duckling_preview.ipynb

       "124  SUBSTITUTION                     dobrze               bardzo\n",
       "125  SUBSTITUTION                         to                 panu\n",
       "126  SUBSTITUTION                   poproszę             dziękuję\n",
       "127  SUBSTITUTION                       pana            internetu\n",
       "128  SUBSTITUTION                    jeszcze           przypadają\n",
       "129  SUBSTITUTION                          o                   na\n",
       "130       CORRECT                    podanie              podanie\n",
       "131      DELETION                   czterech                     \n",
       "132      DELETION                  ostatnich                     \n",
       "133      DELETION                       cyfr                     \n",
       "134      DELETION                          z                     \n",
       "135      DELETION                     numeru                     \n",
       "136      DELETION                      karty                     \n",
       "137      DELETION                      którą                     \n",
       "138      DELETION                    opłacał                     \n",
       "139      DELETION                        pan                     \n",
       "140      DELETION                 rezerwację                     \n",
       "141      DELETION                     cztery                     \n",
       "142      DELETION                   ostatnie                     \n",
       "143      DELETION                      cyfry                     \n",
       "144      DELETION                      tylko                     \n",
       "145      DELETION                 oczywiście                     \n",
       "146  SUBSTITUTION                        jak              ostatni\n",
       "147  SUBSTITUTION                najbardziej                  rok\n",
       "148  SUBSTITUTION                         ja                  dym\n",
       "149  SUBSTITUTION                        też                    i\n",
       "150  SUBSTITUTION                    swojego                hartu\n",
       "151  SUBSTITUTION                     numeru                która\n",
       "152  SUBSTITUTION                      karty            opanowała\n",
       "153  SUBSTITUTION                        nie             rybackie\n",
       "154  SUBSTITUTION                       znam                    i\n",
       "155  SUBSTITUTION                         na                kutry\n",
       "156  SUBSTITUTION                     pamięć             rybackie\n",
       "157  SUBSTITUTION                        tak                    i\n",
       "158  SUBSTITUTION                         że                kutry\n",
       "159  SUBSTITUTION                        nie             rybackie\n",
       "160  SUBSTITUTION                      dziwi                    i\n",
       "161  SUBSTITUTION                       mnie              wybrali\n",
       "162  SUBSTITUTION                         to            pojedynek\n",
       "163  SUBSTITUTION                 absolutnie                 ludy\n",
       "164  SUBSTITUTION                         ma                bantu\n",
       "165  SUBSTITUTION                        pan                    i\n",
       "166  SUBSTITUTION                       tyle           potrzebują\n",
       "167  SUBSTITUTION                      czasu               dużego\n",
       "168  SUBSTITUTION                        ile             dziękuje\n",
       "169  SUBSTITUTION                        pan                    <\n",
       "170  SUBSTITUTION                 potrzebuje                  unk\n",
       "171  SUBSTITUTION                     bardzo                    >\n",
       "172  SUBSTITUTION                       panu               buduje\n",
       "173  SUBSTITUTION                   dziękuję                swoją\n",
       "174  SUBSTITUTION                         no                    <\n",
       "175  SUBSTITUTION                     dobrze                  unk\n",
       "176  SUBSTITUTION                         ja                    >\n",
       "177  SUBSTITUTION                        już                    i\n",
       "178  SUBSTITUTION                 przechodzę                 jego\n",
       "179  SUBSTITUTION                         do        identyfikacja\n",
       "180  SUBSTITUTION                    meritum                 była\n",
       "181  SUBSTITUTION                         bo                tłumy\n",
       "182  SUBSTITUTION                 oczywiście                  nie\n",
       "183  SUBSTITUTION                   wszystko             nadawały\n",
       "184       CORRECT                        się                  się\n",
       "185      DELETION                     zgadza                     \n",
       "186      DELETION                weryfikacja                     \n",
       "187      DELETION                 przebiegła                     \n",
       "188      DELETION                  pomyślnie                     \n",
       "189      DELETION                  natomiast                     \n",
       "190      DELETION                         no                     \n",
       "191      DELETION                    dzwonię                     \n",
       "192      DELETION                         do                     \n",
       "193      DELETION                       pana                     \n",
       "194  SUBSTITUTION                          z               godnym\n",
       "195  SUBSTITUTION                     powodu                miana\n",
       "196  SUBSTITUTION                      dosyć              ugodowy\n",
       "197  SUBSTITUTION                    takiego                    i\n",
       "198       CORRECT                       mało                 mało\n",
       "199  SUBSTITUTION                przyjemnego              wolnego\n",
       "200       CORRECT                        nie                  nie\n",
       "201      DELETION                    ukrywam                     \n",
       "202  SUBSTITUTION                          z             ukrywali\n",
       "203       CORRECT                       tego                 tego\n",
       "204      DELETION                    względu                     \n",
       "205      DELETION                         że                     \n",
       "206  SUBSTITUTION                         za           decydujące\n",
       "207  SUBSTITUTION                      okres                  dla\n",
       "208  SUBSTITUTION                wynajmowana           określenia\n",
       "209  SUBSTITUTION                       prze          najbardziej\n",
       "210       CORRECT               wynajmowania         wynajmowania\n",
       "211      DELETION                przepraszam                     \n",
       "212  SUBSTITUTION                      przez               oprócz\n",
       "213       CORRECT                       pana                 pana\n",
       "214  SUBSTITUTION                    jednego              jerzego\n",
       "215  SUBSTITUTION                          z                tutaj\n",
       "216  SUBSTITUTION                    naszych                 lodu\n",
       "217  SUBSTITUTION                 samochodów            otrzymali\n",
       "218  SUBSTITUTION               otrzymaliśmy               urlopy\n",
       "219  SUBSTITUTION                     mandat                    i\n",
       "220  SUBSTITUTION                         za                udają\n",
       "221  SUBSTITUTION              przekroczenie                  się\n",
       "222  SUBSTITUTION                  prędkości              wkrótce\n",
       "223  SUBSTITUTION                         na                  nie\n",
       "224  SUBSTITUTION                     drodze                tylko\n",
       "225  SUBSTITUTION                   krajowej                   ci\n",
       "226  SUBSTITUTION                          k                  dla\n",
       "227  SUBSTITUTION              osiemdziesiąt                kraju\n",
       "228  SUBSTITUTION                       pięć                    i\n",
       "229  SUBSTITUTION                     między                   za\n",
       "230  SUBSTITUTION                amsterdamem        osiemdziesiąt\n",
       "231  SUBSTITUTION                          a                 pięć\n",
       "232  SUBSTITUTION                rotterdamem              ansbach\n",
       "233       CORRECT                          i                    i\n",
       "234  SUBSTITUTION                          z             bayreuth\n",
       "235  SUBSTITUTION                        tej                    i\n",
       "236  SUBSTITUTION             korespondencji               udręki\n",
       "237  SUBSTITUTION                      którą                które\n",
       "238       CORRECT               otrzymaliśmy         otrzymaliśmy\n",
       "239      DELETION                     wynika                     \n",
       "240      DELETION                         że                     \n",
       "241      DELETION                        ten                     \n",
       "242  SUBSTITUTION                     mandat          komunikacji\n",
       "243  SUBSTITUTION                     został             tyrmanda\n",
       "244  SUBSTITUTION                 wystawiony               odbyły\n",
       "245  SUBSTITUTION                          w                  się\n",
       "246  SUBSTITUTION                    miejscu                   tu\n",
       "247  SUBSTITUTION                      gdzie               będzie\n",
       "248       CORRECT               dopuszczalna         dopuszczalna\n",
       "249       CORRECT                   prędkość             prędkość\n",
       "250     INSERTION                                         wynosił\n",
       "251  SUBSTITUTION                     wynosi                  sto\n",
       "252  SUBSTITUTION              sześćdziesiąt                jeden\n",
       "253       CORRECT                 kilometrów           kilometrów\n",
       "254       CORRECT                         na                   na\n",
       "255       CORRECT                    godzinę              godzinę\n",
       "256  SUBSTITUTION                         no                 ruin\n",
       "257       CORRECT                          i                    i\n",
       "258      DELETION                        ten                     \n",
       "259      DELETION                      limit                     \n",
       "260      DELETION                     został                     \n",
       "261      DELETION               przekroczony                     \n",
       "262  SUBSTITUTION                          o               gruzów\n",
       "263  SUBSTITUTION                dwadzieścia               odbywa\n",
       "264  SUBSTITUTION                       trzy                  się\n",
       "265  SUBSTITUTION                  kilometry           kilometrów\n",
       "266       CORRECT                         na                   na\n",
       "267       CORRECT                    godzinę              godzinę\n",
       "268       CORRECT                          w                    w\n",
       "269      DELETION                    związku                     \n",
       "270      DELETION                          z                     \n",
       "271      DELETION                       czym                     \n",
       "272  SUBSTITUTION                         no                  tym\n",
       "273  SUBSTITUTION                         po                 roku\n",
       "274  SUBSTITUTION                     prostu                 było\n",
       "275  SUBSTITUTION                   będziemy                   tu\n",
       "276  SUBSTITUTION                   zmuszeni                  ani\n",
       "277  SUBSTITUTION                   obciążyć               odciąć\n",
       "278       CORRECT                       pana                 pana\n",
       "279  SUBSTITUTION                 wysokością                 bytu\n",
       "280  SUBSTITUTION                       tego              pociągu\n",
       "281       CORRECT                    mandatu              mandatu\n",
       "282       CORRECT                      który                który\n",
       "283      DELETION                     został                     \n",
       "284      DELETION                         do                     \n",
       "285  SUBSTITUTION                        nas                 znad\n",
       "286  SUBSTITUTION                    wysłany              bałtyku\n",
       "287       CORRECT                          i                    i\n",
       "288      DELETION                         to                     \n",
       "289      DELETION                       jest                     \n",
       "290      DELETION                     mandat                     \n",
       "291      DELETION                          w                     \n",
       "292      DELETION                  wysokości                     \n",
       "293  SUBSTITUTION                    trzystu             feldmana\n",
       "294  SUBSTITUTION             pięćdziesięciu                 ukoi\n",
       "295  SUBSTITUTION                       euro              uniknąć\n",
       "296  SUBSTITUTION                  dokładnie                 losu\n",
       "297  SUBSTITUTION                        tak                 pani\n",
       "298  SUBSTITUTION                        jak                 basi\n",
       "299  SUBSTITUTION                najbardziej           kierującej\n",
       "300  SUBSTITUTION               otrzymaliśmy            otrzymali\n",
       "301       CORRECT                     mandat               mandat\n",
       "302      DELETION                          z                     \n",
       "303      DELETION             korespondencji                     \n",
       "304      DELETION                      którą                     \n",
       "305  SUBSTITUTION               otrzymaliśmy      eksperymentalne\n",
       "306  SUBSTITUTION                         od               studio\n",
       "307  SUBSTITUTION                    policji                który\n",
       "308  SUBSTITUTION              holenderskiej             otrzymał\n",
       "309  SUBSTITUTION                     wynika                 film\n",
       "310  SUBSTITUTION                         że                  oto\n",
       "311  SUBSTITUTION                         ta                    i\n",
       "312  SUBSTITUTION                       data            endeckiej\n",
       "313  SUBSTITUTION                      gdzie        communicating\n",
       "314  SUBSTITUTION                       było             sztandar\n",
       "315  SUBSTITUTION                   zrobione               pokoju\n",
       "316  SUBSTITUTION                    zdjęcie                takie\n",
       "317  SUBSTITUTION                        był              budynki\n",
       "318  SUBSTITUTION                      zapis                    i\n",
       "319  SUBSTITUTION                          z              budowle\n",
       "320  SUBSTITUTION                 fotoradaru                 daje\n",
       "321       CORRECT                        ona                  ona\n",
       "322      DELETION                    pokrywa                     \n",
       "323      DELETION                        się                     \n",
       "324      DELETION                          z                     \n",
       "325  SUBSTITUTION                        tym               ukrywa\n",
       "326       CORRECT                      kiedy                kiedy\n",
       "327  SUBSTITUTION                        pan                 panu\n",
       "328  SUBSTITUTION                 wynajmował          wynajmowała\n",
       "329       CORRECT                         od                   od\n",
       "330     INSERTION                                             lat\n",
       "331  SUBSTITUTION                        nas               budują\n",
       "332  SUBSTITUTION                   samochód                  lud\n",
       "333  SUBSTITUTION                          z             potonęły\n",
       "334  SUBSTITUTION                       tego          dziewięćset\n",
       "335  SUBSTITUTION                    względu         dwudziestego\n",
       "336  SUBSTITUTION                         że             drugiego\n",
       "337  SUBSTITUTION                  fotoradar           cierpienia\n",
       "338  SUBSTITUTION                     zrobił                    i\n",
       "339  SUBSTITUTION                    zdjęcie            dokładnie\n",
       "340  SUBSTITUTION               dwudziestego            odwiedzin\n",
       "341  SUBSTITUTION                   drugiego                   na\n",
       "342  SUBSTITUTION                   sierpnia               kursie\n",
       "343  SUBSTITUTION                          i            szkolenia\n",
       "344  SUBSTITUTION                  dokładnie                medal\n",
       "345  SUBSTITUTION                          o                    i\n",
       "346  SUBSTITUTION                   godzinie              nagrodę\n",
       "347  SUBSTITUTION               siedemnastej                    w\n",
       "348  SUBSTITUTION               czterdzieści                kraju\n",
       "349  SUBSTITUTION                     siedem               pokoju\n",
       "350       CORRECT                         na                   na\n",
       "351  SUBSTITUTION                     drodze                całym\n",
       "352  SUBSTITUTION                   krajowej                  jej\n",
       "353  SUBSTITUTION                          k                życiu\n",
       "354  SUBSTITUTION              sześćdziesiąt            osobistym\n",
       "355  SUBSTITUTION                      osiem                    i\n",
       "356  SUBSTITUTION                    zostało               dotarł\n",
       "357  SUBSTITUTION                    została                   aż\n",
       "358  SUBSTITUTION               przekroczona                tutaj\n",
       "359  SUBSTITUTION                   prędkość            prędkości\n",
       "360       CORRECT                          o                    o\n",
       "361       CORRECT                dwadzieścia          dwadzieścia\n",
       "362       CORRECT                      osiem                osiem\n",
       "363       CORRECT                 kilometrów           kilometrów\n",
       "364       CORRECT                         na                   na\n",
       "365       CORRECT                    godzinę              godzinę\n",
       "366      DELETION                         no                     \n",
       "367       CORRECT                          i                    i\n",
       "368      DELETION                   ponieważ                     \n",
       "369      DELETION                      tutaj                     \n",
       "370  SUBSTITUTION                        nie              broniły\n",
       "371  SUBSTITUTION                         ma              kopalni\n",
       "372  SUBSTITUTION                wątpliwości              ładunki\n",
       "373  SUBSTITUTION                         co                  był\n",
       "374       CORRECT                         do                   do\n",
       "375       CORRECT                       tego                 tego\n",
       "376      DELETION                        kto                     \n",
       "377      DELETION                 wynajmował                     \n",
       "378      DELETION                       auto                     \n",
       "379      DELETION                          w                     \n",
       "380      DELETION                        tym                     \n",
       "381  SUBSTITUTION                     czasie           bynajmniej\n",
       "382  SUBSTITUTION                    dzwonię            autorskie\n",
       "383  SUBSTITUTION                         do                dłoni\n",
       "384  SUBSTITUTION                       pana             dokonali\n",
       "385  SUBSTITUTION                       żeby                   by\n",
       "386       CORRECT               poinformować         poinformować\n",
       "387       CORRECT                          o                    o\n",
       "388       CORRECT               konieczności         konieczności\n",
       "389  SUBSTITUTION                 obciążenia          nieustannej\n",
       "390  SUBSTITUTION                       pana             carlotto\n",
       "391  SUBSTITUTION                      karty                    <\n",
       "392  SUBSTITUTION                         no                  unk\n",
       "393  SUBSTITUTION                        też                    >\n",
       "394  SUBSTITUTION                   chciałam            chciałbym\n",
       "395       CORRECT               poinformować         poinformować\n",
       "396       CORRECT                       pana                 pana\n",
       "397       CORRECT                          o                    o\n",
       "398       CORRECT                        tym                  tym\n",
       "399      DELETION                  wcześniej                     \n",
       "400      DELETION                      zanim                     \n",
       "401  SUBSTITUTION                         do                    i\n",
       "402  SUBSTITUTION                       tego               danuty\n",
       "403  SUBSTITUTION                 obciążenia             ciążenie\n",
       "404  SUBSTITUTION                    dojdzie                   do\n",
       "405  SUBSTITUTION                       żeby                 izby\n",
       "406       CORRECT                      tutaj                tutaj\n",
       "407      DELETION                         no                     \n",
       "408      DELETION                        nie                     \n",
       "409      DELETION                       było                     \n",
       "410      DELETION                    później                     \n",
       "411      DELETION                niejasności                     \n",
       "412      DELETION                   dlaczego                     \n",
       "413      DELETION                         tę                     \n",
       "414      DELETION                  dodatkową                     \n",
       "415  SUBSTITUTION                      kwotę                buntu\n",
       "416  SUBSTITUTION                       panu                    i\n",
       "417  SUBSTITUTION               naliczyliśmy                  nie\n",
       "418  SUBSTITUTION                      mówię               nadaje\n",
       "419  SUBSTITUTION                      tutaj                  się\n",
       "420  SUBSTITUTION                        też                   do\n",
       "421  SUBSTITUTION                          o               upadku\n",
       "422  SUBSTITUTION                          o                   na\n",
       "423  SUBSTITUTION              równowartości                rynku\n",
       "424  SUBSTITUTION                    trzystu                    i\n",
       "425  SUBSTITUTION             pięćdziesięciu                    w\n",
       "426  SUBSTITUTION                       euro                  tym\n",
       "427  SUBSTITUTION                     ciężko              okresie\n",
       "428  SUBSTITUTION                         mi                wojny\n",
       "429  SUBSTITUTION                 powiedzieć               ludowo\n",
       "430  SUBSTITUTION                       jaka               ludowi\n",
       "431  SUBSTITUTION                         to                    i\n",
       "432  SUBSTITUTION                     będzie              uczynić\n",
       "433  SUBSTITUTION                   dokładna               europę\n",
       "434  SUBSTITUTION                      kwota                    i\n",
       "435  SUBSTITUTION                          w                podał\n",
       "436  SUBSTITUTION                 złotówkach                  jej\n",
       "437  SUBSTITUTION                         no                 rękę\n",
       "438  SUBSTITUTION                         bo                  był\n",
       "439  SUBSTITUTION                         to             dokładny\n",
       "440  SUBSTITUTION                        też                wśród\n",
       "441  SUBSTITUTION                        nie             atrybutu\n",
       "442  SUBSTITUTION                     zależy                judel\n",
       "443       CORRECT                         od                   od\n",
       "444  SUBSTITUTION                        nas                  nad\n",
       "445  SUBSTITUTION                      tylko                    i\n",
       "446  SUBSTITUTION                         od                    &\n",
       "447  SUBSTITUTION                      banku                 quot\n",
       "448  SUBSTITUTION                        czy                    ;\n",
       "449  SUBSTITUTION                         od                  bal\n",
       "450  SUBSTITUTION                    wydawcy                    u\n",
       "451  SUBSTITUTION                      karty             odbudowy\n",
       "452  SUBSTITUTION                         od                karty\n",
       "453  SUBSTITUTION                         od            otrzymały\n",
       "454  SUBSTITUTION                organizacji             odbudowę\n",
       "455  SUBSTITUTION                 płatniczej               potęgę\n",
       "456       CORRECT                      która                która\n",
       "457      DELETION                       panu                     \n",
       "458  SUBSTITUTION                      kartę               danuty\n",
       "459       CORRECT                     wydała               wydała\n",
       "460      DELETION                        tak                     \n",
       "461      DELETION                         że                     \n",
       "462      DELETION                         no                     \n",
       "463      DELETION                         ja                     \n",
       "464      DELETION                          w                     \n",
       "465  SUBSTITUTION                        tym                 swój\n",
       "466  SUBSTITUTION                   momencie              czwarty\n",
       "467  SUBSTITUTION                   dokonuję             dokonuje\n",
       "468       CORRECT                    blokady              blokady\n",
       "469       CORRECT                    środków              środków\n",
       "470       CORRECT                         na                   na\n",
       "471       CORRECT               równowartość         równowartość\n",
       "472      DELETION                    trzysta                     \n",
       "473       CORRECT               pięćdziesiąt         pięćdziesiąt\n",
       "474       CORRECT                       euro                 euro\n",
       "475       CORRECT                         na                   na\n",
       "476      DELETION                        tej                     \n",
       "477      DELETION                     karcie                     \n",
       "478  SUBSTITUTION                      którą         amerykańskie\n",
       "479       CORRECT                        nam                  nam\n",
       "480       CORRECT                        pan                  pan\n",
       "481  SUBSTITUTION                      podał                podać\n",
       "482  SUBSTITUTION                      czyli                 swój\n",
       "483       CORRECT             siedemdziesiąt       siedemdziesiąt\n",
       "484      DELETION                       trzy                     \n",
       "485      DELETION           dziewięćdziesiąt                     \n",
       "486      DELETION                      osiem                     \n",
       "487      DELETION                         bo                     \n",
       "488      DELETION                   rozumiem                     \n",
       "489  SUBSTITUTION                         że                nigdy\n",
       "490  SUBSTITUTION                         ta                  nie\n",
       "491  SUBSTITUTION                      karta                 ukoi\n",
       "492  SUBSTITUTION                          w                 bólu\n",
       "493  SUBSTITUTION                    dalszym              tokarka\n",
       "494  SUBSTITUTION                      ciągu                   do\n",
       "495  SUBSTITUTION                       jest              usiłują\n",
       "496       CORRECT                    aktywna              aktywna\n",
       "497      DELETION                        tak                     \n",
       "498      DELETION                     dobrze                     \n",
       "499      DELETION                         to                     \n",
       "500      DELETION                          w                     \n",
       "501      DELETION                      takim                     \n",
       "502      DELETION                      razie                     \n",
       "503       CORRECT                        tak                  tak\n",
       "504      DELETION                         ja                     \n",
       "505      DELETION                   dokonuję                     \n",
       "506      DELETION                      tutaj                     \n",
       "507      DELETION                    blokady                     \n",
       "508  SUBSTITUTION                       tych             dobitnie\n",
       "509  SUBSTITUTION                    środków                  oto\n",
       "510  SUBSTITUTION                     środki                 rytm\n",
       "511  SUBSTITUTION                   rozliczą                pracy\n",
       "512  SUBSTITUTION                        się                kraju\n",
       "513  SUBSTITUTION                         no                    i\n",
       "514  SUBSTITUTION                          w            produkują\n",
       "515  SUBSTITUTION                 zależności                tytoń\n",
       "516  SUBSTITUTION                         od                    i\n",
       "517  SUBSTITUTION                       tego               uniósł\n",
       "518  SUBSTITUTION                        jak             ładunków\n",
       "519  SUBSTITUTION                         to                unrra\n",
       "520  SUBSTITUTION                        tam            ironizuje\n",
       "521  SUBSTITUTION                          u                nadal\n",
       "522  SUBSTITUTION                       pana                  tym\n",
       "523  SUBSTITUTION                          w              upadłym\n",
       "524       CORRECT                      banku                banku\n",
       "525       CORRECT                    wygląda              wygląda\n",
       "526      DELETION                          w                     \n",
       "527  SUBSTITUTION                      ciągu                  mój\n",
       "528  SUBSTITUTION               czterdziestu                tatuś\n",
       "529  SUBSTITUTION                      ośmiu          bollywoodem\n",
       "530  SUBSTITUTION                     godzin                    i\n",
       "531  SUBSTITUTION           siedemdziesięciu               kończy\n",
       "532  SUBSTITUTION                      dwóch                długi\n",
       "533  SUBSTITUTION                     godzin                    i\n",
       "534  SUBSTITUTION                         ja              kieruje\n",
       "535  SUBSTITUTION                 oczywiście                  się\n",
       "536       CORRECT                         na                   na\n",
       "537  SUBSTITUTION                      adres            agresywną\n",
       "538  SUBSTITUTION                    mailowy               narodu\n",
       "539       CORRECT                      który                który\n",
       "540       CORRECT                        nam                  nam\n",
       "541       CORRECT                        pan                  pan\n",
       "542      DELETION                    podawał                     \n",
       "543  SUBSTITUTION                       przy                podać\n",
       "544       CORRECT                dokonywaniu          dokonywaniu\n",
       "545      DELETION                 rezerwacji                     \n",
       "546      DELETION                      czyli                     \n",
       "547      DELETION  arturdrzewieckiinternetpl                     \n",
       "548      DELETION                         na                     \n",
       "549      DELETION                        ten                     \n",
       "550      DELETION                      adres                     \n",
       "551  SUBSTITUTION                    mailowy               rybaku\n",
       "552  SUBSTITUTION                      wyślę                    i\n",
       "553  SUBSTITUTION                         do                artur\n",
       "554  SUBSTITUTION                       pana             gadowski\n",
       "555  SUBSTITUTION                       samą                    i\n",
       "556  SUBSTITUTION                      treść                małpa\n",
       "557  SUBSTITUTION                    mandatu              internę\n",
       "558  SUBSTITUTION                      wyślę                  ale\n",
       "559  SUBSTITUTION                        też                   na\n",
       "560  SUBSTITUTION                         to                  tym\n",
       "561  SUBSTITUTION                    zdjęcie                  aby\n",
       "562  SUBSTITUTION                          z                 były\n",
       "563  SUBSTITUTION                 fotoradaru                  one\n",
       "564  SUBSTITUTION                      wyślę              banalne\n",
       "565  SUBSTITUTION                        też               okresu\n",
       "566  SUBSTITUTION                       taki              mandatu\n",
       "567  SUBSTITUTION                  formularz                  był\n",
       "568  SUBSTITUTION                  oficjalne              budynku\n",
       "569  SUBSTITUTION                      pismo            utrzymana\n",
       "570  SUBSTITUTION                        jak              rubryki\n",
       "571  SUBSTITUTION                         to           formularza\n",
       "572  SUBSTITUTION                    wygląda         nieoficjalne\n",
       "573  SUBSTITUTION                          u                    i\n",
       "574  SUBSTITUTION                        nas           urządzenia\n",
       "575  SUBSTITUTION                     jeżeli          produkcyjne\n",
       "576  SUBSTITUTION                     chodzi               umaluj\n",
       "577  SUBSTITUTION                          o               ludzie\n",
       "578  SUBSTITUTION               przenoszenie                   od\n",
       "579  SUBSTITUTION                       tych                niego\n",
       "580  SUBSTITUTION                    kosztów                kultu\n",
       "581  SUBSTITUTION                          z               narodu\n",
       "582  SUBSTITUTION                     naszej                   po\n",
       "583  SUBSTITUTION               wypożyczalni              lukanii\n",
       "584       CORRECT                         na                   na\n",
       "585      DELETION                    naszych                     \n",
       "586      DELETION                   klientów                     \n",
       "587  SUBSTITUTION                      wyślę                ujęto\n",
       "588  SUBSTITUTION                        raz                    w\n",
       "589  SUBSTITUTION                    jeszcze           belwederze\n",
       "590       CORRECT                  regulamin            regulamin\n",
       "591      DELETION                          z                     \n",
       "592  SUBSTITUTION                zaznaczonym               senatu\n",
       "593  SUBSTITUTION                    punktem               punktu\n",
       "594       CORRECT                      który                który\n",
       "595      DELETION                          o                     \n",
       "596      DELETION                        tym                     \n",
       "597      DELETION                       mówi                     \n",
       "598      DELETION                         bo                     \n",
       "599      DELETION                      tutaj                     \n",
       "600      DELETION                         no                     \n",
       "601      DELETION                          z                     \n",
       "602      DELETION              doświadczenia                     \n",
       "603      DELETION                       wiem                     \n",
       "604      DELETION                         że                     \n",
       "605      DELETION                  większość                     \n",
       "606      DELETION                   klientów                     \n",
       "607  SUBSTITUTION                        nie                    <\n",
       "608  SUBSTITUTION                  zapoznaje                  unk\n",
       "609  SUBSTITUTION                        się                    >\n",
       "610  SUBSTITUTION                          z                 buty\n",
       "611  SUBSTITUTION                regulaminem                    i\n",
       "612  SUBSTITUTION                      przed                    <\n",
       "613  SUBSTITUTION              wypożyczeniem                  unk\n",
       "614  SUBSTITUTION                  samochodu                    >\n",
       "615  SUBSTITUTION                        tak            produkcję\n",
       "616  SUBSTITUTION                         że             niewiele\n",
       "617  SUBSTITUTION                         ja                czasu\n",
       "618  SUBSTITUTION                         po              upłynie\n",
       "619  SUBSTITUTION                     prostu                 dużo\n",
       "620  SUBSTITUTION                    pozwolę              podjęły\n",
       "621  SUBSTITUTION                      sobie            regularną\n",
       "622  SUBSTITUTION                     wysłać               czynem\n",
       "623  SUBSTITUTION                        raz         produkcyjnym\n",
       "624  SUBSTITUTION                    jeszcze               buduje\n",
       "625  SUBSTITUTION                        ten               powrót\n",
       "626  SUBSTITUTION                   fragment             potężnym\n",
       "627  SUBSTITUTION                 regulaminu             drygulec\n",
       "628       CORRECT                      który                który\n",
       "629  SUBSTITUTION                       mówi               mówiąc\n",
       "630       CORRECT                          o                    o\n",
       "631      DELETION                        tym                     \n",
       "632      DELETION              przeniesieniu                     \n",
       "633      DELETION                    kosztów                     \n",
       "634      DELETION                     bardzo                     \n",
       "635      DELETION                         mi                     \n",
       "636  SUBSTITUTION                    przykro                 niej\n",
       "637  SUBSTITUTION                         że            całkowitą\n",
       "638  SUBSTITUTION                      takie                 domy\n",
       "639  SUBSTITUTION                 informacje                    i\n",
       "640  SUBSTITUTION                      muszę              polecił\n",
       "641       CORRECT                       panu                 panu\n",
       "642      DELETION                  przekazać                     \n",
       "643      DELETION                        ale                     \n",
       "644      DELETION                        sam                     \n",
       "645      DELETION                        pan                     \n",
       "646      DELETION                    rozumie                     \n",
       "647      DELETION                         no                     \n",
       "648      DELETION                         że                     \n",
       "649      DELETION                       jest                     \n",
       "650      DELETION                         to                     \n",
       "651      DELETION                konieczność                     \n",
       "652      DELETION                          w                     \n",
       "653      DELETION                      takim                     \n",
       "654      DELETION                      razie                     \n",
       "655      DELETION                         no                     \n",
       "656      DELETION                        cóż                     \n",
       "657      DELETION                         ja                     \n",
       "658  SUBSTITUTION                        nie             akademii\n",
       "659  SUBSTITUTION                   zabieram             wypytuje\n",
       "660  SUBSTITUTION                     więcej                    o\n",
       "661  SUBSTITUTION                      czasu         konieczności\n",
       "662  SUBSTITUTION                      chyba              rozwoju\n",
       "663  SUBSTITUTION                         że                    i\n",
       "664  SUBSTITUTION                         ma               notują\n",
       "665  SUBSTITUTION                        pan               tytuły\n",
       "666  SUBSTITUTION                     jakieś             niestety\n",
       "667  SUBSTITUTION                    pytania             batyskaf\n",
       "668  SUBSTITUTION                         co             uczelnia\n",
       "669       CORRECT                         do                   do\n",
       "670       CORRECT                        tej                  tej\n",
       "671       CORRECT                   sytuacji             sytuacji\n",
       "672      DELETION                 oczywiście                     \n",
       "673      DELETION                        jak                     \n",
       "674      DELETION                najbardziej                     \n",
       "675      DELETION                          w                     \n",
       "676      DELETION                      takim                     \n",
       "677  SUBSTITUTION                      razie                  nie\n",
       "678  SUBSTITUTION                   dziękuję                 były\n",
       "679  SUBSTITUTION                       panu             bardziej\n",
       "680  SUBSTITUTION                         za                curie\n",
       "681  SUBSTITUTION                    rozmowę             dziękuje\n",
       "682  SUBSTITUTION                      życzę                 swym\n",
       "683  SUBSTITUTION                         no         nauczycielom\n",
       "684       CORRECT                       mimo                 mimo\n",
       "685      DELETION                   wszystko                     \n",
       "686       CORRECT                 spokojnego           spokojnego\n",
       "687      DELETION                       dnia                     \n",
       "688  SUBSTITUTION                          i             niejeden\n",
       "689  SUBSTITUTION                 serdecznie              powiatu\n",
       "690  SUBSTITUTION                       pana               dramie\n",
       "691  SUBSTITUTION                 pozdrawiam                  był\n",
       "692  SUBSTITUTION                         do             usunięty\n",
       "693  SUBSTITUTION                 usłyszenia                     "
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "--------------------------------------------------------------\n",
      "--------------------------------------------------------------\n",
      "--------------------------------------------------------------\n"
     ]
    }
   ],
   "source": [
    "for it in list(voicelab_record_provider.get_all_records())[:1]:\n",
    "    show_report_for(it, voicelab_experiment_repository, ajn_connections_config)"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "32e0753f-debf-4946-9e14-1876e9e26e0c",
   "metadata": {},
   "source": [
    "# Statistics of POS incorrect data"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 40,
   "id": "aec7538f-60e7-4eee-9c3a-9875135bed1f",
   "metadata": {},
   "outputs": [],
   "source": [
    "@dataclass\n",
    "class AlignDetails:\n",
    "    value_reference: Optional[str]\n",
    "    value_hypothesis: Optional[str]\n",
    "    \n",
    "    @staticmethod\n",
    "    def _get_string_or_default(value: Optional[str]) -> str:\n",
    "        return value if value is not None else '___'\n",
    "        \n",
    "    def to_string(self) -> str:\n",
    "        return f'{self._get_string_or_default(self.value_reference)} -> {self._get_string_or_default(self.value_hypothesis)}'\n",
    "    \n",
    "    def is_correct(self) -> bool:\n",
    "        return self.value_reference == self.value_hypothesis\n",
    "    \n",
    "    \n",
    "def get_align_details(repository_record: List[Any]) -> List[AlignDetails]:\n",
    "    return [\n",
    "        AlignDetails(\n",
    "            value_reference=it['step_words']['reference_word']['text'] if 'reference_word' in it['step_words'] else None,\n",
    "            value_hypothesis=it['step_words']['hypothesis_word']['text'] if 'hypothesis_word' in it['step_words'] else None\n",
    "        )\n",
    "        for it in repository_record\n",
    "    ]\n",
    "\n",
    "\n",
    "def show_stats(experiment_repository: ExperimentRepository, collections_config: CollectionsConfig):\n",
    "    aggregated_aligned = [\n",
    "        get_align_details(experiment_repository.get_property_for_key(record_id, collections_config.pos_asr_alignment_wer))\n",
    "        for record_id in list(experiment_repository.get_all_record_ids())[:]\n",
    "    ]\n",
    "    list_of_aligns = list(itertools.chain(*aggregated_aligned))\n",
    "    list_of_changes = [it.to_string() for it in list_of_aligns if not it.is_correct()]\n",
    "    unique = np.unique(np.array(list_of_changes), return_counts=True)\n",
    "    return pd.DataFrame({'values': unique[0], 'counts': unique[1]}).sort_values(by=['counts'], ascending=False).head(16)"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "264c3fd3-b890-4e24-ac8d-2756d171ccb8",
   "metadata": {},
   "source": [
    "# Luna Techmo"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 41,
   "id": "1d456e71-301c-416f-b694-03ed7ccfdee0",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/html": [
       "<div>\n",
       "<style scoped>\n",
       "    .dataframe tbody tr th:only-of-type {\n",
       "        vertical-align: middle;\n",
       "    }\n",
       "\n",
       "    .dataframe tbody tr th {\n",
       "        vertical-align: top;\n",
       "    }\n",
       "\n",
       "    .dataframe thead th {\n",
       "        text-align: right;\n",
       "    }\n",
       "</style>\n",
       "<table border=\"1\" class=\"dataframe\">\n",
       "  <thead>\n",
       "    <tr style=\"text-align: right;\">\n",
       "      <th></th>\n",
       "      <th>values</th>\n",
       "      <th>counts</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>131</th>\n",
       "      <td>NUM -&gt; ___</td>\n",
       "      <td>2290</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>192</th>\n",
       "      <td>PUNCT -&gt; ___</td>\n",
       "      <td>1479</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>114</th>\n",
       "      <td>NOUN -&gt; ___</td>\n",
       "      <td>1204</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>169</th>\n",
       "      <td>PROPN -&gt; NOUN</td>\n",
       "      <td>1177</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>45</th>\n",
       "      <td>ADV -&gt; ___</td>\n",
       "      <td>1073</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>30</th>\n",
       "      <td>ADP -&gt; ___</td>\n",
       "      <td>988</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>246</th>\n",
       "      <td>___ -&gt; NOUN</td>\n",
       "      <td>918</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>130</th>\n",
       "      <td>NUM -&gt; X</td>\n",
       "      <td>900</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>225</th>\n",
       "      <td>VERB -&gt; ___</td>\n",
       "      <td>883</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>176</th>\n",
       "      <td>PROPN -&gt; ___</td>\n",
       "      <td>823</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>162</th>\n",
       "      <td>PROPN -&gt; ADJ</td>\n",
       "      <td>662</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>252</th>\n",
       "      <td>___ -&gt; VERB</td>\n",
       "      <td>632</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>161</th>\n",
       "      <td>PRON -&gt; ___</td>\n",
       "      <td>613</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>146</th>\n",
       "      <td>PART -&gt; ___</td>\n",
       "      <td>577</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>15</th>\n",
       "      <td>ADJ -&gt; ___</td>\n",
       "      <td>560</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>73</th>\n",
       "      <td>CCONJ -&gt; ___</td>\n",
       "      <td>495</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "            values  counts\n",
       "131     NUM -> ___    2290\n",
       "192   PUNCT -> ___    1479\n",
       "114    NOUN -> ___    1204\n",
       "169  PROPN -> NOUN    1177\n",
       "45      ADV -> ___    1073\n",
       "30      ADP -> ___     988\n",
       "246    ___ -> NOUN     918\n",
       "130       NUM -> X     900\n",
       "225    VERB -> ___     883\n",
       "176   PROPN -> ___     823\n",
       "162   PROPN -> ADJ     662\n",
       "252    ___ -> VERB     632\n",
       "161    PRON -> ___     613\n",
       "146    PART -> ___     577\n",
       "15      ADJ -> ___     560\n",
       "73    CCONJ -> ___     495"
      ]
     },
     "execution_count": 41,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "show_stats(luna_experiment_repository, techmo_connections_config)"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "7f018f54-75dc-47a5-8b04-839a008a3edc",
   "metadata": {},
   "source": [
    "## Luna AJN"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 42,
   "id": "f7dd73cf-1fff-416b-9fed-d368d2fc65d0",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/html": [
       "<div>\n",
       "<style scoped>\n",
       "    .dataframe tbody tr th:only-of-type {\n",
       "        vertical-align: middle;\n",
       "    }\n",
       "\n",
       "    .dataframe tbody tr th {\n",
       "        vertical-align: top;\n",
       "    }\n",
       "\n",
       "    .dataframe thead th {\n",
       "        text-align: right;\n",
       "    }\n",
       "</style>\n",
       "<table border=\"1\" class=\"dataframe\">\n",
       "  <thead>\n",
       "    <tr style=\"text-align: right;\">\n",
       "      <th></th>\n",
       "      <th>values</th>\n",
       "      <th>counts</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>284</th>\n",
       "      <td>___ -&gt; PUNCT</td>\n",
       "      <td>3344</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>256</th>\n",
       "      <td>VERB -&gt; ___</td>\n",
       "      <td>1645</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>251</th>\n",
       "      <td>VERB -&gt; PUNCT</td>\n",
       "      <td>1582</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>279</th>\n",
       "      <td>___ -&gt; NOUN</td>\n",
       "      <td>1548</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>28</th>\n",
       "      <td>ADP -&gt; PUNCT</td>\n",
       "      <td>1522</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>52</th>\n",
       "      <td>ADV -&gt; ___</td>\n",
       "      <td>1409</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>34</th>\n",
       "      <td>ADP -&gt; ___</td>\n",
       "      <td>1316</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>124</th>\n",
       "      <td>NOUN -&gt; PUNCT</td>\n",
       "      <td>1300</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>46</th>\n",
       "      <td>ADV -&gt; PUNCT</td>\n",
       "      <td>1197</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>246</th>\n",
       "      <td>VERB -&gt; NOUN</td>\n",
       "      <td>1186</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>41</th>\n",
       "      <td>ADV -&gt; NOUN</td>\n",
       "      <td>1167</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>129</th>\n",
       "      <td>NOUN -&gt; ___</td>\n",
       "      <td>1137</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>16</th>\n",
       "      <td>ADJ -&gt; ___</td>\n",
       "      <td>879</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>162</th>\n",
       "      <td>PART -&gt; ___</td>\n",
       "      <td>876</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>178</th>\n",
       "      <td>PRON -&gt; ___</td>\n",
       "      <td>843</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>11</th>\n",
       "      <td>ADJ -&gt; PUNCT</td>\n",
       "      <td>830</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "            values  counts\n",
       "284   ___ -> PUNCT    3344\n",
       "256    VERB -> ___    1645\n",
       "251  VERB -> PUNCT    1582\n",
       "279    ___ -> NOUN    1548\n",
       "28    ADP -> PUNCT    1522\n",
       "52      ADV -> ___    1409\n",
       "34      ADP -> ___    1316\n",
       "124  NOUN -> PUNCT    1300\n",
       "46    ADV -> PUNCT    1197\n",
       "246   VERB -> NOUN    1186\n",
       "41     ADV -> NOUN    1167\n",
       "129    NOUN -> ___    1137\n",
       "16      ADJ -> ___     879\n",
       "162    PART -> ___     876\n",
       "178    PRON -> ___     843\n",
       "11    ADJ -> PUNCT     830"
      ]
     },
     "execution_count": 42,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "show_stats(luna_experiment_repository, ajn_connections_config)"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "e15befae-02ad-4517-9e97-58b3edf8c607",
   "metadata": {},
   "source": [
    "## VoiceLab Techmo"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 43,
   "id": "9acb8d8e-7359-4e89-9902-ec5067d9a65d",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/html": [
       "<div>\n",
       "<style scoped>\n",
       "    .dataframe tbody tr th:only-of-type {\n",
       "        vertical-align: middle;\n",
       "    }\n",
       "\n",
       "    .dataframe tbody tr th {\n",
       "        vertical-align: top;\n",
       "    }\n",
       "\n",
       "    .dataframe thead th {\n",
       "        text-align: right;\n",
       "    }\n",
       "</style>\n",
       "<table border=\"1\" class=\"dataframe\">\n",
       "  <thead>\n",
       "    <tr style=\"text-align: right;\">\n",
       "      <th></th>\n",
       "      <th>values</th>\n",
       "      <th>counts</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>30</th>\n",
       "      <td>ADP -&gt; ___</td>\n",
       "      <td>4527</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>121</th>\n",
       "      <td>NOUN -&gt; ___</td>\n",
       "      <td>4129</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>151</th>\n",