.transkribus.eu/Get?id=NZGMAVAGNRHFMZNNUIJDTRLU https://files.transkribus.eu/Get?id=PCNIVHAJHFRBOYBPQERJTOED https://files.transkribus.eu/Get?id=OJKITNXIXTYUQPCZKYHQYFSD https://files.transkribus.eu/Get?id=SZJTDCPESLWRIHCZXMBEENTJ https://files.transkribus.eu/Get?id=LTREBTDWWJRERTPPUBONYEEH https://files.transkribus.eu/Get?id=KLINCPGDKEXBUQNBUDGJUOPY https://files.transkribus.eu/Get?id=ZGQKNMDFPAZNRBMMKSOZZHKA https://files.transkribus.eu/Get?id=HJFOROVZUNTZMHRZJJOOKJAM https://files.transkribus.eu/Get?id=JAOKFQRGZUHUSDRJUWEGEGUE https://files.transkribus.eu/Get?id=ULKQLRYWHGTQMGGDRWWVKNOG https://files.transkribus.eu/Get?id=FXHKSRQIETJMKXBQZHLEHHZZ https://files.transkribus.eu/Get?id=RQSXJYKCEUBGMPTEHSCBVAAT https://files.transkribus.eu/Get?id=LHMCRHNCINSMVCINCEKNACCG https://files.transkribus.eu/Get?id=LFGZRQTADQHPYNUJPNHFOLMR https://files.transkribus.eu/Get?id=XVTTWUSOAQFYRSZQVLQMMNYB https://files.transkribus.eu/Get?id=MQROMFPZURWSUWVWJAFNJWMS https://files.transkribus.eu/Get?id=XOIOSHOVCUSCCVHNGLOWBUHA https://files.transkribus.eu/Get?id=MQQGCETBKJUWDLIQYUKPRJRU https://files.transkribus.eu/Get?id=HOBEDMUTPAPQPGCVBFSSSUJS https://files.transkribus.eu/Get?id=RYRIDAQAQCWEUCFZSOJMIEVT https://files.transkribus.eu/Get?id=FKUXUMUMEVVMYEPKVXUTCKHY https://files.transkribus.eu/Get?id=ROOUZCYQSEZSDEBSHEMGPQCB https://files.transkribus.eu/Get?id=HYFOIAPWWUYVLBKKNCWWVQDR https://files.transkribus.eu/Get?id=MBLXSOIAFBZPRGCBEAWLSXIQ https://files.transkribus.eu/Get?id=JPTFEKQFLOIJABAHZPJGWTIX https://files.transkribus.eu/Get?id=GVOZSINXFOQINGIHOFLHZEXW https://files.transkribus.eu/Get?id=UIUAZQUCPLZMULJHNUGEYUKR https://files.transkribus.eu/Get?id=AJJVYMFQTWWIWJFCUKMJKJQP https://files.transkribus.eu/Get?id=GIJZOLKNHKZXBTBWPJJJLYYR https://files.transkribus.eu/Get?id=KRXUUEYRYVHLMMUFLKCGUQQH https://files.transkribus.eu/Get?id=RKUEWSQAKAWVGCJMBNOSIVNR https://files.transkribus.eu/Get?id=MNOPSPMISKMPTXUIJUSOEOCW https://files.transkribus.eu/Get?id=BFYUPHCCJDCUAZWITORDMLGX https://files.transkribus.eu/Get?id=APIACVAWACKQMPYJILFINMZT https://files.transkribus.eu/Get?id=TVBDRWEFCSENIIYFTLRYPEAA https://files.transkribus.eu/Get?id=GJYRHFRDWSHRLJVIGNKWWBZN https://files.transkribus.eu/Get?id=BQDKZIVQUJEXYYBCNVTBLHZM https://files.transkribus.eu/Get?id=DOIIRJFDXVHNAXFMUVFBPPYE https://files.transkribus.eu/Get?id=CSELZDAMABXKPWZNCSOHCBAF https://files.transkribus.eu/Get?id=SGFNBPQGXZMZAKGUSWXWOPVR https://files.transkribus.eu/Get?id=XQQIAQLQRMKMYCVHCLWHUYFQ https://files.transkribus.eu/Get?id=ROMIWFNFIBTGGCFLIZJJFWKU https://files.transkribus.eu/Get?id=KTUVRXMZBRHUYFVDAULYHKBC https://files.transkribus.eu/Get?id=AQHXCBAWQVKHUVUHZGDZPZUT https://files.transkribus.eu/Get?id=RXKIXYMKMLQBTJGXFYGCSGDY https://files.transkribus.eu/Get?id=UBEIKSGVTBMHIJPQOEGZSGWD https://files.transkribus.eu/Get?id=RWMVMWDCPNFBTQFGWQOAAVQQ https://files.transkribus.eu/Get?id=NYDHDAZGYHUDGXENOAASLFYF https://files.transkribus.eu/Get?id=SMVJNOMKZTUVAXKVDAJXSAZS https://files.transkribus.eu/Get?id=VIHRDXWGGOATSMCPZJJXAAXJ https://files.transkribus.eu/Get?id=YWFJNUVWHSZGMBDEZTJMCGFM https://files.transkribus.eu/Get?id=OXHBEFWYPNGMTTARWUPYSYUU https://files.transkribus.eu/Get?id=NDCENLOGOVSMHFPOTPOUCQKG https://files.transkribus.eu/Get?id=TIGUOIFRBCUMNOFZYARPGGRS https://files.transkribus.eu/Get?id=NKDMDKVMRTZWTKTRMCYDHFLW https://files.transkribus.eu/Get?id=IDIHIPZTDOXTOQCAWCBQMNOQ https://files.transkribus.eu/Get?id=IQOTZUSGWOKZEPBJBSYYJGTM https://files.transkribus.eu/Get?id=FRWNLSWXMYIPGQJGVPLWPIXT https://files.transkribus.eu/Get?id=QNXDOMQULVGUWUGHIKQDPCOB https://files.transkribus.eu/Get?id=EYISIDNNZSWKOQOYMQUTXVSK https://files.transkribus.eu/Get?id=RGFHPRMBPRYNEFLVEJTCOILG https://files.transkribus.eu/Get?id=FBZILEXIIRGVPPJLOYYSGHCD https://files.transkribus.eu/Get?id=KOIUIIXSXVJDFERSFIXMSZUL https://files.transkribus.eu/Get?id=OYLYXYLBLJYUJYOPLEFDHKLH https://files.transkribus.eu/Get?id=KJVSKBNNSJOQFGSRXVMRZYQR https://files.transkribus.eu/Get?id=UHAHGFFAMRNSTNQGMOBNQTUR https://files.transkribus.eu/Get?id=ADDQMOOXTWMOXGEEYMANWNJZ https://files.transkribus.eu/Get?id=HESNYZGFTYGUDFGZANLQLMOF https://files.transkribus.eu/Get?id=NQCXKFIDTGDACYHFUJZKJQDS https://files.transkribus.eu/Get?id=JZUMSCTEYEHMAOXDAXZNXGUX https://files.transkribus.eu/Get?id=IBERFVXKQCGEAHQNKPNCSNMA https://files.transkribus.eu/Get?id=FGLWDUKYAMWQIHEHKKQOOLMW https://files.transkribus.eu/Get?id=BCUVIEVSNFOUAYXSYCHNPXVP https://files.transkribus.eu/Get?id=XGPDURDECYWHYWZLJHDMIGLU https://files.transkribus.eu/Get?id=YSSYTELQQLMAHAVXQOGPUTUG https://files.transkribus.eu/Get?id=JOYKOTRPSTUFZXBORFKRCZJC https://files.transkribus.eu/Get?id=USOQSWUKNUEATWIZHMUYLDAF https://files.transkribus.eu/Get?id=VKIXEWUCBSZSKBBGIBOVUSUG https://files.transkribus.eu/Get?id=MEYLSJNTPSHUCXXUAJKCZPKA https://files.transkribus.eu/Get?id=XJDDZZAHHOSJGEPLXOBBRDOA https://files.transkribus.eu/Get?id=MYHOGTLJMXZUVRLJOSCUWWFC https://files.transkribus.eu/Get?id=BLASBUUDLXUNZMUIEVJUEJKJ https://files.transkribus.eu/Get?id=WGPNKBRBCKXVXKFONPETTISL https://files.transkribus.eu/Get?id=QMZIHCDKWOSFAPQVPENYRPHJ https://files.transkribus.eu/Get?id=KONWXWVKBGRGADHIGCFGFOXO https://files.transkribus.eu/Get?id=XXQAPZNSVNXIPGRGVHPTABFI https://files.transkribus.eu/Get?id=IPQPOXCWANMWHGBRMRKIYZLS https://files.transkribus.eu/Get?id=VIMEFFPNDFRWXEJQNXODRBEO https://files.transkribus.eu/Get?id=ZAABPSRTGFETETNGYMGLZCOL https://files.transkribus.eu/Get?id=VLSHOPUQJNHGWWSZKKXQAAHK https://files.transkribus.eu/Get?id=HZIGXWAUPNIABHHIBQHBTGEC https://files.transkribus.eu/Get?id=WDHNPKASTGLNHDIVJKSRVFMX https://files.transkribus.eu/Get?id=ZQWHSMWYZFDJUUDMGEYJYDDY https://files.transkribus.eu/Get?id=ZECCNUCTLIHMSLJPJVGAWUXP https://files.transkribus.eu/Get?id=XKKHGWQTAHWAULBAZBLBQCBQ https://files.transkribus.eu/Get?id=BDFTIFSQHNPGIQHLUSQBNZGL https://files.transkribus.eu/Get?id=NMFPJUPMQPKXBBLCBLNVDUQA https://files.transkribus.eu/Get?id=ZANEBGCYUOUQLYXHKOVSPMER https://files.transkribus.eu/Get?id=EGEUWBWWBYWLKJISSMKERPRZ https://files.transkribus.eu/Get?id=YFACZWEGHUMIJOSGFDIFIIXI https://files.transkribus.eu/Get?id=WXWUUGZKGYIKYFJVUPXLXNLN https://files.transkribus.eu/Get?id=BHERQQEMBDPKHNQCTTIABRAN https://files.transkribus.eu/Get?id=DKFQEVFTRHFUYTNWUWKGMDCC https://files.transkribus.eu/Get?id=MHUFOPWVDKPNENGOTPHMRBPG https://files.transkribus.eu/Get?id=PMKYLRJEIQBIIFQVGAAKBAYE https://files.transkribus.eu/Get?id=BTHWVSEFZFEQVWQXNLLVMQQX https://files.transkribus.eu/Get?id=YWYAOOYGHXSNXEMOWXFVFBBV https://files.transkribus.eu/Get?id=ZARBXUTLMISJJORKXYCCLKLJ https://files.transkribus.eu/Get?id=VIIJFLMYIRGLEPGPPAEUOHEG https://files.transkribus.eu/Get?id=BJAVSUFRATHIZMZJBFFVZLWY https://files.transkribus.eu/Get?id=XAFFBZVXTIOMURIZJANEDCRL https://files.transkribus.eu/Get?id=EADIASJDESBUIGBHBBRVBENC https://files.transkribus.eu/Get?id=UHXRNRWZURNNIITNZLBHIKAT https://files.transkribus.eu/Get?id=LYMSCWGIRTABXAXYSWMKLXFF https://files.transkribus.eu/Get?id=RVHBELCYHNQXRVOVSFHFPHUF https://files.transkribus.eu/Get?id=WYPAPFXUXDIYSHGCVUHPCDOE https://files.transkribus.eu/Get?id=RAAORHEZEHXFRPDCFVOUDSNE https://files.transkribus.eu/Get?id=AUMCSARBRKNJLWQGOEFAVEVN https://files.transkribus.eu/Get?id=LHPLORVBMFYIQXIFUOOQNWLF https://files.transkribus.eu/Get?id=YECVIISDTADFSAVYVVRULJED https://files.transkribus.eu/Get?id=BFQTOOGEOSYCFSTIJFYASZKN https://files.transkribus.eu/Get?id=PJXUZAFEEIUWRLDOENGGGYQM https://files.transkribus.eu/Get?id=YOHJAMQIAVSFHWRUESLIHEJH https://files.transkribus.eu/Get?id=EJXPQXSUQUIUSHKMECISPLRE https://files.transkribus.eu/Get?id=ETRCBMFJJFFSCBWGCWMKVYPU https://files.transkribus.eu/Get?id=RSYKJGUJTWKLWHUVWCZGVHBK https://files.transkribus.eu/Get?id=LHMYRDDJSVVHXCFZAPPIUYWS https://files.transkribus.eu/Get?id=JUQMUFIOQAYNHJSATWZVAVJK https://files.transkribus.eu/Get?id=WJLWJKXFVIOXLXEELWFAXEWS https://files.transkribus.eu/Get?id=XVFBFYOABAYUUBGEETZFNDUU https://files.transkribus.eu/Get?id=GSSITABWCVXAEYCPEGGAGWHO https://files.transkribus.eu/Get?id=HHQPEJLPUUSTOJVMGYHEMCWO https://files.transkribus.eu/Get?id=ETKNGULSXICMTXMSUNGUHAGR https://files.transkribus.eu/Get?id=EVVBDQATMBBOWZQVWWBGWFTE https://files.transkribus.eu/Get?id=JQNZVEEWHOTGPSGQWBJTYDXZ https://files.transkribus.eu/Get?id=HDJMNCKXBJTYRBANCDDNBOTV https://files.transkribus.eu/Get?id=FXWRRMVBGADVUBCLXQTUYPAT https://files.transkribus.eu/Get?id=YHIBTJOSJFQPMEFCMUASZYRR https://files.transkribus.eu/Get?id=WUALYWRDKBXDJKNGZGXEYHKQ https://files.transkribus.eu/Get?id=IENQXLHPZTOZJEENEIKRAKNU https://files.transkribus.eu/Get?id=ABWEDGWBHNRUHCDAZQBCHDHR https://files.transkribus.eu/Get?id=HOHZGRKUZEEHTQFOZIUVWRKG https://files.transkribus.eu/Get?id=TYEHTOEKTZMMCEFACTOKAUFI https://files.transkribus.eu/Get?id=DWSKPMUEJAIOJJYYGIRBIGCQ https://files.transkribus.eu/Get?id=JOUYTXSVTWDUMSRJHPBXFGKX https://files.transkribus.eu/Get?id=FGDLEJDZPQNYAMEAQXIQLDNK https://files.transkribus.eu/Get?id=JUXAJMEYQARQLPURIWTWGBBP https://files.transkribus.eu/Get?id=KRNJXKZJCTBMJJDLHHMMSINE https://files.transkribus.eu/Get?id=YULOHZIMFHEOJKWDHCFOLBUL https://files.transkribus.eu/Get?id=OFTLVWCMKVUPUZZJCRLXEIYS https://files.transkribus.eu/Get?id=YHRJAICVWUIQGVCSIPBMKEKH https://files.transkribus.eu/Get?id=JIYRQKQXDYYQJOFINPLTOQSZ https://files.transkribus.eu/Get?id=SSASFJCYPAANTCROPGETZUPP https://files.transkribus.eu/Get?id=QEXKWYLPSRKUWBCIGQNHHPTP https://files.transkribus.eu/Get?id=SVZAIDARQWLMQTKQVIOOSAKV https://files.transkribus.eu/Get?id=XADEUHKOGGIWCOCZMTGGFHDP https://files.transkribus.eu/Get?id=QRUEXOAFCJBJXSXVTYKDCVKF https://files.transkribus.eu/Get?id=TZWAQKCGYPBWVZNSHVAKWCLZ https://files.transkribus.eu/Get?id=HVTGPLBHUEUXGJMGBNHIJIVS https://files.transkribus.eu/Get?id=KGTWYGJEQVUTLYQJJHDYRSDR https://files.transkribus.eu/Get?id=UCCYYBWUNOZJTXPTLDSGMKED https://files.transkribus.eu/Get?id=FTRLAADUOFFQLRZKSCULMDKF https://files.transkribus.eu/Get?id=BYLJWEGQKLYXLDPIEERCHAHG https://files.transkribus.eu/Get?id=SYZATLLCHFOLZWDODQRXNKNF https://files.transkribus.eu/Get?id=WZJQYUAGMQINLJKWSLBKSGTG https://files.transkribus.eu/Get?id=WOTWYEYWCOLUIZDUZMKQOBPA https://files.transkribus.eu/Get?id=IZHOVYUENLZPLEPYGPDACJXR https://files.transkribus.eu/Get?id=ITMLHBAAKCXDNTLMKMXNWOVM https://files.transkribus.eu/Get?id=GTEWMTDGWLEQMNDTONNPWVAV https://files.transkribus.eu/Get?id=XHVQSIUMEGUOGCMLVIBNPDIZ https://files.transkribus.eu/Get?id=IYDRMOZDVMAEHZLCLYWJFKZU https://files.transkribus.eu/Get?id=OZGAGQKATLEVBZMKNPUSZJZA https://files.transkribus.eu/Get?id=BMYKTWTVVTYYYQQSTXWQHKLC https://files.transkribus.eu/Get?id=IYUGXNCSKYCGDZZATBMXIPRX https://files.transkribus.eu/Get?id=BBTDFLKQWLYLTYYEMEYWRCFX
# Get XML files.
wget --content-disposition https://files.transkribus.eu/Get?id=JHVUSEFVRTWKTDYBEFVBEPFB https://files.transkribus.eu/Get?id=BYFBODHLFRJDQBVOROTLEWET https://files.transkribus.eu/Get?id=RWSSDKJHNAPTDXVLREBGBCLW https://files.transkribus.eu/Get?id=YEPDJBAQSQUMXTBRNXGGFREU https://files.transkribus.eu/Get?id=FZQWWFZCNWCYAPQRVIZTAOSX https://files.transkribus.eu/Get?id=IMXDAERWVVGYBIOMIEFTXFMK https://files.transkribus.eu/Get?id=LVZAFONLFVUFHYVDCAHAERYP https://files.transkribus.eu/Get?id=CIMPTJQYTQESVVGCBQRALMEX https://files.transkribus.eu/Get?id=CVSIQYLKNTXSVXRHBRBHIKKM https://files.transkribus.eu/Get?id=UIOFSVNICSMOSBKPJBZOATHM https://files.transkribus.eu/Get?id=NIEOIIUKDXUNFYJHHXADOUZH https://files.transkribus.eu/Get?id=RECFLPURJRTWVCNLRPARZWCT https://files.transkribus.eu/Get?id=BIIMISGELCYEAVCSFBEOBEXE https://files.transkribus.eu/Get?id=YSCXXSWLKELAWMDBRLAXKZPM https://files.transkribus.eu/Get?id=PPKQXCNQTIPCWWKKYEYKKFWR https://files.transkribus.eu/Get?id=TMGWJSEMDVZZUHIMCJMBQSAS https://files.transkribus.eu/Get?id=RQQGZJRNLALMLBHXWZAQRJJB https://files.transkribus.eu/Get?id=NKRITEGYTBOJEQKHIYFVPIKG https://files.transkribus.eu/Get?id=OHMVVRMMPUDTYDRDFHJZPKHZ https://files.transkribus.eu/Get?id=IQWIGWEOZAOZGEBTROXULIFN https://files.transkribus.eu/Get?id=XKLRDXMZFCGBBDKXXWQUOGTP https://files.transkribus.eu/Get?id=KBBFFENIEDVCCEJKGINYYOIA https://files.transkribus.eu/Get?id=VHHUSIBLNZVDYJDEUGTFNFIB https://files.transkribus.eu/Get?id=NWOLIMGYYSFJGHASEFSAHPSU https://files.transkribus.eu/Get?id=CMXZGEPQCEQPOYXCVMUAAMXL https://files.transkribus.eu/Get?id=UJMUBEOLKJSRNPSVFBSPIIQE https://files.transkribus.eu/Get?id=QNKLLUDVYNWHQFYPDGDKOSWG https://files.transkribus.eu/Get?id=UBELSVTLYVTCJSWPBNAKXCWN https://files.transkribus.eu/Get?id=UAQBIBAIRXGBSPALZPERMJQN https://files.transkribus.eu/Get?id=ABGYGICZRGMJPDKBTQRFMZWY https://files.transkribus.eu/Get?id=IEOAQAMGVZGEDZPSJCWPXJOW https://files.transkribus.eu/Get?id=JLXONNHBYYYPWIACEFKYYIGW https://files.transkribus.eu/Get?id=VJUZWLEDNECLNPBDKLPQIXPP https://files.transkribus.eu/Get?id=IWOOMDSPRWIHAUSGCDHMDRDY https://files.transkribus.eu/Get?id=NNHSAVUBAHYCJNDDFERAQIPK https://files.transkribus.eu/Get?id=ORCJXYGXILWSHLOOHPSECMQL https://files.transkribus.eu/Get?id=DDJGAXYOUQPGAQUBYVKNPSZI https://files.transkribus.eu/Get?id=ONNXZWVCJEGUTVJFDUVLSPIG https://files.transkribus.eu/Get?id=IIAVPDPIIJCOHAZDKJDBAMHL https://files.transkribus.eu/Get?id=VNVJLQRTQLJKWIUJPFDMRNPT https://files.transkribus.eu/Get?id=UYHHWDUMYRMHUMQXPBAWJFFA https://files.transkribus.eu/Get?id=YSSLAKWGNUJKXPMFGMFCHPNW https://files.transkribus.eu/Get?id=AXPZFHVZZQROPNGGQEYZNZXO https://files.transkribus.eu/Get?id=HHQOOTMTFVPDCARHHHGBIAUX https://files.transkribus.eu/Get?id=POTHXBKQPCHCXROHCSKQIWBJ https://files.transkribus.eu/Get?id=DSCRKYNYRSZFFHCQSDFCYUCO https://files.transkribus.eu/Get?id=JZPDYBTUFJAFSUACHHIZBWXH https://files.transkribus.eu/Get?id=YOECZHMWXUKAVVZYRONOQLJX https://files.transkribus.eu/Get?id=URNYUAFDAUHQPAHFMUUBTKHT https://files.transkribus.eu/Get?id=REFTLJUUJNMDPSVNZTSBEJEB https://files.transkribus.eu/Get?id=XMUKGQJUGFOGGGPLQNECYVBK https://files.transkribus.eu/Get?id=TCPFSFMTPGDZEWSMJZBXQEGH https://files.transkribus.eu/Get?id=YHMLVPJHADPBXBWKVJFDNTEL https://files.transkribus.eu/Get?id=JYXIBXPCZESDNNSHCBETTLLB https://files.transkribus.eu/Get?id=RQVWZBXVHZHRBPDJLSYDINFB https://files.transkribus.eu/Get?id=KIUANWLYIDEJYTDSVUFAWUCV https://files.transkribus.eu/Get?id=POZKJCZOPOKGILWSBPFUYDNK https://files.transkribus.eu/Get?id=PLIDQKLLDDSTURVBSYUKEHOJ https://files.transkribus.eu/Get?id=IRPJIVREFPZBZYXFELHDZWTH https://files.transkribus.eu/Get?id=GCAWPLAWQVTKWUDKTXBCTRJY https://files.transkribus.eu/Get?id=LTGBEPJCSYDUNHIIFKLVUJQX https://files.transkribus.eu/Get?id=AYRNLZJFWCBVZOFUNVQBTDYE https://files.transkribus.eu/Get?id=FHLBDWEYSNDJJZOAOBJTSVBU https://files.transkribus.eu/Get?id=DLDNAFADKWZLSBRCBLVCXCSL https://files.transkribus.eu/Get?id=YQTEGXGRQLFNNOYEUWBFRYWL https://files.transkribus.eu/Get?id=XHVQSESDDDCAALHGRMUBTXHV https://files.transkribus.eu/Get?id=RQLANQXXACUDSKBSYKIQNODK https://files.transkribus.eu/Get?id=FYQCLMNREDKIKKOUOOLGWEAX https://files.transkribus.eu/Get?id=UNFTXZTWQDELXHLHXDZSDWYX https://files.transkribus.eu/Get?id=ORBMIMGYCXLYQMYIAEIOXPJV https://files.transkribus.eu/Get?id=VEITMPANUCBOUFALNYCRWHTI https://files.transkribus.eu/Get?id=ECNNPBOIQRTBDNLSPXDMPGLP https://files.transkribus.eu/Get?id=QKCSSFYBNNVCOSFIADEDCXDA https://files.transkribus.eu/Get?id=LBPAHTIOKUBENKAYUXZECLSN https://files.transkribus.eu/Get?id=UXPPULKUSALSNGREBVUEWAHK https://files.transkribus.eu/Get?id=EIPMIKZZMACPQAYNQHTJCNKX https://files.transkribus.eu/Get?id=NFVGPKGIVBNNHLUSALCJKVQD https://files.transkribus.eu/Get?id=DZHPULUWLVNRBEVGOOWJPHAY https://files.transkribus.eu/Get?id=DTPIETEDOPLJBMYMVJRASTVP https://files.transkribus.eu/Get?id=TLGKJAWVGBPWWSUPTXHZYTWA https://files.transkribus.eu/Get?id=VQLTSTPZFUURQZWASJKPPLRR https://files.transkribus.eu/Get?id=TVUSOUXQBJCRBWDGZWKBUJSL https://files.transkribus.eu/Get?id=PXWWTCWWTEWUONCKNUBJSJDQ https://files.transkribus.eu/Get?id=JJBWWXFHRTPDNVEJHVMEEPRO https://files.transkribus.eu/Get?id=BVNFMMNMGEOFCNOXKNKAODQH https://files.transkribus.eu/Get?id=VCZYYLYPGLGDUEEOLTSKGRCP https://files.transkribus.eu/Get?id=YRBFGZUNUWESRFIZAWFZDLXA https://files.transkribus.eu/Get?id=RTMXEFVCUSQYBUFGXJROOSQW https://files.transkribus.eu/Get?id=WHUQMBKMTMBXEAJTENUGAWTY https://files.transkribus.eu/Get?id=EOFPMFLTWMSWKVWNUOOBFPWG https://files.transkribus.eu/Get?id=XQFIWGCRJEASTRDJAEJUCKKD https://files.transkribus.eu/Get?id=KZRXAIFVAXZSUELIMALBPERE https://files.transkribus.eu/Get?id=AOZASWDDOAXNSZAZUUAKZUVW https://files.transkribus.eu/Get?id=NQHOQQTPPIGHMJOXLHZCMKIK https://files.transkribus.eu/Get?id=ZWGLSCBYSFNEBDLBAZUXLAEC https://files.transkribus.eu/Get?id=ZNKWVPEGXHBULCAOEKGGGJKR https://files.transkribus.eu/Get?id=TKANIGGKMDYJJBZJVOMVKBHL https://files.transkribus.eu/Get?id=WVQFNOXILOFEARUYWIHZDCVP https://files.transkribus.eu/Get?id=CGUNPASSZQOASOYNDPAPWUAI https://files.transkribus.eu/Get?id=WFYPPCWSRVAMVXIJROZTESKN https://files.transkribus.eu/Get?id=PLGVGOXDAMGBNCISZQHKBVVJ https://files.transkribus.eu/Get?id=BJXKMZOQAUZCFJGHDKIDIKWY https://files.transkribus.eu/Get?id=CMDZVVKNEJZKHZKZRTRQOLBC https://files.transkribus.eu/Get?id=IUCKTFOEMGCBWQMSNKBJBMLE https://files.transkribus.eu/Get?id=EIZMZATKMGACEAALZFSMTHQJ https://files.transkribus.eu/Get?id=YJLDTJWFIEZYUDLALRXVNIBK https://files.transkribus.eu/Get?id=QSJICLTFSXESUQEFXPMEVACO https://files.transkribus.eu/Get?id=JVEAELRWZRPXATUPJNMPSAIO https://files.transkribus.eu/Get?id=XLISFOZJUAXSOZXKWGTTHKDD https://files.transkribus.eu/Get?id=VVCOOOIHGMRDLSMDHMPGYLFE https://files.transkribus.eu/Get?id=GZEEXENYIDTXCTPBSAORHHSA https://files.transkribus.eu/Get?id=IDGJOMRHAFKFUEHXKQGCUFNG https://files.transkribus.eu/Get?id=UKXMGZJJCWNYHCVKXSOCWTBI https://files.transkribus.eu/Get?id=BARPNPDXBCVDSYLRXQITLSLU https://files.transkribus.eu/Get?id=HTWTCHRHVYIYHVMITIYNPGOZ https://files.transkribus.eu/Get?id=NZSTGDRQKMIECMYFKIXRGWQF https://files.transkribus.eu/Get?id=ICTGHWNQUZZZGZOMQBKWFDIP https://files.transkribus.eu/Get?id=QZYDJWWQDHCWRRWCIKBFHLVB https://files.transkribus.eu/Get?id=XDUTNFFNKXYTPZGFHFLEHTSY https://files.transkribus.eu/Get?id=ISXNKHWMHCVPRJLHBZPOWXNR https://files.transkribus.eu/Get?id=QXXAZLBIQPTNGSQBIAYVNGWB https://files.transkribus.eu/Get?id=NYSQTMYOGWTKDEPTPKNCJUWC https://files.transkribus.eu/Get?id=EONVKVKOIAHEROQVYEQZIXKR https://files.transkribus.eu/Get?id=XVNYPROHFPLWSUQKMCBYVOTI https://files.transkribus.eu/Get?id=HUDJOFZWWGELFRBJVVZNRJAS https://files.transkribus.eu/Get?id=FFDRJLMVQTQGUITZNOWSRFHT https://files.transkribus.eu/Get?id=OIAVJGZJSNRVPIQOQGKELTTS https://files.transkribus.eu/Get?id=NKYVSVOAKQUOCKKVRYHBGZYH https://files.transkribus.eu/Get?id=KJNSHSVFLRTXPBRLABNVJNEZ https://files.transkribus.eu/Get?id=MBSKPVBMWWDUAAXUEJDYNLPZ https://files.transkribus.eu/Get?id=MHLGDEADRYDIURDNKUFEGPOB https://files.transkribus.eu/Get?id=CTDTEHIJXZCTRQOMLHCQAJIE https://files.transkribus.eu/Get?id=SNZPGLFMDJIHJTIJWPUWMNLO https://files.transkribus.eu/Get?id=NICXUBXEUYOWGTCAOHUXLQYP https://files.transkribus.eu/Get?id=YKDBOQISRUEPRNOGCQRJSKZO https://files.transkribus.eu/Get?id=DCKSZDCZAELVYSCOGSYMZCRV https://files.transkribus.eu/Get?id=JAZYPAUWVLULYUMXQXSXPDUP https://files.transkribus.eu/Get?id=KJFYLNDRPHMNWWNBRASILZSO https://files.transkribus.eu/Get?id=GXZXDWEKOUMRRWVGCKBVOYDX https://files.transkribus.eu/Get?id=XXASBUPYNMROTAPFYCPEKTFO https://files.transkribus.eu/Get?id=UTWJIKOCOWILGTHAVQELNXZP https://files.transkribus.eu/Get?id=UQWTMHFCAGGNCRPDHMISHEZQ https://files.transkribus.eu/Get?id=QOLIGHKUCUPILRHJZTQBMAQW https://files.transkribus.eu/Get?id=QKGCVEHTCAHRULIYFKOKKCFY https://files.transkribus.eu/Get?id=KHQDPSYPSBRGAJUFRJMOYYHV https://files.transkribus.eu/Get?id=VKWKDCOHWOCNAROYRJKHLFYC https://files.transkribus.eu/Get?id=XNLDKCGFMCHVEZTAJQLOHVYS https://files.transkribus.eu/Get?id=LSUTSNAWNILOCLCJLJVHHPPP https://files.transkribus.eu/Get?id=VBICXRUQAIGVEXAVMWTWIUDK https://files.transkribus.eu/Get?id=KJFZNMXHNIVVXNRWRLNPFOOT https://files.transkribus.eu/Get?id=JBOQSXZLWOLKVMLXOVQDOKJU https://files.transkribus.eu/Get?id=IMDEJHWZKACFXIGKEFXPQLII https://files.transkribus.eu/Get?id=QZKWGDEXJPZKTDUCDFWDEOET https://files.transkribus.eu/Get?id=EQSKRRFMYFDNFAJKYSBEAGQZ https://files.transkribus.eu/Get?id=ECTQOJPXUKYFSXVWRRSGVQRO https://files.transkribus.eu/Get?id=XRDHUZNHGKFBUMJFWMZSNYGM https://files.transkribus.eu/Get?id=VXBCUSYDWISJNFPVLOPGMHQL https://files.transkribus.eu/Get?id=HASZIIPNMFNRBGRFNAHLLUBR https://files.transkribus.eu/Get?id=MUKOMKPEEGXLSMZSNNZVZQVF https://files.transkribus.eu/Get?id=TEYUPVVQHWBKCNMEBDOYBTNY https://files.transkribus.eu/Get?id=YYEOVUJKGVARJPNSIFGLTJDA https://files.transkribus.eu/Get?id=OZBOZQQGNFNCAIZRFBXBZTQU https://files.transkribus.eu/Get?id=SPDIYTHJUBLTXUFRDZHXVVNF https://files.transkribus.eu/Get?id=GBDRCZDQRFRWUHIBQMQQFWOB https://files.transkribus.eu/Get?id=PUXPSBYVMNNKUVFDVBXBUFGU https://files.transkribus.eu/Get?id=VWQGXYKCTNVODXWYFXCXIESE https://files.transkribus.eu/Get?id=QAGEQHUIBPCKCVZKCAYSKJAS https://files.transkribus.eu/Get?id=DXHLEYZKFOUPCGDJKILGLEEN https://files.transkribus.eu/Get?id=XJIGEZRWLZCHXNFMCJVGRJYO https://files.transkribus.eu/Get?id=CDFNKPNPFNLUDLMHNSVWIOJY https://files.transkribus.eu/Get?id=GMJIJQMKUDSKYOZBYNUSAXNB https://files.transkribus.eu/Get?id=ZIKEBIHOCZEPWRSXGLQIFANV https://files.transkribus.eu/Get?id=UMXWNECUPMBVUSYXQSQTLJXA https://files.transkribus.eu/Get?id=DOYOKEEZUPSEPOLQFFOSDXLI https://files.transkribus.eu/Get?id=BXVPCUAOJBDIXZTHJNVJQUVG https://files.transkribus.eu/Get?id=ZKGAJZFHGCNQPJACKBVUONZZ https://files.transkribus.eu/Get?id=KAAROZFEMHTPFBBZRROKOWBF https://files.transkribus.eu/Get?id=KGDGOZQBVHZIPFABKPUYMOPG https://files.transkribus.eu/Get?id=QEJFWXTMXGFJTFWCORHVCGCR https://files.transkribus.eu/Get?id=LPDOYQEQTNASZIWTBRVUZILS https://files.transkribus.eu/Get?id=YLQKKITJQKAOCOHWYOTYZUMC https://files.transkribus.eu/Get?id=LOTMIHEOLSGYTTJQPJCUDXIF https://files.transkribus.eu/Get?id=NLDJIZLJUKUYPWQZHQPRBACG https://files.transkribus.eu/Get?id=PNLSPNFUBYEFSRAMKEGHZRMX https://files.transkribus.eu/Get?id=TFLTRJJTYRGVLRUZKWXQJTSL https://files.transkribus.eu/Get?id=IGKKGDXZEUPHQEHRNVGGGSHF https://files.transkribus.eu/Get?id=KDNTAWHVDCRMWUJYHCXXNAEJ https://files.transkribus.eu/Get?id=CJNSNRLHKBUNFWVVDPFBOYGC https://files.transkribus.eu/Get?id=OZJOAVTDJANNGFWRCSPSSMNT https://files.transkribus.eu/Get?id=ZAKYZRJPMEUGAFMEETNWJHYD https://files.transkribus.eu/Get?id=AAKZVULGHVARRSMQCUZFRQOT https://files.transkribus.eu/Get?id=ANJKLTOEBRJAOFAZIQAOPDVL https://files.transkribus.eu/Get?id=QQENWYLYGEORJOOWXQGDDOYQ https://files.transkribus.eu/Get?id=RWKPNTKIKQTIMRNWWYGVBHXM https://files.transkribus.eu/Get?id=IAOLLMALNIHPDWGDSPCAKUFB https://files.transkribus.eu/Get?id=TOZWZVEROLTDQSSGQVVIDPFW https://files.transkribus.eu/Get?id=NZAQUQJMEOMMALKIAJTKYCEB https://files.transkribus.eu/Get?id=IYTYIEJULURSVZXZJOKNLGTP https://files.transkribus.eu/Get?id=PDDUAQELWDADQVLKYKTMUIEO
cd ..
mkdir Validation
cd Validation
# Get image files.
wget --content-disposition https://files.transkribus.eu/Get?id=VKHSBYKDRSPCXNYVVDUUJHIJ https://files.transkribus.eu/Get?id=ZFDKMFYHZSSWJRWJDBSKXLME https://files.transkribus.eu/Get?id=KUNKUUMWRVICWVMISJETKKSP https://files.transkribus.eu/Get?id=VCYDPLIKHZNSMBXZNEARDMQG https://files.transkribus.eu/Get?id=DSZCVDYESIQHIKUWNBMEGUMI https://files.transkribus.eu/Get?id=UMTPCFPNQPEMZGGSGOZWRHZZ https://files.transkribus.eu/Get?id=XGLLYKKRWGKDJTSTQZTQSBDB https://files.transkribus.eu/Get?id=GMDPHGGENKKZXGUNAYDUYWLO https://files.transkribus.eu/Get?id=YWGNYDFWVHJLFFSYFHFOPPWV https://files.transkribus.eu/Get?id=IVUIBDCTBUYBGAEWVULXXFST https://files.transkribus.eu/Get?id=SQCNQZEQTSUHTQWYPFYQALUH https://files.transkribus.eu/Get?id=SCOYGUKQFUEPENAKBDSOXYGF https://files.transkribus.eu/Get?id=UHGKWLEZJGEGEUDTJWHQHWGT https://files.transkribus.eu/Get?id=RLXCXHKMEZPDOPKXEQWALGPY https://files.transkribus.eu/Get?id=REUHQRVBWHKVAGYKNOBKCSNP https://files.transkribus.eu/Get?id=FARONPPATEJYGRXVRHXRBPPI https://files.transkribus.eu/Get?id=BYXLCAPXMRURUGUHGOCDUAWJ https://files.transkribus.eu/Get?id=RHZLBFIAGXPNVSEKDMXVNAZT https://files.transkribus.eu/Get?id=HMXEFFBXKKVTASJBJNLMYPNL https://files.transkribus.eu/Get?id=JRNXMKFSVZEOMVCMPGWFRPYJ https://files.transkribus.eu/Get?id=NTGOQDNHUKAKGEUWVHRHDUOR https://files.transkribus.eu/Get?id=JJLLOATKBIRFPVVUMQETOTTN https://files.transkribus.eu/Get?id=CAOYGGOXCZVJYMBKTZGNCFCT https://files.transkribus.eu/Get?id=WSAKKKUEQCSGPKCLCVHHCKVD
# Get XML files.
wget --content-disposition https://files.transkribus.eu/Get?id=CHDVOIIYLHWZEPXMAUJWWNKU https://files.transkribus.eu/Get?id=ZRLEVECHRBPRETNOKXOFWGPU https://files.transkribus.eu/Get?id=WCIRBUXVTCXPGZTDSGJLGVWR https://files.transkribus.eu/Get?id=LMXKCJUHLQTOTWZKQOYBXRRH https://files.transkribus.eu/Get?id=RVNPPKIXIUTUFYAYGDUMWPRB https://files.transkribus.eu/Get?id=BOCCILWYTULGOIMRGCUHLLBF https://files.transkribus.eu/Get?id=PHXCDTHIHWHPCPVZANZLMEZW https://files.transkribus.eu/Get?id=ERCWXYVDDXSQTNDNRADLLNSL https://files.transkribus.eu/Get?id=VWCEXDWQJWMZPPBWPGVUFLSU https://files.transkribus.eu/Get?id=REJFIMVSKOBVFYCGVPRYXSIL https://files.transkribus.eu/Get?id=QPDBOUSCBUNSAVIKCHKWKLOP https://files.transkribus.eu/Get?id=XVCPFSVVGHVLQYTRGUPECDBO https://files.transkribus.eu/Get?id=KAHBRRJHNECLTFIDYKEVHIKN https://files.transkribus.eu/Get?id=WPWCKGCRUCANVXXOAXMMMJHM https://files.transkribus.eu/Get?id=XQUUXHYOWQGAUAQIYITHIMBX https://files.transkribus.eu/Get?id=XJYJHKBQWTLIIBGYXZYMOMTE https://files.transkribus.eu/Get?id=RWJTCQCXVXQOTBIMHLYRZXYN https://files.transkribus.eu/Get?id=JBLGRLYTPUUOYLKCPNNSMGCY https://files.transkribus.eu/Get?id=AYSTOXAKCOUKGCLBAGEDTBQN https://files.transkribus.eu/Get?id=KRNQGZWNGROPPIIJXPATHOHH https://files.transkribus.eu/Get?id=QDARVKDOIPYISUFZBCOTHCXV https://files.transkribus.eu/Get?id=EMFJCMSQWEELQTLIQDVDHBHJ https://files.transkribus.eu/Get?id=FDIATJURTJQLLGMJTHTQIQTC https://files.transkribus.eu/Get?id=CUMMKSQMFFWKWEEBFRYKWNRL
cd ..

Install kraken

python3.9 -m venv venv3.9
source venv3.9/bin/activate
pip install -U pip setuptools wheel
pip install kraken
pip install -U torch torchvision --extra-index-url https://download.pytorch.org/whl/cu113

Run training

The training is running on a Debian GNU Linux server with Nvidia RTX A5000 GPU.

(venv3.9) stweil@ocr-01:~/src/gitlab/scripta/escriptorium/Juristische_Konsilien_Tuebingen/Transkribus_Exporte$ time nice ketos train -f page -t list.train -e list.eval -o Juristische_Konsilien_Tuebingen -d cuda:0 --lag 20 -r 0.0001 -B 1 -w 0 -s '[1,120,0,1 Cr3,13,32 Do0.1,2 Mp2,2 Cr3,13,32 Do0.1,2 Mp2,2 Cr3,9,64 Do0.1,2 Mp2,2 Cr3,9,64 Do0.1,2 S1(1x0)1,3 Lbx200 Do0.1,2 Lbx200 Do.1,2 Lbx200 Do]'
WARNING:root:Torch version 1.11.0+cu113 has not been tested with coremltools. You may run into unexpected errors. Torch 1.10.2 is the most recent version that has been tested.
[05/24/22 17:38:52] WARNING  alphabet mismatch: chars in training set only: {'‡', '[', '†', 'ꝟ', '=', 'û', 'ꝸ', 'º', 'X', '╒', 'Ü', '’', '♃', '½', ']',   train.py:304
                             'ꝯ'} (not included in accuracy test during training)                                                                                     
                    WARNING  alphabet mismatch: chars in validation set only: {'Ä', 'ù', 'ͦ'} (not trained)                                                train.py:308
Trainer already configured with model summary callbacks: [<class 'pytorch_lightning.callbacks.rich_model_summary.RichModelSummary'>]. Skipping setting a default `ModelSummary` callback.
GPU available: True, used: True
TPU available: False, using: 0 TPU cores
IPU available: False, using: 0 IPUs
HPU available: False, using: 0 HPUs
`Trainer(val_check_interval=1.0)` was configured so validation will run at the end of the training epoch..
[05/24/22 17:38:54] WARNING  Non-encodable sequence ùltis... encountered. Advancing one code point.                                                       codec.py:131
                    WARNING  Non-encodable sequence Är„... encountered. Advancing one code point.                                                         codec.py:131
                    WARNING  Non-encodable sequence ͦ 166... encountered. Advancing one code point.                                                        codec.py:131
LOCAL_RANK: 0 - CUDA_VISIBLE_DEVICES: [0]
┏━━━━┳━━━━━━━━━━━┳━━━━━━━━━━━━━━━━━━━━━━━━━━┳━━━━━━━━┓
┃    ┃ Name      ┃ Type                     ┃ Params ┃
┡━━━━╇━━━━━━━━━━━╇━━━━━━━━━━━━━━━━━━━━━━━━━━╇━━━━━━━━┩
│ 0  │ net       │ MultiParamSequential     │  4.0 M │
│ 1  │ net.C_0   │ ActConv2D                │  1.3 K │
│ 2  │ net.Do_1  │ Dropout                  │      0 │
│ 3  │ net.Mp_2  │ MaxPool                  │      0 │
│ 4  │ net.C_3   │ ActConv2D                │ 40.0 K │
│ 5  │ net.Do_4  │ Dropout                  │      0 │
│ 6  │ net.Mp_5  │ MaxPool                  │      0 │
│ 7  │ net.C_6   │ ActConv2D                │ 55.4 K │
│ 8  │ net.Do_7  │ Dropout                  │      0 │
│ 9  │ net.Mp_8  │ MaxPool                  │      0 │
│ 10 │ net.C_9   │ ActConv2D                │  110 K │
│ 11 │ net.Do_10 │ Dropout                  │      0 │
│ 12 │ net.S_11  │ Reshape                  │      0 │
│ 13 │ net.L_12  │ TransposedSummarizingRNN │  1.9 M │
│ 14 │ net.Do_13 │ Dropout                  │      0 │
│ 15 │ net.L_14  │ TransposedSummarizingRNN │  963 K │
│ 16 │ net.Do_15 │ Dropout                  │      0 │
│ 17 │ net.L_16  │ TransposedSummarizingRNN │  963 K │
│ 18 │ net.Do_17 │ Dropout                  │      0 │
│ 19 │ net.O_18  │ LinSoftmax               │ 48.5 K │
└────┴───────────┴──────────────────────────┴────────┘
Trainable params: 4.0 M                                                                                                                                               
Non-trainable params: 0                                                                                                                                               
Total params: 4.0 M                                                                                                                                                   
Total estimated model params size (MB): 16                                                                                                                            
stage 0/∞  ━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━ 7366/7366 0:00:00 1:05:58 val_accuracy: 0.00000  early_stopping: 0/20 0.00000
stage 1/∞  ━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━ 7366/7366 0:00:00 1:05:57 val_accuracy: 0.27963  early_stopping: 0/20 0.27963
stage 2/∞  ━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━ 7366/7366 0:00:00 1:05:44 val_accuracy: 0.63872  early_stopping: 0/20 0.63872
stage 3/∞  ━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━ 7366/7366 0:00:00 1:06:02 val_accuracy: 0.77172  early_stopping: 0/20 0.77172
stage 4/∞  ━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━ 7366/7366 0:00:00 1:05:59 val_accuracy: 0.82858  early_stopping: 0/20 0.82858
stage 5/∞  ━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━ 7366/7366 0:00:00 1:06:05 val_accuracy: 0.86218  early_stopping: 0/20 0.86218
stage 6/∞  ━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━ 7366/7366 0:00:00 1:05:50 val_accuracy: 0.88192  early_stopping: 0/20 0.88192
stage 7/∞  ━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━ 7366/7366 0:00:00 1:06:01 val_accuracy: 0.89528  early_stopping: 0/20 0.89528
stage 8/∞  ━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━ 7366/7366 0:00:00 1:05:55 val_accuracy: 0.90674  early_stopping: 0/20 0.90674
stage 9/∞  ━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━ 7366/7366 0:00:00 1:06:02 val_accuracy: 0.90923  early_stopping: 0/20 0.90923
stage 10/∞ ━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━ 7366/7366 0:00:00 1:06:03 val_accuracy: 0.91599  early_stopping: 0/20 0.91599
stage 11/∞ ━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━ 7366/7366 0:00:00 1:05:48 val_accuracy: 0.91963  early_stopping: 0/20 0.91963
stage 12/∞ ━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━ 7366/7366 0:00:00 1:05:42 val_accuracy: 0.92259  early_stopping: 0/20 0.92259
stage 13/∞ ━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━ 7366/7366 0:00:00 1:05:35 val_accuracy: 0.92418  early_stopping: 0/20 0.92418
stage 14/∞ ━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━ 7366/7366 0:00:00 1:05:59 val_accuracy: 0.92860  early_stopping: 0/20 0.92860
stage 15/∞ ━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━ 7366/7366 0:00:00 1:05:46 val_accuracy: 0.92850  early_stopping: 1/20 0.92860
stage 16/∞ ━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━ 7366/7366 0:00:00 1:05:48 val_accuracy: 0.92978  early_stopping: 0/20 0.92978
stage 17/∞ ━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━ 7366/7366 0:00:00 1:05:49 val_accuracy: 0.93181  early_stopping: 0/20 0.93181
stage 18/∞ ━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━ 7366/7366 0:00:00 1:06:02 val_accuracy: 0.93016  early_stopping: 1/20 0.93181
stage 19/∞ ━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━ 7366/7366 0:00:00 1:05:55 val_accuracy: 0.93448  early_stopping: 0/20 0.93448
stage 20/∞ ━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━ 7366/7366 0:00:00 1:05:33 val_accuracy: 0.93439  early_stopping: 1/20 0.93448
stage 21/∞ ━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━ 7366/7366 0:00:00 1:05:27 val_accuracy: 0.93539  early_stopping: 0/20 0.93539
stage 22/∞ ━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━ 7366/7366 0:00:00 1:05:33 val_accuracy: 0.93567  early_stopping: 0/20 0.93567
stage 23/∞ ━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━ 7366/7366 0:00:00 1:05:32 val_accuracy: 0.93588  early_stopping: 0/20 0.93588
stage 24/∞ ━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━ 7366/7366 0:00:00 1:05:39 val_accuracy: 0.93722  early_stopping: 0/20 0.93722
stage 25/∞ ━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━ 7366/7366 0:00:00 1:05:41 val_accuracy: 0.93679  early_stopping: 1/20 0.93722
stage 26/∞ ━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━ 7366/7366 0:00:00 1:05:31 val_accuracy: 0.93707  early_stopping: 2/20 0.93722
stage 27/∞ ━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━ 7366/7366 0:00:00 1:05:34 val_accuracy: 0.93847  early_stopping: 0/20 0.93847
stage 28/∞ ━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━ 7366/7366 0:00:00 1:05:32 val_accuracy: 0.93747  early_stopping: 1/20 0.93847
stage 29/∞ ━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━ 7366/7366 0:00:00 1:05:42 val_accuracy: 0.93884  early_stopping: 0/20 0.93884
stage 30/∞ ━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━ 7366/7366 0:00:00 1:05:23 val_accuracy: 0.93884  early_stopping: 1/20 0.93884
stage 31/∞ ━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━ 7366/7366 0:00:00 1:05:21 val_accuracy: 0.93682  early_stopping: 2/20 0.93884
stage 32/∞ ━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━ 7366/7366 0:00:00 1:05:44 val_accuracy: 0.93869  early_stopping: 3/20 0.93884
stage 33/∞ ━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━ 7366/7366 0:00:00 1:05:49 val_accuracy: 0.93657  early_stopping: 4/20 0.93884
stage 34/∞ ━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━ 7366/7366 0:00:00 1:05:37 val_accuracy: 0.93919  early_stopping: 0/20 0.93919
stage 35/∞ ━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━ 7366/7366 0:00:00 1:05:39 val_accuracy: 0.93782  early_stopping: 1/20 0.93919
stage 36/∞ ━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━ 7366/7366 0:00:00 1:04:27 val_accuracy: 0.93875  early_stopping: 2/20 0.93919
stage 37/∞ ━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━ 7366/7366 0:00:00 1:05:49 val_accuracy: 0.94028  early_stopping: 0/20 0.94028
stage 38/∞ ━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━ 7366/7366 0:00:00 1:05:50 val_accuracy: 0.94112  early_stopping: 0/20 0.94112
stage 39/∞ ━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━ 7366/7366 0:00:00 1:06:05 val_accuracy: 0.93978  early_stopping: 1/20 0.94112
stage 40/∞ ━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━ 7366/7366 0:00:00 1:08:45 val_accuracy: 0.93922  early_stopping: 2/20 0.94112
stage 41/∞ ━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━ 7366/7366 0:00:00 1:09:56 val_accuracy: 0.94046  early_stopping: 3/20 0.94112
stage 42/∞ ━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━ 7366/7366 0:00:00 1:07:38 val_accuracy: 0.94186  early_stopping: 0/20 0.94186
stage 43/∞ ━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━ 7366/7366 0:00:00 1:09:34 val_accuracy: 0.93940  early_stopping: 1/20 0.94186
stage 44/∞ ━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━ 7366/7366 0:00:00 1:09:55 val_accuracy: 0.93909  early_stopping: 2/20 0.94186
stage 45/∞ ━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━ 7366/7366 0:00:00 1:10:04 val_accuracy: 0.94040  early_stopping: 3/20 0.94186
stage 46/∞ ━━━━━━━━━━╺━━━━━━━━━━━━━━━━━━━━━━━━━━━━━ 1930/7366 0:51:25 0:18:01 val_accuracy: 0.94040  early_stopping: 3/20 0.94186

Run training with augmentation

Using more than one worker process makes the training much faster, and augmentation should result in a better model.

pip install albumentations
(venv3.9) stweil@ocr-01:~/src/gitlab/scripta/escriptorium/Juristische_Konsilien_Tuebingen/Transkribus_Exporte$ time nice ketos train -f page -t list.train -e list.eval -o Juristische_Konsilien_Tuebingen+ -d cuda:0 --augment --workers 4 -r 0.0001 -B 1 --min-epochs 20 -w 0 -s '[1,120,0,1 Cr3,13,32 Do0.1,2 Mp2,2 Cr3,13,32 Do0.1,2 Mp2,2 Cr3,9,64 Do0.1,2 Mp2,2 Cr3,9,64 Do0.1,2 S1(1x0)1,3 Lbx200 Do0.1,2 Lbx200 Do.1,2 Lbx200 Do]'
WARNING:root:scikit-learn version 1.1.1 is not supported. Minimum required version: 0.17. Maximum required version: 0.19.2. Disabling scikit-learn conversion API.
WARNING:root:Torch version 1.11.0+cu113 has not been tested with coremltools. You may run into unexpected errors. Torch 1.10.2 is the most recent version that has been tested.
[05/26/22 13:43:36] WARNING  alphabet mismatch: chars in training set only: {'’', 'ꝯ', 'û', ']', 'ꝟ', 'ꝸ', 'Ü', '[', '†', '=', '‡', 'º', '½', 'X', '♃', '╒'} (not included in accuracy test during        train.py:304
                             training)                                                                                                                                                                                
                    WARNING  alphabet mismatch: chars in validation set only: {'ͦ', 'Ä', 'ù'} (not trained)                                                                                                train.py:308
Trainer already configured with model summary callbacks: [<class 'pytorch_lightning.callbacks.rich_model_summary.RichModelSummary'>]. Skipping setting a default `ModelSummary` callback.
GPU available: True, used: True
TPU available: False, using: 0 TPU cores
IPU available: False, using: 0 IPUs
HPU available: False, using: 0 HPUs
`Trainer(val_check_interval=1.0)` was configured so validation will run at the end of the training epoch..
[05/26/22 13:43:38] WARNING  Non-encodable sequence ùltis... encountered. Advancing one code point.                                                                                                       codec.py:131
                    WARNING  Non-encodable sequence Är„... encountered. Advancing one code point.                                                                                                         codec.py:131
                    WARNING  Non-encodable sequence ͦ 166... encountered. Advancing one code point.                                                                                                        codec.py:131
LOCAL_RANK: 0 - CUDA_VISIBLE_DEVICES: [0]
┏━━━━┳━━━━━━━━━━━┳━━━━━━━━━━━━━━━━━━━━━━━━━━┳━━━━━━━━┓
┃    ┃ Name      ┃ Type                     ┃ Params ┃
┡━━━━╇━━━━━━━━━━━╇━━━━━━━━━━━━━━━━━━━━━━━━━━╇━━━━━━━━┩
│ 0  │ net       │ MultiParamSequential     │  4.0 M │
│ 1  │ net.C_0   │ ActConv2D                │  1.3 K │
│ 2  │ net.Do_1  │ Dropout                  │      0 │
│ 3  │ net.Mp_2  │ MaxPool                  │      0 │
│ 4  │ net.C_3   │ ActConv2D                │ 40.0 K │
│ 5  │ net.Do_4  │ Dropout                  │      0 │
│ 6  │ net.Mp_5  │ MaxPool                  │      0 │
│ 7  │ net.C_6   │ ActConv2D                │ 55.4 K │
│ 8  │ net.Do_7  │ Dropout                  │      0 │
│ 9  │ net.Mp_8  │ MaxPool                  │      0 │
│ 10 │ net.C_9   │ ActConv2D                │  110 K │
│ 11 │ net.Do_10 │ Dropout                  │      0 │
│ 12 │ net.S_11  │ Reshape                  │      0 │
│ 13 │ net.L_12  │ TransposedSummarizingRNN │  1.9 M │
│ 14 │ net.Do_13 │ Dropout                  │      0 │
│ 15 │ net.L_14  │ TransposedSummarizingRNN │  963 K │
│ 16 │ net.Do_15 │ Dropout                  │      0 │
│ 17 │ net.L_16  │ TransposedSummarizingRNN │  963 K │
│ 18 │ net.Do_17 │ Dropout                  │      0 │
│ 19 │ net.O_18  │ LinSoftmax               │ 48.5 K │
└────┴───────────┴──────────────────────────┴────────┘
Trainable params: 4.0 M                                                                                                                                                                                               
Non-trainable params: 0                                                                                                                                                                                               
Total params: 4.0 M                                                                                                                                                                                                   
Total estimated model params size (MB): 16                                                                                                                                                                            
stage 0/∞  ━━━╸━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━ 673/7366 0:16:39 0:01:43  early_stopping: 0/5 -inf

Increasing the number of workers further reduces the time per epoch even more.

(venv3.9) stweil@ocr-01:~/src/gitlab/scripta/escriptorium/Juristische_Konsilien_Tuebingen/Transkribus_Exporte$ time nice ketos train -f page -t list.train -e list.eval -o Juristische_Konsilien_Tuebingen+ -d cuda:0 
--augment --workers 16 -r 0.0001 -B 1 --min-epochs 20 -w 0 -s '[1,120,0,1 Cr3,13,32 Do0.1,2 Mp2,2 Cr3,13,32 Do0.1,2 Mp2,2 Cr3,9,64 Do0.1,2 Mp2,2 Cr3,9,64 Do0.1,2 S1(1x0)1,3 Lbx200 Do0.1,2 Lbx200 Do.1,2 Lbx200 Do]'
WARNING:root:scikit-learn version 1.1.1 is not supported. Minimum required version: 0.17. Maximum required version: 0.19.2. Disabling scikit-learn conversion API.
WARNING:root:Torch version 1.11.0+cu113 has not been tested with coremltools. You may run into unexpected errors. Torch 1.10.2 is the most recent version that has been tested.
[05/26/22 13:49:52] WARNING  alphabet mismatch: chars in training set only: {'ꝯ', '½', 'ꝟ', '╒', '=', '[', 'û', '’', 'Ü', '†', '♃', 'ꝸ', '‡', 'X', 'º', ']'} (not included in accuracy test during        train.py:304
                             training)                                                                                                                                                                                
                    WARNING  alphabet mismatch: chars in validation set only: {'Ä', 'ù', 'ͦ'} (not trained)                                                                                                train.py:308
Trainer already configured with model summary callbacks: [<class 'pytorch_lightning.callbacks.rich_model_summary.RichModelSummary'>]. Skipping setting a default `ModelSummary` callback.
GPU available: True, used: True
TPU available: False, using: 0 TPU cores
IPU available: False, using: 0 IPUs
HPU available: False, using: 0 HPUs
`Trainer(val_check_interval=1.0)` was configured so validation will run at the end of the training epoch..
[05/26/22 13:49:55] WARNING  Non-encodable sequence ùltis... encountered. Advancing one code point.                                                                                                       codec.py:131
                    WARNING  Non-encodable sequence Är„... encountered. Advancing one code point.                                                                                                         codec.py:131
                    WARNING  Non-encodable sequence ͦ 166... encountered. Advancing one code point.                                                                                                        codec.py:131
LOCAL_RANK: 0 - CUDA_VISIBLE_DEVICES: [0]
┏━━━━┳━━━━━━━━━━━┳━━━━━━━━━━━━━━━━━━━━━━━━━━┳━━━━━━━━┓
┃    ┃ Name      ┃ Type                     ┃ Params ┃
┡━━━━╇━━━━━━━━━━━╇━━━━━━━━━━━━━━━━━━━━━━━━━━╇━━━━━━━━┩
│ 0  │ net       │ MultiParamSequential     │  4.0 M │
│ 1  │ net.C_0   │ ActConv2D                │  1.3 K │
│ 2  │ net.Do_1  │ Dropout                  │      0 │
│ 3  │ net.Mp_2  │ MaxPool                  │      0 │
│ 4  │ net.C_3   │ ActConv2D                │ 40.0 K │
│ 5  │ net.Do_4  │ Dropout                  │      0 │
│ 6  │ net.Mp_5  │ MaxPool                  │      0 │
│ 7  │ net.C_6   │ ActConv2D                │ 55.4 K │
│ 8  │ net.Do_7  │ Dropout                  │      0 │
│ 9  │ net.Mp_8  │ MaxPool                  │      0 │
│ 10 │ net.C_9   │ ActConv2D                │  110 K │
│ 11 │ net.Do_10 │ Dropout                  │      0 │
│ 12 │ net.S_11  │ Reshape                  │      0 │
│ 13 │ net.L_12  │ TransposedSummarizingRNN │  1.9 M │
│ 14 │ net.Do_13 │ Dropout                  │      0 │
│ 15 │ net.L_14  │ TransposedSummarizingRNN │  963 K │
│ 16 │ net.Do_15 │ Dropout                  │      0 │
│ 17 │ net.L_16  │ TransposedSummarizingRNN │  963 K │
│ 18 │ net.Do_17 │ Dropout                  │      0 │
│ 19 │ net.O_18  │ LinSoftmax               │ 48.5 K │
└────┴───────────┴──────────────────────────┴────────┘
Trainable params: 4.0 M                                                                                                                                                                                               
Non-trainable params: 0                                                                                                                                                                                               
Total params: 4.0 M                                                                                                                                                                                                   
Total estimated model params size (MB): 16                                                                                                                                                                            
stage 0/∞  ━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━ 7366/7366 0:00:00 0:07:03 val_accuracy: 0.00000  early_stopping: 0/5 0.00000
stage 1/∞  ━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━ 7366/7366 0:00:00 0:06:24 val_accuracy: 0.20782  early_stopping: 0/5 0.20782
stage 2/∞  ━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━ 7366/7366 0:00:00 0:06:21 val_accuracy: 0.57978  early_stopping: 0/5 0.57978
stage 3/∞  ━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━ 7366/7366 0:00:00 0:06:23 val_accuracy: 0.71006  early_stopping: 0/5 0.71006
stage 4/∞  ━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━ 7366/7366 0:00:00 0:06:20 val_accuracy: 0.78446  early_stopping: 0/5 0.78446
stage 5/∞  ━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━ 7366/7366 0:00:00 0:06:23 val_accuracy: 0.82139  early_stopping: 0/5 0.82139
stage 6/∞  ━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━ 7366/7366 0:00:00 0:06:57 val_accuracy: 0.83555  early_stopping: 0/5 0.83555
stage 7/∞  ━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━ 7366/7366 0:00:00 0:06:25 val_accuracy: 0.86224  early_stopping: 0/5 0.86224
stage 8/∞  ━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━ 7366/7366 0:00:00 0:06:21 val_accuracy: 0.86741  early_stopping: 0/5 0.86741
stage 9/∞  ━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━ 7366/7366 0:00:00 0:06:21 val_accuracy: 0.88089  early_stopping: 0/5 0.88089
stage 10/∞ ━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━ 7366/7366 0:00:00 0:06:22 val_accuracy: 0.89192  early_stopping: 0/5 0.89192
stage 11/∞ ━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━ 7366/7366 0:00:00 0:06:23 val_accuracy: 0.87604  early_stopping: 1/5 0.89192
stage 12/∞ ━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━ 7366/7366 0:00:00 0:06:23 val_accuracy: 0.89733  early_stopping: 0/5 0.89733
stage 13/∞ ━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━ 7366/7366 0:00:00 0:06:23 val_accuracy: 0.90216  early_stopping: 0/5 0.90216
stage 14/∞ ━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━ 7366/7366 0:00:00 0:06:26 val_accuracy: 0.90222  early_stopping: 0/5 0.90222
stage 15/∞ ━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━ 7366/7366 0:00:00 0:06:24 val_accuracy: 0.90182  early_stopping: 1/5 0.90222
stage 16/∞ ━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━ 7366/7366 0:00:00 0:06:48 val_accuracy: 0.90608  early_stopping: 0/5 0.90608
stage 17/∞ ━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━ 7366/7366 0:00:00 0:07:44 val_accuracy: 0.91259  early_stopping: 0/5 0.91259
stage 18/∞ ━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━ 7366/7366 0:00:00 0:06:25 val_accuracy: 0.90496  early_stopping: 1/5 0.91259
stage 19/∞ ━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━ 7366/7366 0:00:00 0:06:25 val_accuracy: 0.91141  early_stopping: 2/5 0.91259
stage 20/∞ ━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━ 7366/7366 0:00:00 0:06:25 val_accuracy: 0.90126  early_stopping: 3/5 0.91259
stage 21/∞ ━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━ 7366/7366 0:00:00 0:06:24 val_accuracy: 0.91119  early_stopping: 4/5 0.91259
stage 22/∞ ━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━ 7366/7366 0:00:00 0:06:24 val_accuracy: 0.91100  early_stopping: 5/5 0.91259
Moving best model Juristische_Konsilien_Tuebingen+_17.mlmodel (0.9125926494598389) to Juristische_Konsilien_Tuebingen+_best.mlmodel

real    151m8.293s
user    2131m47.894s
sys     1193m39.862s

Resume the training:

(venv3.9) stweil@ocr-01:~/src/gitlab/scripta/escriptorium/Juristische_Konsilien_Tuebingen/Transkribus_Exporte$ time nice ketos train -i Juristische_Konsilien_Tuebingen+_17.mlmodel -f page -t list.train -e list.eval -o Juristische_Konsilien_Tuebingen+ -d cuda:0 --augment --workers 16 -r 0.0001 -B 1 --min-epochs 200 --lag 10 -w 0 -s '[1,120,0,1 Cr3,13,32 Do0.1,2 Mp2,2 Cr3,13,32 Do0.1,2 Mp2,2 Cr3,9,64 Do0.1,2 Mp2,2 Cr3,9,64 Do0
.1,2 S1(1x0)1,3 Lbx200 Do0.1,2 Lbx200 Do.1,2 Lbx200 Do]'
WARNING:root:scikit-learn version 1.1.1 is not supported. Minimum required version: 0.17. Maximum required version: 0.19.2. Disabling scikit-learn conversion API.
WARNING:root:Torch version 1.11.0+cu113 has not been tested with coremltools. You may run into unexpected errors. Torch 1.10.2 is the most recent version that has been tested.
[05/26/22 16:56:31] WARNING  alphabet mismatch: chars in training set only: {'Ü', '‡', 'û', '♃', '╒', 'ꝯ', 'º', 'ꝸ', '†', '=', 'X', '[', '½', ']', 'ꝟ', '’'} (not included in accuracy test during        train.py:304
                             training)                                                                                                                                                                                
Trainer already configured with model summary callbacks: [<class 'pytorch_lightning.callbacks.rich_model_summary.RichModelSummary'>]. Skipping setting a default `ModelSummary` callback.
GPU available: True, used: True
TPU available: False, using: 0 TPU cores
IPU available: False, using: 0 IPUs
HPU available: False, using: 0 HPUs
`Trainer(val_check_interval=1.0)` was configured so validation will run at the end of the training epoch..
LOCAL_RANK: 0 - CUDA_VISIBLE_DEVICES: [0]
┏━━━━┳━━━━━━━━━━━┳━━━━━━━━━━━━━━━━━━━━━━━━━━┳━━━━━━━━┓
┃    ┃ Name      ┃ Type                     ┃ Params ┃
┡━━━━╇━━━━━━━━━━━╇━━━━━━━━━━━━━━━━━━━━━━━━━━╇━━━━━━━━┩
│ 0  │ net       │ MultiParamSequential     │  4.0 M │
│ 1  │ net.C_0   │ ActConv2D                │  1.3 K │
│ 2  │ net.Do_1  │ Dropout                  │      0 │
│ 3  │ net.Mp_2  │ MaxPool                  │      0 │
│ 4  │ net.C_3   │ ActConv2D                │ 40.0 K │
│ 5  │ net.Do_4  │ Dropout                  │      0 │
│ 6  │ net.Mp_5  │ MaxPool                  │      0 │
│ 7  │ net.C_6   │ ActConv2D                │ 55.4 K │
│ 8  │ net.Do_7  │ Dropout                  │      0 │
│ 9  │ net.Mp_8  │ MaxPool                  │      0 │
│ 10 │ net.C_9   │ ActConv2D                │  110 K │
│ 11 │ net.Do_10 │ Dropout                  │      0 │
│ 12 │ net.S_11  │ Reshape                  │      0 │
│ 13 │ net.L_12  │ TransposedSummarizingRNN │  1.9 M │
│ 14 │ net.Do_13 │ Dropout                  │      0 │
│ 15 │ net.L_14  │ TransposedSummarizingRNN │  963 K │
│ 16 │ net.Do_15 │ Dropout                  │      0 │
│ 17 │ net.L_16  │ TransposedSummarizingRNN │  963 K │
│ 18 │ net.Do_17 │ Dropout                  │      0 │
│ 19 │ net.O_18  │ LinSoftmax               │ 48.5 K │
└────┴───────────┴──────────────────────────┴────────┘
Trainable params: 4.0 M                                                                                                                                                                                               
Non-trainable params: 0                                                                                                                                                                                               
Total params: 4.0 M                                                                                                                                                                                                   
Total estimated model params size (MB): 16                                                                                                                                                                            
stage 0/∞  ━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━ 7366/7366 0:00:00 0:07:05 val_accuracy: 0.91661  early_stopping: 0/10 0.91661
stage 1/∞  ━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━ 7366/7366 0:00:00 0:06:23 val_accuracy: 0.89668  early_stopping: 1/10 0.91661
stage 2/∞  ━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━ 7366/7366 0:00:00 0:06:23 val_accuracy: 0.91540  early_stopping: 2/10 0.91661
stage 3/∞  ━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━ 7366/7366 0:00:00 0:06:21 val_accuracy: 0.92044  early_stopping: 0/10 0.92044
stage 4/∞  ━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━ 7366/7366 0:00:00 0:06:19 val_accuracy: 0.90824  early_stopping: 1/10 0.92044
stage 5/∞  ━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━ 7366/7366 0:00:00 0:06:24 val_accuracy: 0.91755  early_stopping: 2/10 0.92044
stage 6/∞  ━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━ 7366/7366 0:00:00 0:06:24 val_accuracy: 0.90014  early_stopping: 3/10 0.92044
stage 7/∞  ━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━ 7366/7366 0:00:00 0:06:26 val_accuracy: 0.91244  early_stopping: 4/10 0.92044
stage 8/∞  ━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━ 7366/7366 0:00:00 0:06:26 val_accuracy: 0.91060  early_stopping: 5/10 0.92044
stage 9/∞  ━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━ 7366/7366 0:00:00 0:07:02 val_accuracy: 0.91213  early_stopping: 6/10 0.92044
stage 10/∞ ━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━ 7366/7366 0:00:00 0:06:24 val_accuracy: 0.90397  early_stopping: 7/10 0.92044
stage 11/∞ ━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━ 7366/7366 0:00:00 0:06:23 val_accuracy: 0.92128  early_stopping: 0/10 0.92128
stage 12/∞ ━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━ 7366/7366 0:00:00 0:06:29 val_accuracy: 0.90992  early_stopping: 1/10 0.92128
stage 13/∞ ━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━ 7366/7366 0:00:00 0:06:26 val_accuracy: 0.92181  early_stopping: 0/10 0.92181
stage 14/∞ ━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━ 7366/7366 0:00:00 0:06:23 val_accuracy: 0.91278  early_stopping: 1/10 0.92181
stage 15/∞ ━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━ 7366/7366 0:00:00 0:06:24 val_accuracy: 0.92343  early_stopping: 0/10 0.92343
stage 16/∞ ━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━ 7366/7366 0:00:00 0:06:21 val_accuracy: 0.92577  early_stopping: 0/10 0.92577
stage 17/∞ ━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━ 7366/7366 0:00:00 0:06:21 val_accuracy: 0.92166  early_stopping: 1/10 0.92577
stage 18/∞ ━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━ 7366/7366 0:00:00 0:06:27 val_accuracy: 0.92197  early_stopping: 2/10 0.92577
stage 19/∞ ━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━ 7366/7366 0:00:00 0:06:25 val_accuracy: 0.92240  early_stopping: 3/10 0.92577
stage 20/∞ ━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━ 7366/7366 0:00:00 0:06:57 val_accuracy: 0.91297  early_stopping: 4/10 0.92577
stage 21/∞ ━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━ 7366/7366 0:00:00 0:06:24 val_accuracy: 0.92078  early_stopping: 5/10 0.92577
stage 22/∞ ━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━ 7366/7366 0:00:00 0:06:23 val_accuracy: 0.92418  early_stopping: 6/10 0.92577
stage 23/∞ ━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━ 7366/7366 0:00:00 0:06:22 val_accuracy: 0.92611  early_stopping: 0/10 0.92611
stage 24/∞ ━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━ 7366/7366 0:00:00 0:06:26 val_accuracy: 0.91758  early_stopping: 1/10 0.92611
stage 25/∞ ━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━ 7366/7366 0:00:00 0:06:25 val_accuracy: 0.92412  early_stopping: 2/10 0.92611
stage 26/∞ ━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━ 7366/7366 0:00:00 0:06:25 val_accuracy: 0.91991  early_stopping: 3/10 0.92611
stage 27/∞ ━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━ 7366/7366 0:00:00 0:06:21 val_accuracy: 0.92287  early_stopping: 4/10 0.92611
stage 28/∞ ━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━ 7366/7366 0:00:00 0:06:22 val_accuracy: 0.92069  early_stopping: 5/10 0.92611
stage 29/∞ ━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━ 7366/7366 0:00:00 0:06:23 val_accuracy: 0.92767  early_stopping: 0/10 0.92767
stage 30/∞ ━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━ 7366/7366 0:00:00 0:06:53 val_accuracy: 0.92234  early_stopping: 1/10 0.92767
stage 31/∞ ━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━ 7366/7366 0:00:00 0:07:41 val_accuracy: 0.92848  early_stopping: 0/10 0.92848
stage 32/∞ ━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━ 7366/7366 0:00:00 0:06:25 val_accuracy: 0.91652  early_stopping: 1/10 0.92848
stage 33/∞ ━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━ 7366/7366 0:00:00 0:06:27 val_accuracy: 0.92704  early_stopping: 2/10 0.92848
stage 34/∞ ━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━ 7366/7366 0:00:00 0:06:25 val_accuracy: 0.93280  early_stopping: 0/10 0.93280
stage 35/∞ ━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━ 7366/7366 0:00:00 0:06:24 val_accuracy: 0.91870  early_stopping: 1/10 0.93280
stage 36/∞ ━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━ 7366/7366 0:00:00 0:06:23 val_accuracy: 0.92726  early_stopping: 2/10 0.93280
stage 37/∞ ━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━ 7366/7366 0:00:00 0:06:23 val_accuracy: 0.92041  early_stopping: 3/10 0.93280
stage 38/∞ ━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━ 7366/7366 0:00:00 0:06:26 val_accuracy: 0.92673  early_stopping: 4/10 0.93280
stage 39/∞ ━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━ 7366/7366 0:00:00 0:06:23 val_accuracy: 0.92894  early_stopping: 5/10 0.93280
stage 40/∞ ━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━ 7366/7366 0:00:00 0:06:27 val_accuracy: 0.92633  early_stopping: 6/10 0.93280
stage 41/∞ ━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━ 7366/7366 0:00:00 0:07:03 val_accuracy: 0.92695  early_stopping: 7/10 0.93280
stage 42/∞ ━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━ 7366/7366 0:00:00 0:06:27 val_accuracy: 0.91484  early_stopping: 8/10 0.93280
stage 43/∞ ━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━ 7366/7366 0:00:00 0:06:23 val_accuracy: 0.92997  early_stopping: 9/10 0.93280
stage 44/∞ ━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━ 7366/7366 0:00:00 0:06:24 val_accuracy: 0.93296  early_stopping: 0/10 0.93296
stage 45/∞ ━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━ 7366/7366 0:00:00 0:06:24 val_accuracy: 0.91892  early_stopping: 1/10 0.93296
stage 46/∞ ━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━ 7366/7366 0:00:00 0:06:25 val_accuracy: 0.91798  early_stopping: 2/10 0.93296
stage 47/∞ ━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━ 7366/7366 0:00:00 0:06:24 val_accuracy: 0.92888  early_stopping: 3/10 0.93296
stage 48/∞ ━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━ 7366/7366 0:00:00 0:06:22 val_accuracy: 0.93000  early_stopping: 4/10 0.93296
stage 49/∞ ━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━ 7366/7366 0:00:00 0:06:23 val_accuracy: 0.92135  early_stopping: 5/10 0.93296
stage 50/∞ ━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━ 7366/7366 0:00:00 0:06:24 val_accuracy: 0.92256  early_stopping: 6/10 0.93296
stage 51/∞ ━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━ 7366/7366 0:00:00 0:06:25 val_accuracy: 0.92975  early_stopping: 7/10 0.93296
stage 52/∞ ━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━ 7366/7366 0:00:00 0:07:03 val_accuracy: 0.92468  early_stopping: 8/10 0.93296
stage 53/∞ ━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━ 7366/7366 0:00:00 0:06:29 val_accuracy: 0.92589  early_stopping: 9/10 0.93296
stage 54/∞ ━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━ 7366/7366 0:00:00 0:06:25 val_accuracy: 0.92589  early_stopping: 9/10 0.93296Trainer was signaled to stop but required minimum epochs (200) or minimum steps (None) has not been met. Training will continue...
stage 54/∞ ━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━ 7366/7366 0:00:00 0:06:25 val_accuracy: 0.92265  early_stopping: 10/10 0.93296
stage 55/∞ ━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━ 7366/7366 0:00:00 0:06:23 val_accuracy: 0.92265  early_stopping: 10/10 0.93296Trainer was signaled to stop but required minimum epochs (200) or minimum steps (None) has not been met. Training will continue...
stage 55/∞ ━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━ 7366/7366 0:00:00 0:06:23 val_accuracy: 0.92901  early_stopping: 11/10 0.93296
stage 56/∞ ━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━ 7366/7366 0:00:00 0:06:24 val_accuracy: 0.92901  early_stopping: 11/10 0.93296Trainer was signaled to stop but required minimum epochs (200) or minimum steps (None) has not been met. Training will continue...
stage 56/∞ ━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━ 7366/7366 0:00:00 0:06:24 val_accuracy: 0.92493  early_stopping: 12/10 0.93296
stage 57/∞ ━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━ 7366/7366 0:00:00 0:06:22 val_accuracy: 0.92493  early_stopping: 12/10 0.93296Trainer was signaled to stop but required minimum epochs (200) or minimum steps (None) has not been met. Training will continue...
stage 57/∞ ━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━ 7366/7366 0:00:00 0:06:22 val_accuracy: 0.92159  early_stopping: 13/10 0.93296
stage 58/∞ ━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━ 7366/7366 0:00:00 0:06:25 val_accuracy: 0.92159  early_stopping: 13/10 0.93296Trainer was signaled to stop but required minimum epochs (200) or minimum steps (None) has not been met. Training will continue...
stage 58/∞ ━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━ 7366/7366 0:00:00 0:06:25 val_accuracy: 0.92097  early_stopping: 14/10 0.93296
stage 59/∞ ━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━ 7366/7366 0:00:00 0:06:21 val_accuracy: 0.92097  early_stopping: 14/10 0.93296Trainer was signaled to stop but required minimum epochs (200) or minimum steps (None) has not been met. Training will continue...
stage 59/∞ ━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━ 7366/7366 0:00:00 0:06:21 val_accuracy: 0.92620  early_stopping: 15/10 0.93296
stage 60/∞ ━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━ 7366/7366 0:00:00 0:06:24 val_accuracy: 0.93495  early_stopping: 0/10 0.93495
stage 61/∞ ━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━ 7366/7366 0:00:00 0:06:25 val_accuracy: 0.92433  early_stopping: 1/10 0.93495
stage 62/∞ ━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━ 7366/7366 0:00:00 0:06:24 val_accuracy: 0.92957  early_stopping: 2/10 0.93495
stage 63/∞ ━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━ 7366/7366 0:00:00 0:06:59 val_accuracy: 0.92882  early_stopping: 3/10 0.93495
stage 64/∞ ━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━ 7366/7366 0:00:00 0:06:26 val_accuracy: 0.92863  early_stopping: 4/10 0.93495
stage 65/∞ ━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━ 7366/7366 0:00:00 0:06:25 val_accuracy: 0.93377  early_stopping: 5/10 0.93495
stage 66/∞ ━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━ 7366/7366 0:00:00 0:06:22 val_accuracy: 0.93218  early_stopping: 6/10 0.93495
stage 67/∞ ━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━ 7366/7366 0:00:00 0:06:23 val_accuracy: 0.93744  early_stopping: 0/10 0.93744
stage 68/∞ ━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━ 7366/7366 0:00:00 0:06:23 val_accuracy: 0.92683  early_stopping: 1/10 0.93744
stage 69/∞ ━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━ 7366/7366 0:00:00 0:06:23 val_accuracy: 0.93293  early_stopping: 2/10 0.93744
stage 70/∞ ━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━ 7366/7366 0:00:00 0:06:25 val_accuracy: 0.93520  early_stopping: 3/10 0.93744
stage 71/∞ ━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━ 7366/7366 0:00:00 0:06:27 val_accuracy: 0.93246  early_stopping: 4/10 0.93744
stage 72/∞ ━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━ 7366/7366 0:00:00 0:06:24 val_accuracy: 0.92340  early_stopping: 5/10 0.93744
stage 73/∞ ━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━ 7366/7366 0:00:00 0:07:10 val_accuracy: 0.93548  early_stopping: 6/10 0.93744
stage 74/∞ ━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━ 7366/7366 0:00:00 0:07:16 val_accuracy: 0.92944  early_stopping: 7/10 0.93744
stage 75/∞ ━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━ 7366/7366 0:00:00 0:06:25 val_accuracy: 0.93078  early_stopping: 8/10 0.93744
stage 76/∞ ━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━ 7366/7366 0:00:00 0:06:23 val_accuracy: 0.93094  early_stopping: 9/10 0.93744
stage 77/∞ ━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━ 7366/7366 0:00:00 0:06:26 val_accuracy: 0.93094  early_stopping: 9/10 0.93744Trainer was signaled to stop but required minimum epochs (200) or minimum steps (None) has not been met. Training will continue...
stage 77/∞ ━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━ 7366/7366 0:00:00 0:06:26 val_accuracy: 0.92941  early_stopping: 10/10 0.93744
stage 78/∞ ━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━ 7366/7366 0:00:00 0:06:24 val_accuracy: 0.92941  early_stopping: 10/10 0.93744Trainer was signaled to stop but required minimum epochs (200) or minimum steps (None) has not been met. Training will continue...
stage 78/∞ ━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━ 7366/7366 0:00:00 0:06:24 val_accuracy: 0.92044  early_stopping: 11/10 0.93744
stage 79/∞ ━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━ 7366/7366 0:00:00 0:06:25 val_accuracy: 0.92044  early_stopping: 11/10 0.93744Trainer was signaled to stop but required minimum epochs (200) or minimum steps (None) has not been met. Training will continue...
stage 79/∞ ━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━ 7366/7366 0:00:00 0:06:25 val_accuracy: 0.93489  early_stopping: 12/10 0.93744
stage 80/∞ ━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━ 7366/7366 0:00:00 0:06:26 val_accuracy: 0.93489  early_stopping: 12/10 0.93744Trainer was signaled to stop but required minimum epochs (200) or minimum steps (None) has not been met. Training will continue...
stage 80/∞ ━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━ 7366/7366 0:00:00 0:06:26 val_accuracy: 0.92359  early_stopping: 13/10 0.93744
stage 81/∞ ━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━ 7366/7366 0:00:00 0:06:25 val_accuracy: 0.93751  early_stopping: 0/10 0.93751
stage 82/∞ ━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━ 7366/7366 0:00:00 0:06:24 val_accuracy: 0.92848  early_stopping: 1/10 0.93751
stage 83/∞ ━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━ 7366/7366 0:00:00 0:06:27 val_accuracy: 0.93937  early_stopping: 0/10 0.93937
stage 84/∞ ━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━ 7366/7366 0:00:00 0:07:00 val_accuracy: 0.92841  early_stopping: 1/10 0.93937
stage 85/∞ ━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━ 7366/7366 0:00:00 0:06:26 val_accuracy: 0.93716  early_stopping: 2/10 0.93937
stage 86/∞ ━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━ 7366/7366 0:00:00 0:06:23 val_accuracy: 0.92876  early_stopping: 3/10 0.93937
stage 87/∞ ━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━ 7366/7366 0:00:00 0:06:23 val_accuracy: 0.92412  early_stopping: 4/10 0.93937
stage 88/∞ ━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━ 7366/7366 0:00:00 0:06:24 val_accuracy: 0.93461  early_stopping: 5/10 0.93937
stage 89/∞ ━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━ 7366/7366 0:00:00 0:06:26 val_accuracy: 0.92627  early_stopping: 6/10 0.93937
stage 90/∞ ━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━ 7366/7366 0:00:00 0:06:26 val_accuracy: 0.93081  early_stopping: 7/10 0.93937
stage 91/∞ ━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━ 7366/7366 0:00:00 0:06:24 val_accuracy: 0.93542  early_stopping: 8/10 0.93937
stage 92/∞ ━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━ 7366/7366 0:00:00 0:06:21 val_accuracy: 0.93000  early_stopping: 9/10 0.93937
stage 93/∞ ━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━ 7366/7366 0:00:00 0:06:29 val_accuracy: 0.93000  early_stopping: 9/10 0.93937Trainer was signaled to stop but required minimum epochs (200) or minimum steps (None) has not been met. Training will continue...
stage 93/∞ ━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━ 7366/7366 0:00:00 0:06:29 val_accuracy: 0.93667  early_stopping: 10/10 0.93937
stage 94/∞ ━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━ 7366/7366 0:00:00 0:06:29 val_accuracy: 0.93667  early_stopping: 10/10 0.93937Trainer was signaled to stop but required minimum epochs (200) or minimum steps (None) has not been met. Training will continue...
stage 94/∞ ━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━ 7366/7366 0:00:00 0:06:29 val_accuracy: 0.93150  early_stopping: 11/10 0.93937
stage 95/∞ ━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━ 7366/7366 0:00:00 0:07:03 val_accuracy: 0.93150  early_stopping: 11/10 0.93937Trainer was signaled to stop but required minimum epochs (200) or minimum steps (None) has not been met. Training will continue...
stage 95/∞ ━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━ 7366/7366 0:00:00 0:07:03 val_accuracy: 0.93486  early_stopping: 12/10 0.93937
stage 96/∞ ━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━ 7366/7366 0:00:00 0:06:24 val_accuracy: 0.93486  early_stopping: 12/10 0.93937Trainer was signaled to stop but required minimum epochs (200) or minimum steps (None) has not been met. Training will continue...
stage 96/∞ ━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━ 7366/7366 0:00:00 0:06:24 val_accuracy: 0.93673  early_stopping: 13/10 0.93937
stage 97/∞ ━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━ 7366/7366 0:00:00 0:06:27 val_accuracy: 0.93673  early_stopping: 13/10 0.93937Trainer was signaled to stop but required minimum epochs (200) or minimum steps (None) has not been met. Training will continue...
stage 97/∞ ━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━ 7366/7366 0:00:00 0:06:27 val_accuracy: 0.93623  early_stopping: 14/10 0.93937
stage 98/∞ ━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━ 7366/7366 0:00:00 0:06:28 val_accuracy: 0.93623  early_stopping: 14/10 0.93937Trainer was signaled to stop but required minimum epochs (200) or minimum steps (None) has not been met. Training will continue...
stage 98/∞ ━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━ 7366/7366 0:00:00 0:06:28 val_accuracy: 0.91889  early_stopping: 15/10 0.93937
stage 99/∞ ━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━ 7366/7366 0:00:00 0:06:22 val_accuracy: 0.91889  early_stopping: 15/10 0.93937Trainer was signaled to stop but required minimum epochs (200) or minimum steps (None) has not been met. Training will continue...
stage 99/∞ ━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━ 7366/7366 0:00:00 0:06:22 val_accuracy: 0.93053  early_stopping: 16/10 0.93937
stage 100/∞ ━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━ 7366/7366 0:00:00 0:06:23 val_accuracy: 0.93053  early_stopping: 16/10 0.93937Trainer was signaled to stop but required minimum epochs (200) or minimum steps (None) has not been met. Training will continue...
stage 100/∞ ━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━ 7366/7366 0:00:00 0:06:23 val_accuracy: 0.93377  early_stopping: 17/10 0.93937
stage 101/∞ ━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━ 7366/7366 0:00:00 0:06:25 val_accuracy: 0.93377  early_stopping: 17/10 0.93937Trainer was signaled to stop but required minimum epochs (200) or minimum steps (None) has not been met. Training will continue...
stage 101/∞ ━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━ 7366/7366 0:00:00 0:06:25 val_accuracy: 0.93196  early_stopping: 18/10 0.93937
stage 102/∞ ━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━ 7366/7366 0:00:00 0:06:25 val_accuracy: 0.93196  early_stopping: 18/10 0.93937Trainer was signaled to stop but required minimum epochs (200) or minimum steps (None) has not been met. Training will continue...
stage 102/∞ ━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━ 7366/7366 0:00:00 0:06:25 val_accuracy: 0.93358  early_stopping: 19/10 0.93937
stage 103/∞ ━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━ 7366/7366 0:00:00 0:06:22 val_accuracy: 0.93358  early_stopping: 19/10 0.93937Trainer was signaled to stop but required minimum epochs (200) or minimum steps (None) has not been met. Training will continue...
stage 103/∞ ━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━ 7366/7366 0:00:00 0:06:22 val_accuracy: 0.93595  early_stopping: 20/10 0.93937
stage 104/∞ ━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━ 7366/7366 0:00:00 0:06:25 val_accuracy: 0.93595  early_stopping: 20/10 0.93937Trainer was signaled to stop but required minimum epochs (200) or minimum steps (None) has not been met. Training will continue...
stage 104/∞ ━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━ 7366/7366 0:00:00 0:06:25 val_accuracy: 0.92745  early_stopping: 21/10 0.93937
stage 105/∞ ━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━ 7366/7366 0:00:00 0:06:45 val_accuracy: 0.92745  early_stopping: 21/10 0.93937Trainer was signaled to stop but required minimum epochs (200) or minimum steps (None) has not been met. Training will continue...
stage 105/∞ ━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━ 7366/7366 0:00:00 0:06:45 val_accuracy: 0.93128  early_stopping: 22/10 0.93937
stage 106/∞ ━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━ 7366/7366 0:00:00 0:07:50 val_accuracy: 0.93128  early_stopping: 22/10 0.93937Trainer was signaled to stop but required minimum epochs (200) or minimum steps (None) has not been met. Training will continue...
stage 106/∞ ━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━ 7366/7366 0:00:00 0:07:50 val_accuracy: 0.92026  early_stopping: 23/10 0.93937
stage 107/∞ ━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━ 7366/7366 0:00:00 0:06:22 val_accuracy: 0.92026  early_stopping: 23/10 0.93937Trainer was signaled to stop but required minimum epochs (200) or minimum steps (None) has not been met. Training will continue...
stage 107/∞ ━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━ 7366/7366 0:00:00 0:06:22 val_accuracy: 0.93916  early_stopping: 24/10 0.93937
stage 108/∞ ━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━ 7366/7366 0:00:00 0:06:26 val_accuracy: 0.93916  early_stopping: 24/10 0.93937Trainer was signaled to stop but required minimum epochs (200) or minimum steps (None) has not been met. Training will continue...
stage 108/∞ ━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━ 7366/7366 0:00:00 0:06:26 val_accuracy: 0.93274  early_stopping: 25/10 0.93937
stage 109/∞ ━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━ 7366/7366 0:00:00 0:06:25 val_accuracy: 0.93274  early_stopping: 25/10 0.93937Trainer was signaled to stop but required minimum epochs (200) or minimum steps (None) has not been met. Training will continue...
stage 109/∞ ━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━ 7366/7366 0:00:00 0:06:25 val_accuracy: 0.92549  early_stopping: 26/10 0.93937
stage 110/∞ ━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━ 7366/7366 0:00:00 0:06:23 val_accuracy: 0.93635  early_stopping: 27/10 0.93937
stage 111/∞ ━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━ 7366/7366 0:00:00 0:06:23 val_accuracy: 0.93589  early_stopping: 28/10 0.93937
stage 112/∞ ━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━ 7366/7366 0:00:00 0:06:23 val_accuracy: 0.92695  early_stopping: 29/10 0.93937
stage 113/∞ ━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━ 7366/7366 0:00:00 0:06:25 val_accuracy: 0.93414  early_stopping: 30/10 0.93937
stage 114/∞ ━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━ 7366/7366 0:00:00 0:06:22 val_accuracy: 0.92760  early_stopping: 31/10 0.93937
stage 115/∞ ━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━ 7366/7366 0:00:00 0:06:23 val_accuracy: 0.93772  early_stopping: 32/10 0.93937
stage 116/∞ ━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━ 7366/7366 0:00:00 0:07:10 val_accuracy: 0.93368  early_stopping: 33/10 0.93937
stage 117/∞ ━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━ 7366/7366 0:00:00 0:06:25 val_accuracy: 0.93947  early_stopping: 0/10 0.93947
stage 118/∞ ━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━ 7366/7366 0:00:00 0:06:21 val_accuracy: 0.93816  early_stopping: 1/10 0.93947
stage 119/∞ ━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━ 7366/7366 0:00:00 0:06:24 val_accuracy: 0.93732  early_stopping: 2/10 0.93947
stage 120/∞ ━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━ 7366/7366 0:00:00 0:06:21 val_accuracy: 0.92882  early_stopping: 3/10 0.93947
stage 121/∞ ━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━ 7366/7366 0:00:00 0:06:27 val_accuracy: 0.93084  early_stopping: 4/10 0.93947
stage 122/∞ ━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━ 7366/7366 0:00:00 0:06:26 val_accuracy: 0.93284  early_stopping: 5/10 0.93947
stage 123/∞ ━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━ 7366/7366 0:00:00 0:06:23 val_accuracy: 0.93271  early_stopping: 6/10 0.93947
stage 124/∞ ━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━ 7366/7366 0:00:00 0:06:27 val_accuracy: 0.93389  early_stopping: 7/10 0.93947
stage 125/∞ ━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━ 7366/7366 0:00:00 0:06:24 val_accuracy: 0.92788  early_stopping: 8/10 0.93947
stage 126/∞ ━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━ 7366/7366 0:00:00 0:06:26 val_accuracy: 0.94028  early_stopping: 0/10 0.94028
stage 127/∞ ━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━ 7366/7366 0:00:00 0:07:04 val_accuracy: 0.93582  early_stopping: 1/10 0.94028
stage 128/∞ ━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━ 7366/7366 0:00:00 0:06:25 val_accuracy: 0.92848  early_stopping: 2/10 0.94028
stage 129/∞ ━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━ 7366/7366 0:00:00 0:06:22 val_accuracy: 0.93598  early_stopping: 3/10 0.94028
stage 130/∞ ━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━ 7366/7366 0:00:00 0:06:24 val_accuracy: 0.92518  early_stopping: 4/10 0.94028
stage 131/∞ ━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━ 7366/7366 0:00:00 0:06:23 val_accuracy: 0.93461  early_stopping: 5/10 0.94028
stage 132/∞ ━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━ 7366/7366 0:00:00 0:06:24 val_accuracy: 0.92764  early_stopping: 6/10 0.94028
stage 133/∞ ━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━ 7366/7366 0:00:00 0:06:22 val_accuracy: 0.94162  early_stopping: 0/10 0.94162
stage 134/∞ ━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━ 7366/7366 0:00:00 0:06:26 val_accuracy: 0.94411  early_stopping: 0/10 0.94411
stage 135/∞ ━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━ 7366/7366 0:00:00 0:06:23 val_accuracy: 0.93835  early_stopping: 1/10 0.94411
stage 136/∞ ━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━ 7366/7366 0:00:00 0:06:28 val_accuracy: 0.93794  early_stopping: 2/10 0.94411
stage 137/∞ ━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━ 7366/7366 0:00:00 0:06:22 val_accuracy: 0.94000  early_stopping: 3/10 0.94411
stage 138/∞ ━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━ 7366/7366 0:00:00 0:06:57 val_accuracy: 0.93875  early_stopping: 4/10 0.94411
stage 139/∞ ━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━ 7366/7366 0:00:00 0:06:23 val_accuracy: 0.93735  early_stopping: 5/10 0.94411
stage 140/∞ ━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━ 7366/7366 0:00:00 0:06:23 val_accuracy: 0.93894  early_stopping: 6/10 0.94411
stage 141/∞ ━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━ 7366/7366 0:00:00 0:06:23 val_accuracy: 0.93327  early_stopping: 7/10 0.94411
stage 142/∞ ━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━ 7366/7366 0:00:00 0:06:26 val_accuracy: 0.93632  early_stopping: 8/10 0.94411
stage 143/∞ ━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━ 7366/7366 0:00:00 0:06:26 val_accuracy: 0.93729  early_stopping: 9/10 0.94411
stage 144/∞ ━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━ 7366/7366 0:00:00 0:06:26 val_accuracy: 0.92988  early_stopping: 10/10 0.94411
stage 145/∞ ━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━ 7366/7366 0:00:00 0:06:21 val_accuracy: 0.93181  early_stopping: 11/10 0.94411
stage 146/∞ ━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━ 7366/7366 0:00:00 0:06:29 val_accuracy: 0.93701  early_stopping: 12/10 0.94411
stage 147/∞ ━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━ 7366/7366 0:00:00 0:06:27 val_accuracy: 0.94056  early_stopping: 13/10 0.94411
stage 148/∞ ━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━ 7366/7366 0:00:00 0:07:03 val_accuracy: 0.94401  early_stopping: 14/10 0.94411
stage 149/∞ ━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━ 7366/7366 0:00:00 0:07:35 val_accuracy: 0.94548  early_stopping: 0/10 0.94548
stage 150/∞ ━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━ 7366/7366 0:00:00 0:06:23 val_accuracy: 0.92340  early_stopping: 1/10 0.94548
stage 151/∞ ━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━ 7366/7366 0:00:00 0:06:25 val_accuracy: 0.93997  early_stopping: 2/10 0.94548
stage 152/∞ ━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━ 7366/7366 0:00:00 0:06:29 val_accuracy: 0.93903  early_stopping: 3/10 0.94548
stage 153/∞ ━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━ 7366/7366 0:00:00 0:06:22 val_accuracy: 0.93421  early_stopping: 4/10 0.94548
stage 154/∞ ━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━ 7366/7366 0:00:00 0:06:31 val_accuracy: 0.93856  early_stopping: 5/10 0.94548
stage 155/∞ ━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━ 7366/7366 0:00:00 0:06:26 val_accuracy: 0.93860  early_stopping: 6/10 0.94548
stage 156/∞ ━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━ 7366/7366 0:00:00 0:06:25 val_accuracy: 0.92138  early_stopping: 7/10 0.94548
stage 157/∞ ━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━ 7366/7366 0:00:00 0:06:25 val_accuracy: 0.93825  early_stopping: 8/10 0.94548
stage 158/∞ ━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━ 7366/7366 0:00:00 0:06:27 val_accuracy: 0.93171  early_stopping: 9/10 0.94548
stage 159/∞ ━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━ 7366/7366 0:00:00 0:06:59 val_accuracy: 0.93087  early_stopping: 10/10 0.94548
stage 160/∞ ━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━ 7366/7366 0:00:00 0:06:24 val_accuracy: 0.93695  early_stopping: 11/10 0.94548
stage 161/∞ ━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━ 7366/7366 0:00:00 0:06:22 val_accuracy: 0.93364  early_stopping: 12/10 0.94548
stage 162/∞ ━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━ 7366/7366 0:00:00 0:06:24 val_accuracy: 0.94330  early_stopping: 13/10 0.94548
stage 163/∞ ━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━ 7366/7366 0:00:00 0:06:23 val_accuracy: 0.93249  early_stopping: 14/10 0.94548
stage 164/∞ ━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━ 7366/7366 0:00:00 0:06:22 val_accuracy: 0.94252  early_stopping: 15/10 0.94548
stage 165/∞ ━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━ 7366/7366 0:00:00 0:06:24 val_accuracy: 0.93324  early_stopping: 16/10 0.94548
stage 166/∞ ━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━ 7366/7366 0:00:00 0:06:23 val_accuracy: 0.94255  early_stopping: 17/10 0.94548
stage 167/∞ ━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━ 7366/7366 0:00:00 0:06:26 val_accuracy: 0.93212  early_stopping: 18/10 0.94548
stage 168/∞ ━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━ 7366/7366 0:00:00 0:06:23 val_accuracy: 0.93785  early_stopping: 19/10 0.94548
stage 169/∞ ━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━ 7366/7366 0:00:00 0:06:32 val_accuracy: 0.93906  early_stopping: 20/10 0.94548
stage 170/∞ ━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━ 7366/7366 0:00:00 0:07:08 val_accuracy: 0.93794  early_stopping: 21/10 0.94548
stage 171/∞ ━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━ 7366/7366 0:00:00 0:06:33 val_accuracy: 0.93003  early_stopping: 22/10 0.94548
stage 172/∞ ━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━ 7366/7366 0:00:00 0:06:24 val_accuracy: 0.92888  early_stopping: 23/10 0.94548
stage 173/∞ ━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━ 7366/7366 0:00:00 0:06:21 val_accuracy: 0.93140  early_stopping: 24/10 0.94548
stage 174/∞ ━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━ 7366/7366 0:00:00 0:06:28 val_accuracy: 0.93941  early_stopping: 25/10 0.94548
stage 175/∞ ━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━ 7366/7366 0:00:00 0:06:24 val_accuracy: 0.93255  early_stopping: 26/10 0.94548
stage 176/∞ ━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━ 7366/7366 0:00:00 0:06:25 val_accuracy: 0.93190  early_stopping: 27/10 0.94548
stage 177/∞ ━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━ 7366/7366 0:00:00 0:06:26 val_accuracy: 0.93414  early_stopping: 28/10 0.94548
stage 178/∞ ━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━ 7366/7366 0:00:00 0:06:26 val_accuracy: 0.93682  early_stopping: 29/10 0.94548
stage 179/∞ ━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━ 7366/7366 0:00:00 0:06:25 val_accuracy: 0.93069  early_stopping: 30/10 0.94548
stage 180/∞ ━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━ 7366/7366 0:00:00 0:06:24 val_accuracy: 0.93327  early_stopping: 31/10 0.94548
stage 181/∞ ━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━ 7366/7366 0:00:00 0:07:01 val_accuracy: 0.92605  early_stopping: 32/10 0.94548
stage 182/∞ ━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━ 7366/7366 0:00:00 0:06:26 val_accuracy: 0.93193  early_stopping: 33/10 0.94548
stage 183/∞ ━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━ 7366/7366 0:00:00 0:06:23 val_accuracy: 0.93760  early_stopping: 34/10 0.94548
stage 184/∞ ━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━ 7366/7366 0:00:00 0:06:23 val_accuracy: 0.93265  early_stopping: 35/10 0.94548
stage 185/∞ ━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━ 7366/7366 0:00:00 0:06:24 val_accuracy: 0.92767  early_stopping: 36/10 0.94548
stage 186/∞ ━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━ 7366/7366 0:00:00 0:06:25 val_accuracy: 0.93747  early_stopping: 37/10 0.94548
stage 187/∞ ━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━ 7366/7366 0:00:00 0:06:22 val_accuracy: 0.93909  early_stopping: 38/10 0.94548
stage 188/∞ ━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━ 7366/7366 0:00:00 0:06:24 val_accuracy: 0.93695  early_stopping: 39/10 0.94548
stage 189/∞ ━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━ 7366/7366 0:00:00 0:06:21 val_accuracy: 0.93844  early_stopping: 40/10 0.94548
stage 190/∞ ━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━ 7366/7366 0:00:00 0:06:23 val_accuracy: 0.93053  early_stopping: 41/10 0.94548
stage 191/∞ ━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━ 7366/7366 0:00:00 0:07:16 val_accuracy: 0.94034  early_stopping: 42/10 0.94548
stage 192/∞ ━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━ 7366/7366 0:00:00 0:06:58 val_accuracy: 0.93421  early_stopping: 43/10 0.94548
stage 193/∞ ━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━ 7366/7366 0:00:00 0:06:24 val_accuracy: 0.93455  early_stopping: 44/10 0.94548
stage 194/∞ ━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━ 7366/7366 0:00:00 0:06:26 val_accuracy: 0.93383  early_stopping: 45/10 0.94548
stage 195/∞ ━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━ 7366/7366 0:00:00 0:06:23 val_accuracy: 0.93090  early_stopping: 46/10 0.94548
stage 196/∞ ━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━ 7366/7366 0:00:00 0:06:27 val_accuracy: 0.94563  early_stopping: 0/10 0.94563
stage 197/∞ ━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━ 7366/7366 0:00:00 0:06:25 val_accuracy: 0.93657  early_stopping: 1/10 0.94563
stage 198/∞ ━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━ 7366/7366 0:00:00 0:06:26 val_accuracy: 0.94068  early_stopping: 2/10 0.94563
stage 199/∞ ━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━ 7366/7366 0:00:00 0:06:24 val_accuracy: 0.93483  early_stopping: 3/10 0.94563
stage 200/∞ ━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━ 7366/7366 0:00:00 0:06:26 val_accuracy: 0.93050  early_stopping: 4/10 0.94563
stage 201/∞ ━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━ 7366/7366 0:00:00 0:06:26 val_accuracy: 0.93187  early_stopping: 5/10 0.94563
stage 202/∞ ━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━ 7366/7366 0:00:00 0:07:09 val_accuracy: 0.94731  early_stopping: 0/10 0.94731
stage 203/∞ ━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━ 7366/7366 0:00:00 0:06:28 val_accuracy: 0.93396  early_stopping: 1/10 0.94731
stage 204/∞ ━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━ 7366/7366 0:00:00 0:06:26 val_accuracy: 0.93642  early_stopping: 2/10 0.94731
stage 205/∞ ━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━ 7366/7366 0:00:00 0:06:28 val_accuracy: 0.93501  early_stopping: 3/10 0.94731
stage 206/∞ ━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━ 7366/7366 0:00:00 0:06:27 val_accuracy: 0.93224  early_stopping: 4/10 0.94731
stage 207/∞ ━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━ 7366/7366 0:00:00 0:06:20 val_accuracy: 0.94221  early_stopping: 5/10 0.94731
stage 208/∞ ━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━ 7366/7366 0:00:00 0:06:24 val_accuracy: 0.93473  early_stopping: 6/10 0.94731
stage 209/∞ ━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━ 7366/7366 0:00:00 0:06:21 val_accuracy: 0.93726  early_stopping: 7/10 0.94731
stage 210/∞ ━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━ 7366/7366 0:00:00 0:06:25 val_accuracy: 0.94062  early_stopping: 8/10 0.94731
stage 211/∞ ━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━ 7366/7366 0:00:00 0:06:23 val_accuracy: 0.93928  early_stopping: 9/10 0.94731
stage 212/∞ ━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━ 7366/7366 0:00:00 0:06:24 val_accuracy: 0.93931  early_stopping: 10/10 0.94731
Moving best model Juristische_Konsilien_Tuebingen+_202.mlmodel (0.9473143219947815) to Juristische_Konsilien_Tuebingen+_best.mlmodel

real    1387m15.438s
user    19775m16.197s
sys     11007m8.085s

Run training with HTR+ like network specification

(venv3.9) stweil@ocr-01:~/src/gitlab/scripta/escriptorium/Juristische_Konsilien_Tuebingen/Transkribus_Exporte$ time nice ketos train -f page -t list.train -e list.eval -o Juristische_Konsilien_Tuebingen+256 -d cuda:0 --augment --workers 24 -r 0.0001 -B 1 --min-epochs 200 --lag 20 -w 0 -s '[256,64,0,1 Cr4,2,8,4,2 Cr4,2,32,1,1 Mp4,2,4,2 Cr3,3,64,1,1 Mp1,2,1,2 S1(1x0)1,3 Lbx256 Do0.5 Lbx256 Do0.5 Lbx256 Do0.5 Cr255,1,85,1,1]'
WARNING:root:scikit-learn version 1.1.1 is not supported. Minimum required version: 0.17. Maximum required version: 0.19.2. Disabling scikit-learn conversion API.
WARNING:root:Torch version 1.11.0+cu113 has not been tested with coremltools. You may run into unexpected errors. Torch 1.10.2 is the most recent version that has been tested.
[05/27/22 17:05:44] WARNING  alphabet mismatch: chars in training set only: {'½', 'ꝟ', '‡', 'Ü', '’', 'û', ']', '[', 'º', '♃', 'X', '╒', 'ꝸ', 'ꝯ', '=', '†'} (not included in accuracy test during        train.py:304
                             training)                                                                                                                                                                                
Trainer already configured with model summary callbacks: [<class 'pytorch_lightning.callbacks.rich_model_summary.RichModelSummary'>]. Skipping setting a default `ModelSummary` callback.
GPU available: True, used: True
TPU available: False, using: 0 TPU cores
IPU available: False, using: 0 IPUs
HPU available: False, using: 0 HPUs
`Trainer(val_check_interval=1.0)` was configured so validation will run at the end of the training epoch..
LOCAL_RANK: 0 - CUDA_VISIBLE_DEVICES: [0]
┏━━━━┳━━━━━━━━━━━┳━━━━━━━━━━━━━━━━━━━━━━━━━━┳━━━━━━━━┓
┃    ┃ Name      ┃ Type                     ┃ Params ┃
┡━━━━╇━━━━━━━━━━━╇━━━━━━━━━━━━━━━━━━━━━━━━━━╇━━━━━━━━┩
│ 0  │ net       │ MultiParamSequential     │ 15.2 M │
│ 1  │ net.C_0   │ ActConv2D                │     72 │
│ 2  │ net.C_1   │ ActConv2D                │  2.1 K │
│ 3  │ net.Mp_2  │ MaxPool                  │      0 │
│ 4  │ net.C_3   │ ActConv2D                │ 18.5 K │
│ 5  │ net.Mp_4  │ MaxPool                  │      0 │
│ 6  │ net.S_5   │ Reshape                  │      0 │
│ 7  │ net.L_6   │ TransposedSummarizingRNN │  921 K │
│ 8  │ net.Do_7  │ Dropout                  │      0 │
│ 9  │ net.L_8   │ TransposedSummarizingRNN │  1.6 M │
│ 10 │ net.Do_9  │ Dropout                  │      0 │
│ 11 │ net.L_10  │ TransposedSummarizingRNN │  1.6 M │
│ 12 │ net.Do_11 │ Dropout                  │      0 │
│ 13 │ net.C_12  │ ActConv2D                │ 11.1 M │
│ 14 │ net.O_13  │ LinSoftmax               │ 10.4 K │
└────┴───────────┴──────────────────────────┴────────┘
Trainable params: 15.2 M                                                                                                                                                                                              
Non-trainable params: 0                                                                                                                                                                                               
Total params: 15.2 M                                                                                                                                                                                                  
Total estimated model params size (MB): 60                                                                                                                                                                            
stage 0/∞  ━━━━━╺━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━ 956/7366 0:05:12 0:00:58  early_stopping: 0/20 -inf