diff --git a/.gitignore b/.gitignore
index 7e144ff..ff076b7 100644
--- a/.gitignore
+++ b/.gitignore
@@ -174,3 +174,6 @@ cython_debug/
# option (not recommended) you can uncomment the following to ignore the entire idea folder.
#.idea/
+*~
+*.log
+*.aux
diff --git a/README.md b/README.md
index 11398e7..c9d210e 100644
--- a/README.md
+++ b/README.md
@@ -1,11 +1,11 @@
-# Modelowanie języka 2023
+# Modelowanie języka 2024
Tutaj znajdują się informacje dotyczące wykładu i ćwiczeń z
„Modelowania języka”, zajęć prowadzonych przez prof. UAM dr. hab.
-Filip Gralińskiego.
+Filip Gralińskiego i dr. Pawła Skórzewskiego.
**Uwaga** Informacje podane w plikach w podkatalogach `wyk/` i `cw/`
-nie muszą być aktualne (są to materiały przeniesione z poprzedniego
+nie muszą być aktualne (częściowo są to materiały przeniesione z poprzedniego
roku). W szczególności proszę nie wykonywać żadnych zadań z materiałów
ćwiczeniowych, chyba że wyraźnie wskazano je tutaj lub pojawiły się
jako zadanie na platformie Teams.
@@ -21,28 +21,7 @@ na trzy, mniej więcej równe objętością, działy:
## Zasady zaliczenia wykładu
-Wykład kończy się egzaminem. Z egzaminu można uzyskać zwolnienie pod
-następującymi warunkami:
-
-* ocena co najmniej dobry plus z ćwiczeń,
-* uzyskanie co najmniej 20 punktów wykładowych.
-
-Punkt wykładowy można zdobyć za różnego rodzaju aktywności związane z wykładem:
-
-* zgłoszenie (jakiegokolwiek: merytorycznego, językowego,
- technicznego) błędu w materiałach do wykładu (punkt otrzymuje
- pierwszy zgłaszający),
-* zadanie sensownego pytania na wykładzie,
-* odpowiedź na pytanie na wykładzie,
-* udział w dyskusji po wykładzie na platformie Teams (zadanie pytania
- lub odpowiedź na pytanie innego studenta).
-
-Punkt wykładowy jest zaliczony tylko pod warunkiem obecności na danym wykładzie.
-
-Dodatkowo przy liczeniu sumy punktów wykładowych obowiązują następujące ograniczenia:
-
-* liczymy maksymalnie 3 punkty na wykład,
-* z każdego działu liczymy maksymalnie 8 punktów.
+Wykład kończy się egzaminem.
## Zasady zaliczenia ćwiczeń
diff --git a/wyk/01_Jezyk.ipynb b/wyk/01_Jezyk.ipynb
new file mode 100644
index 0000000..d640520
--- /dev/null
+++ b/wyk/01_Jezyk.ipynb
@@ -0,0 +1,773 @@
+{
+ "cells": [
+ {
+ "cell_type": "markdown",
+ "metadata": {},
+ "source": [
+ "![Logo 1](https://git.wmi.amu.edu.pl/AITech/Szablon/raw/branch/master/Logotyp_AITech1.jpg)\n",
+ "
\n",
+ "
Modelowanie języka \n",
+ " 1. Język i jego zapis [wykład] \n",
+ " Filip Graliński (2022) \n",
+ "\n",
+ "\n",
+ "![Logo 2](https://git.wmi.amu.edu.pl/AITech/Szablon/raw/branch/master/Logotyp_AITech2.jpg)\n",
+ "\n"
+ ]
+ },
+ {
+ "cell_type": "markdown",
+ "metadata": {},
+ "source": [
+ "## Język — różne perspektywy\n",
+ "\n"
+ ]
+ },
+ {
+ "cell_type": "markdown",
+ "metadata": {},
+ "source": [
+ "### Słowo wstępne\n",
+ "\n"
+ ]
+ },
+ {
+ "cell_type": "markdown",
+ "metadata": {},
+ "source": [
+ "W matematyce istnieją dwa spojrzenia na rzeczywistość: ciągłe i dyskretne.\n",
+ "\n",
+ "Otaczająca nas rzeczywistość fizyczna jest z natury ciągła\n",
+ "(przynajmniej jeśli nie operujemy w mikroskali), lecz język\n",
+ "jest dyskretnym wyłomem w ciągłej rzeczywistości.\n",
+ "\n"
+ ]
+ },
+ {
+ "cell_type": "markdown",
+ "metadata": {},
+ "source": [
+ "### Lingwistyka matematyczna\n",
+ "\n"
+ ]
+ },
+ {
+ "cell_type": "markdown",
+ "metadata": {},
+ "source": [
+ "Przypomnijmy sobie definicję języka przyjętą w lingwistyce\n",
+ "matematycznej, w kontekście, na przykład, teorii automatów.\n",
+ "\n",
+ "****Alfabetem**** nazywamy skończony zbiór symboli.\n",
+ "\n",
+ "****Łańcuchem**** (****napisem****) nad alfabetem $\\Sigma$ nazywamy dowolny, skończony,\n",
+ "ciąg złożony z symboli z $\\Sigma$.\n",
+ "\n",
+ "****Językiem**** nazywamy dowolny, skończony bądź nieskończony, zbiór łańcuchów.\n",
+ "\n",
+ "W tym formalnym ujęciu językami są na przykład następujące zbiory:\n",
+ "\n",
+ "- $\\{\\mathit{poniedziałek},\\mathit{wtorek},\\mathit{środa},\\mathit{czwartek},\\mathit{piątek},\\mathit{sobota},\\mathit{niedziela}\\}$\n",
+ "- $\\{\\mathit{ab},\\mathit{abb},\\mathit{abbb},\\mathit{abbbb},\\ldots\\}$\n",
+ "\n",
+ "To podejście, z jednej strony oczywiście nie do końca się pokrywa się z potocznym\n",
+ "rozumieniem słowa *język*, z drugiej kojarzy nam się z takimi\n",
+ "narzędziami informatyki jak wyrażenia regularne, automaty skończenie\n",
+ "stanowe czy gramatyki języków programowania.\n",
+ "\n"
+ ]
+ },
+ {
+ "cell_type": "code",
+ "execution_count": 1,
+ "metadata": {},
+ "outputs": [
+ {
+ "name": "stdout",
+ "output_type": "stream",
+ "text": [
+ "abbb"
+ ]
+ }
+ ],
+ "source": [
+ "import regex as re\n",
+ "rx = re.compile(r'ab+')\n",
+ "\n",
+ "rx.search('żabbba').group(0)"
+ ]
+ },
+ {
+ "cell_type": "code",
+ "execution_count": 1,
+ "metadata": {},
+ "outputs": [
+ {
+ "name": "stdout",
+ "output_type": "stream",
+ "text": [
+ "abbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbbb"
+ ]
+ }
+ ],
+ "source": [
+ "import rstr\n",
+ "\n",
+ "rstr.xeger(r'ab+')"
+ ]
+ },
+ {
+ "cell_type": "markdown",
+ "metadata": {},
+ "source": [
+ "### Ujęcie probabilistyczne\n",
+ "\n"
+ ]
+ },
+ {
+ "cell_type": "markdown",
+ "metadata": {},
+ "source": [
+ "Na tym wykładzie przyjmiemy inną perspektywą, częściowo ciągłą, opartą\n",
+ "na probabilistyce. Język będziemy definiować poprzez ****rozkład\n",
+ "prawdopodobieństwa****: sensownym wypowiedziom czy tekstom będziemy\n",
+ "przypisywać stosunkowe wysokie prawdopodobieństwo, „ułomnym” tekstom — niższe (być może zerowe).\n",
+ "\n",
+ "Na ogół nie mamy jednak do czynienia z językiem jako takim tylko z\n",
+ "jego przybliżeniami, ****modelami**** (model może być lepszy lub gorszy,\n",
+ "ale przynajmniej powinien być użyteczny…). Formalnie $M$ nazywamy\n",
+ "modelem języka (nad skończonym alfabetem $\\Sigma$), jeśli określa dyskretny rozkład prawdopodobieństwa $P_M$:\n",
+ "\n",
+ "$$P_M \\colon \\Sigma^{*} \\rightarrow [0,1].$$\n",
+ "\n",
+ "Rzecz jasna, skoro mamy do czynienia z rozkładem prawdopodobieństwa, to:\n",
+ "\n",
+ "$$\\sum_{\\alpha \\in \\Sigma^{*}} P_M(\\alpha) = 1.$$\n",
+ "\n",
+ "Jeśli $M$ ma być modelem języka polskiego, oczekiwalibyśmy, że dla\n",
+ "napisów:\n",
+ "\n",
+ "- $z_1$ — *W tym stanie rzeczy pan Ignacy coraz częściej myślał o Wokulskim.*\n",
+ "- $z_2$ — *Po wypełniony zbiornik pełny i należne kwotę, usłyszała w attendant*\n",
+ "- $z_3$ — *xxxxyźźźźźit backspace hoooooooooop x y z*\n",
+ "\n",
+ "zachodzić będzie:\n",
+ "\n",
+ "$$ P_M(z_1) > P_M(z_2) > P_M(z_3). $$\n",
+ "\n",
+ "****Pytanie**** Jakiej konkretnie wartości prawdopodobieństwa\n",
+ "spodziewalibyśmy się dla zdania *Dzisiaj rano kupiłem w piekarni sześć bułek*\n",
+ "dla sensownego modelu języka polskiego?\n",
+ "\n",
+ "Moglibyśmy sprowadzić tę definicję języka do tej „dyskretnej”, tzn.\n",
+ "moglibyśmy przyjąć, że łańcuch $\\alpha$ należy do języka wyznaczonego\n",
+ "przez model $M$, jeśli $P_M(\\alpha) > 0$.\n",
+ "\n",
+ "****Pytanie**** Czy moglibyśmy w ten sposób opisać język nieskończony? Czy może istnieć\n",
+ "dyskretny rozkład prawdopodobieństwa dla nieskończonego zbioru?\n",
+ "\n"
+ ]
+ },
+ {
+ "cell_type": "markdown",
+ "metadata": {},
+ "source": [
+ "### Co jest symbolem?\n",
+ "\n"
+ ]
+ },
+ {
+ "cell_type": "markdown",
+ "metadata": {},
+ "source": [
+ "Model języka daje rozkład prawdopodobieństwa nad zbiorem łańcuchów\n",
+ "opartym na skończonym alfabecie, tj. zbiorze symboli. W praktyce\n",
+ "alfabet nie musi być zgodny z potocznym czy językoznawczym rozumieniem\n",
+ "tego słowa. To znaczy alfabet może być zbiorem znaków (liter), ale\n",
+ "modelować język możemy też przyjmując inny typ symboli: sylaby,\n",
+ "morfemy (cząstki wyrazów) czy po prostu całe wyrazy.\n",
+ "\n",
+ "Powinniśmy przy tym pamiętać, że, koniec końców, w pamięci komputera\n",
+ "wszelkiego rodzaju łańcuchy są zapisywane jako ciągi zer i jedynek — bitów.\n",
+ "Omówmy pokrótce techniczną stronę modelowania języka.\n",
+ "\n"
+ ]
+ },
+ {
+ "cell_type": "markdown",
+ "metadata": {},
+ "source": [
+ "## Kodowanie znaków\n",
+ "\n"
+ ]
+ },
+ {
+ "cell_type": "markdown",
+ "metadata": {},
+ "source": [
+ "### Cóż może być prostszego od pliku tekstowego?\n",
+ "\n"
+ ]
+ },
+ {
+ "cell_type": "markdown",
+ "metadata": {},
+ "source": [
+ " Ala ma kota.\n",
+ "\n",
+ "Komputer nic nie wie o literach.\n",
+ "\n",
+ "… w rzeczywistości operuje tylko na liczbach …\n",
+ "\n",
+ "… czy raczej na zerach i jedynkach …\n",
+ "\n",
+ "… a tak naprawdę na ciągłym sygnale elektrycznym …\n",
+ "\n",
+ "![img](./01_Jezyk/digitalsignal.jpg)\n",
+ "\n",
+ "… zera i jedynki są w naszej głowie …\n",
+ "\n",
+ "… co jest dziwne, *naprawdę* dziwne …\n",
+ "\n",
+ "… ale nikt normalny się tym nie przejmuje.\n",
+ "\n"
+ ]
+ },
+ {
+ "cell_type": "markdown",
+ "metadata": {},
+ "source": [
+ "### Jak zakodować literę?\n",
+ "\n"
+ ]
+ },
+ {
+ "cell_type": "markdown",
+ "metadata": {},
+ "source": [
+ "Zakodowanie pikseli składających się na kształtu (****glyfu****) litery A\n",
+ "*oczywiście* nie jest dobrym pomysłem.\n",
+ "\n",
+ "![img](./01_Jezyk/raster.png)\n",
+ "\n",
+ "Nie, potrzebujemy *arbitralnego* kodowania dla wszystkich możliwych\n",
+ "kształtów litery A (*w naszych głowach*): A, $\\mathcal{A}$,\n",
+ "$\\mathbb{A}$, $\\mathfrak{A}$ powinny otrzymać ten sam kod, powiedzmy 65\n",
+ "(binarnie: 1000001).\n",
+ "\n"
+ ]
+ },
+ {
+ "cell_type": "markdown",
+ "metadata": {},
+ "source": [
+ "### ASCII\n",
+ "\n"
+ ]
+ },
+ {
+ "cell_type": "markdown",
+ "metadata": {},
+ "source": [
+ "ASCII to 7-bitowy (****nie**** 8-bitowy!) system kodowania znaków.\n",
+ "\n"
+ ]
+ },
+ {
+ "cell_type": "code",
+ "execution_count": 1,
+ "metadata": {},
+ "outputs": [
+ {
+ "name": "stdout",
+ "output_type": "stream",
+ "text": [
+ "40: (\n",
+ "41: )\n",
+ "42: *\n",
+ "43: +\n",
+ "44: ,\n",
+ "45: -\n",
+ "46: .\n",
+ "47: /\n",
+ "48: 0\n",
+ "49: 1\n",
+ "50: 2\n",
+ "51: 3\n",
+ "52: 4\n",
+ "53: 5\n",
+ "54: 6\n",
+ "55: 7\n",
+ "56: 8\n",
+ "57: 9\n",
+ "58: :\n",
+ "59: ;\n",
+ "60: <\n",
+ "61: =\n",
+ "62: >\n",
+ "63: ?\n",
+ "64: @\n",
+ "65: A\n",
+ "66: B\n",
+ "67: C\n",
+ "68: D\n",
+ "69: E\n",
+ "70: F\n",
+ "71: G\n",
+ "72: H\n",
+ "73: I\n",
+ "74: J\n",
+ "75: K\n",
+ "76: L\n",
+ "77: M\n",
+ "78: N\n",
+ "79: O\n",
+ "80: P\n",
+ "81: Q\n",
+ "82: R\n",
+ "83: S\n",
+ "84: T\n",
+ "85: U\n",
+ "86: V\n",
+ "87: W\n",
+ "88: X\n",
+ "89: Y\n",
+ "90: Z\n",
+ "91: [\n",
+ "92: \\\n",
+ "93: ]\n",
+ "94: ^\n",
+ "95: _\n",
+ "96: `\n",
+ "97: a\n",
+ "98: b\n",
+ "99: c\n",
+ "100: d\n",
+ "101: e\n",
+ "102: f\n",
+ "103: g\n",
+ "104: h\n",
+ "105: i\n",
+ "106: j\n",
+ "107: k\n",
+ "108: l\n",
+ "109: m\n",
+ "110: n\n",
+ "111: o\n",
+ "112: p\n",
+ "113: q\n",
+ "114: r\n",
+ "115: s\n",
+ "116: t\n",
+ "117: u\n",
+ "118: v\n",
+ "119: w\n",
+ "120: x\n",
+ "121: y\n",
+ "122: z\n",
+ "123: {\n",
+ "124: |\n",
+ "125: }\n",
+ "126: ~\n",
+ "127: \n"
+ ]
+ }
+ ],
+ "source": [
+ "for code in range(40, 128):\n",
+ " print(f'{code}: {chr(code)}')"
+ ]
+ },
+ {
+ "cell_type": "markdown",
+ "metadata": {},
+ "source": [
+ "### Jak zejść na poziom bitów?\n",
+ "\n"
+ ]
+ },
+ {
+ "cell_type": "markdown",
+ "metadata": {},
+ "source": [
+ "#### Linux — wiersz poleceń\n",
+ "\n"
+ ]
+ },
+ {
+ "cell_type": "markdown",
+ "metadata": {},
+ "source": [
+ "Linux command line:\n",
+ "\n",
+ " $ echo 'Ala ma kota' > file.txt\n",
+ " $ hexdump -C file.txt\n",
+ " 00000000 41 6c 61 20 6d 61 20 6b 6f 74 61 0a |Ala ma kota.|\n",
+ " 0000000c\n",
+ "\n"
+ ]
+ },
+ {
+ "cell_type": "markdown",
+ "metadata": {},
+ "source": [
+ "#### Edytor tekstu (Emacs)\n",
+ "\n"
+ ]
+ },
+ {
+ "cell_type": "markdown",
+ "metadata": {},
+ "source": [
+ "![img](./01_Jezyk/hexl-mode.png)\n",
+ "\n"
+ ]
+ },
+ {
+ "cell_type": "markdown",
+ "metadata": {},
+ "source": [
+ "#### Uwaga!\n",
+ "\n"
+ ]
+ },
+ {
+ "cell_type": "markdown",
+ "metadata": {},
+ "source": [
+ "- kiedy dzieje się coś dziwnego, sprawdź, co tak *naprawdę* jest w pliku\n",
+ "- ASCII jest 7-bitowym kodowaniem (128 znaków)\n",
+ " - choć zazwyczaj uzupełnionym (ang. *padded*) do 8 bitów\n",
+ " - nie mów plik *plik ASCII*, kiedy masz na myśli *prosty/czysty plik tekstowy* (ang. *plain text file*)\n",
+ "\n"
+ ]
+ },
+ {
+ "cell_type": "markdown",
+ "metadata": {},
+ "source": [
+ "### Higiena plików tekstowych\n",
+ "\n"
+ ]
+ },
+ {
+ "cell_type": "markdown",
+ "metadata": {},
+ "source": [
+ "#### Piekło końca wiersza\n",
+ "\n"
+ ]
+ },
+ {
+ "cell_type": "markdown",
+ "metadata": {},
+ "source": [
+ "![img](./01_Jezyk/dante.jpg)\n",
+ "\n",
+ "Więcej na [https://re-research.pl/pl/post/2017-01-28-00042-anatomia-pliku-tekstowego-2.html](https://re-research.pl/pl/post/2017-01-28-00042-anatomia-pliku-tekstowego-2.html)\n",
+ "\n"
+ ]
+ },
+ {
+ "cell_type": "markdown",
+ "metadata": {},
+ "source": [
+ "#### Dobre rady\n",
+ "\n"
+ ]
+ },
+ {
+ "cell_type": "markdown",
+ "metadata": {},
+ "source": [
+ "- żadnych niepotrzebnych spacji na końcu wiersza\n",
+ "\n",
+ "- żadnych niepotrzebnych pustych wierszy na końcu pliku\n",
+ "\n",
+ "- … ale ostatni wiersz powinien zakończyć się znakiem końca wiersza\n",
+ "\n",
+ "- nie używać znaków tabulacji (zamiast tego 4 spacje)\n",
+ " - wyjątek: pliki TSV\n",
+ " - wyjątek: pliki Makefile\n",
+ "\n",
+ "- uwaga na niestandardowe spacje i dziwne znaki o zerowej długości\n",
+ "\n"
+ ]
+ },
+ {
+ "cell_type": "markdown",
+ "metadata": {},
+ "source": [
+ "### Unikod\n",
+ "\n"
+ ]
+ },
+ {
+ "cell_type": "markdown",
+ "metadata": {},
+ "source": [
+ "ASCII obejmuje 128 znaków: litery alfabetu łacińskiego (właściwie angielskiego),\n",
+ "cyfry, znaki interpunkcyjne, znaki specjalne itd.\n",
+ "\n",
+ "Co z pozostałymi znakami? Polskimi ogonkami, czeskimi haczykami,\n",
+ "francuskimi akcentami, cyrylicą, koreańskim alfabetem, chińskimi\n",
+ "znakami, rongorongo?\n",
+ "\n",
+ "워싱턴, 부산, 삼성\n",
+ "\n",
+ "Rozwiązaniem jest Unikod (ang. *Unicode*) system, który przypisuje\n",
+ "znakom używanym przez ludzkość liczby (kody, ang. *code points*).\n",
+ "\n",
+ "| Znak|Kod ASCII|Kod Unikodowy|\n",
+ "|---|---|---|\n",
+ "| 9|57|57|\n",
+ "| a|97|97|\n",
+ "| ą|-|261|\n",
+ "| ł|-|322|\n",
+ "| $\\aleph$|-|1488|\n",
+ "| ặ|-|7861|\n",
+ "| ☣|-|9763|\n",
+ "| 😇|-|128519|\n",
+ "\n"
+ ]
+ },
+ {
+ "cell_type": "markdown",
+ "metadata": {},
+ "source": [
+ "### UTF-8\n",
+ "\n"
+ ]
+ },
+ {
+ "cell_type": "markdown",
+ "metadata": {},
+ "source": [
+ "Kody znaków są pojęciem abstrakcyjnym. Potrzebujemy konkretnego ****kodowania****\n",
+ "by zamienić kody w sekwencję bajtów. Najpopularniejszym kodowaniem jest UTF-8.\n",
+ "\n",
+ "W kodowaniu UTF-8 znaki zapisywane za pomocą 1, 2, 3, 4, 5 lub 6 bajtów\n",
+ "(w praktyce — raczej to 4 bajtów).\n",
+ "\n",
+ "| Znak|Kod Unikodowy|Szesnastkowo|UTF-8 (binarnie)|\n",
+ "|---|---|---|---|\n",
+ "| 9|57|U+0049|01001001|\n",
+ "| a|97|U+0061|01100001|\n",
+ "| ą|261|U+0105|11000100:10000101|\n",
+ "| ł|322|U+0142|11000101:10000010|\n",
+ "| $\\aleph$|1488|U+05D0|11010111:10010000|\n",
+ "| ặ|7861|U+1EB7|11100001:10111010:10110111|\n",
+ "| ☣|9763|U+2623|11100010:10011000:10100011|\n",
+ "| 😇|128519|U+1f607|11110000:10011111:10011000:10000111|\n",
+ "\n"
+ ]
+ },
+ {
+ "cell_type": "markdown",
+ "metadata": {},
+ "source": [
+ "#### UTF-8 — ogólny schemat zamiany kodu na bajty\n",
+ "\n"
+ ]
+ },
+ {
+ "cell_type": "markdown",
+ "metadata": {},
+ "source": [
+ "- 0x00 do 0x7F – 0xxxxxxx,\n",
+ "- 0x80 do 0x7FF – 110xxxxx 10xxxxxx\n",
+ "- 0x800 do 0xFFFF — 1110xxxx 10xxxxxx 10xxxxxx\n",
+ "- 0x10000 do 0x1FFFFF – 11110xxx 10xxxxxx 10xxxxxx 10xxxxxx\n",
+ "- 0x200000 do 0x3FFFFFF – 111110xx 10xxxxxx 10xxxxxx 10xxxxxx 10xxxxxx\n",
+ "- 0x4000000 do 0x7FFFFFFF – 1111110x 10xxxxxx 10xxxxxx 10xxxxxx 10xxxxxx 10xxxxxx\n",
+ "\n",
+ "Symbol x oznacza znaczący bit.\n",
+ "\n"
+ ]
+ },
+ {
+ "cell_type": "markdown",
+ "metadata": {},
+ "source": [
+ "#### *Źdźbło* to ile bajtów w UTF-8?\n",
+ "\n"
+ ]
+ },
+ {
+ "cell_type": "markdown",
+ "metadata": {},
+ "source": [
+ "Jeśli wczytać jako wiersz w języku C, 11 bajtów!\n",
+ "\n",
+ "![img](./01_Jezyk/zdzblo.png)\n",
+ "\n"
+ ]
+ },
+ {
+ "cell_type": "markdown",
+ "metadata": {},
+ "source": [
+ "### Dlaczego UTF-8 jest doskonałym systemem kodowania?\n",
+ "\n"
+ ]
+ },
+ {
+ "cell_type": "markdown",
+ "metadata": {},
+ "source": [
+ "- wstecznie kompatybilny z ASCII\n",
+ " - plik ASCII jest poprawnym plikiem UTF-8\n",
+ "- nie zajmuje dużo miejsca\n",
+ " - chyba że w tekście jest dużo „dziwnych” znaków\n",
+ "- proste grepowanie działa\n",
+ " - `grep UAM text-in-utf8.txt` zadziała\n",
+ " - ale nawet nie próbuj: `grep SRPOL text-in-utf16.txt`\n",
+ "\n"
+ ]
+ },
+ {
+ "cell_type": "markdown",
+ "metadata": {},
+ "source": [
+ "### Porady\n",
+ "\n"
+ ]
+ },
+ {
+ "cell_type": "markdown",
+ "metadata": {},
+ "source": [
+ "- zawsze używaj UTF-8\n",
+ " - bądź asertywny! jeśli w pracy każą używać czegoś innego — rezygnuj z pracy\n",
+ " - **NIE** używaj innych unikodowych kodowań: UTF-16, UTF-32, UCS-2\n",
+ " - **NIE** używaj nieunikodowych systemów kodowania\n",
+ " - ISO-8859-2, Windows-1250, Mazovia, IEA Świerk, …\n",
+ "- uwaga na pułapki UTF-8\n",
+ " - ustalenie długości napisu w znakach wymaga przejścia znak po znaku\n",
+ " - jeśli napis w kodowaniu UTF-8 zajmuje 9 bajtów, ile to znaków?\n",
+ " 3, 4, 5, 6, 7, 8 lub 9!\n",
+ "\n"
+ ]
+ },
+ {
+ "cell_type": "markdown",
+ "metadata": {},
+ "source": [
+ "### **NIE** używaj sekwencji BOM\n",
+ "\n"
+ ]
+ },
+ {
+ "cell_type": "markdown",
+ "metadata": {},
+ "source": [
+ "![img](./01_Jezyk/evil-bom.png)\n",
+ "\n"
+ ]
+ },
+ {
+ "cell_type": "markdown",
+ "metadata": {},
+ "source": [
+ "### Unikod/UTF-8 a języki programowania\n",
+ "\n"
+ ]
+ },
+ {
+ "cell_type": "markdown",
+ "metadata": {},
+ "source": [
+ "Pamiętaj, żeby być konsekwentnym!\n",
+ "\n",
+ "- kodowanie kodu źródłowego (literały!)\n",
+ " - czasami podawane na początku pliku\n",
+ " - … albo brane z ustawień *locale*\n",
+ " - … albo — domyślnie — UTF-8 (w nowszych językach programowania)\n",
+ "- kodowanie standardowego wejścia/wyjścia i plików\n",
+ "- jak sekwencje bajtów są interpretowane w czasie działania programu?\n",
+ " - *Źdźbło* jest łańcuchem złożonym z 6 czy 9 elementów??\n",
+ " - 9 bajtów\n",
+ " - 6 kodów\n",
+ " - `\"Źdźbło\"[1]` …\n",
+ " - `d`\n",
+ " - … albo śmieci\n",
+ "\n"
+ ]
+ },
+ {
+ "cell_type": "markdown",
+ "metadata": {},
+ "source": [
+ "#### Python 2\n",
+ "\n"
+ ]
+ },
+ {
+ "cell_type": "code",
+ "execution_count": 1,
+ "metadata": {},
+ "outputs": [],
+ "source": [
+ "#!/usr/bin/python2\n",
+ "# -*- coding: utf-8 -*-\n",
+ "import sys\n",
+ "for line in sys.stdin:\n",
+ " line = line.decode('utf-8').rstrip()\n",
+ " if \"źdźbło\".decode('utf-8') in line:\n",
+ " print len(line), ' ', line"
+ ]
+ },
+ {
+ "cell_type": "markdown",
+ "metadata": {},
+ "source": [
+ "#### Python3\n",
+ "\n"
+ ]
+ },
+ {
+ "cell_type": "code",
+ "execution_count": 1,
+ "metadata": {},
+ "outputs": [],
+ "source": [
+ "#!/usr/bin/python3\n",
+ "import sys\n",
+ "for line in sys.stdin:\n",
+ " line = line.strip()\n",
+ " if \"źdźbło\" in line:\n",
+ " print(len(line), ' ', line)"
+ ]
+ },
+ {
+ "cell_type": "markdown",
+ "metadata": {},
+ "source": [
+ "Uwaga: zakładając, że zmienna środowiskowa `LANG` jest ustawiona na UTF-8.\n",
+ "\n"
+ ]
+ }
+ ],
+ "metadata": {
+ "kernelspec": {
+ "display_name": "Python 3 (ipykernel)",
+ "language": "python",
+ "name": "python3"
+ },
+ "language_info": {
+ "codemirror_mode": {
+ "name": "ipython",
+ "version": 3
+ },
+ "file_extension": ".py",
+ "mimetype": "text/x-python",
+ "name": "python",
+ "nbconvert_exporter": "python",
+ "pygments_lexer": "ipython3",
+ "version": "3.10.9"
+ },
+ "org": null
+ },
+ "nbformat": 4,
+ "nbformat_minor": 1
+}
diff --git a/wyk/01_Jezyk.org b/wyk/01_Jezyk.org
new file mode 100644
index 0000000..efcd79e
Binary files /dev/null and b/wyk/01_Jezyk.org differ
diff --git a/wyk/01_Jezyk/dante.jpg b/wyk/01_Jezyk/dante.jpg
new file mode 100644
index 0000000..6b44156
Binary files /dev/null and b/wyk/01_Jezyk/dante.jpg differ
diff --git a/wyk/01_Jezyk/digitalsignal.jpg b/wyk/01_Jezyk/digitalsignal.jpg
new file mode 100644
index 0000000..8cffde0
Binary files /dev/null and b/wyk/01_Jezyk/digitalsignal.jpg differ
diff --git a/wyk/01_Jezyk/evil-bom.png b/wyk/01_Jezyk/evil-bom.png
new file mode 100644
index 0000000..ce9961a
Binary files /dev/null and b/wyk/01_Jezyk/evil-bom.png differ
diff --git a/wyk/01_Jezyk/hexl-mode.png b/wyk/01_Jezyk/hexl-mode.png
new file mode 100644
index 0000000..bb8d712
Binary files /dev/null and b/wyk/01_Jezyk/hexl-mode.png differ
diff --git a/wyk/01_Jezyk/raster.png b/wyk/01_Jezyk/raster.png
new file mode 100644
index 0000000..1113dca
Binary files /dev/null and b/wyk/01_Jezyk/raster.png differ
diff --git a/wyk/01_Jezyk/zdzblo.pdf b/wyk/01_Jezyk/zdzblo.pdf
new file mode 100644
index 0000000..9db7742
Binary files /dev/null and b/wyk/01_Jezyk/zdzblo.pdf differ
diff --git a/wyk/01_Jezyk/zdzblo.png b/wyk/01_Jezyk/zdzblo.png
new file mode 100644
index 0000000..0632af1
Binary files /dev/null and b/wyk/01_Jezyk/zdzblo.png differ
diff --git a/wyk/01_Jezyk/zdzblo.svg b/wyk/01_Jezyk/zdzblo.svg
new file mode 100644
index 0000000..89b2a5a
--- /dev/null
+++ b/wyk/01_Jezyk/zdzblo.svg
@@ -0,0 +1,320 @@
+
+
+
+
+Created by potrace 1.16, written by Peter Selinger 2001-2019
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
diff --git a/wyk/01_Jezyk/zdzblo.tex b/wyk/01_Jezyk/zdzblo.tex
new file mode 100644
index 0000000..fefc053
--- /dev/null
+++ b/wyk/01_Jezyk/zdzblo.tex
@@ -0,0 +1,39 @@
+
+\documentclass{article}
+\usepackage[a6paper]{geometry}
+\usepackage[T1]{fontenc}
+\usepackage{bytefield}
+\thispagestyle{empty}
+\begin{document}
+
+\begin{bytefield}{8}
+ \bitheader[endianness=big]{0-7} \\
+ \begin{leftwordgroup}{Ź}
+ \bitboxes{1}{11000101} \\
+ \bitboxes{1}{10111001}
+ \end{leftwordgroup} \\
+ \begin{leftwordgroup}{d}
+ \bitboxes{1}{01100100}
+ \end{leftwordgroup} \\
+ \begin{leftwordgroup}{ź}
+ \bitboxes{1}{11000101} \\
+ \bitboxes{1}{10111010}
+ \end{leftwordgroup} \\
+ \begin{leftwordgroup}{b}
+ \bitboxes{1}{01100010}
+ \end{leftwordgroup} \\
+ \begin{leftwordgroup}{ł}
+ \bitboxes{1}{11000101} \\
+ \bitboxes{1}{10000010}
+ \end{leftwordgroup} \\
+ \begin{leftwordgroup}{o}
+ \bitboxes{1}{01101111}
+ \end{leftwordgroup} \\
+ \begin{leftwordgroup}{\textit{koniec wiersza}}
+ \bitboxes{1}{00001010}
+ \end{leftwordgroup} \\
+ \begin{leftwordgroup}{\textit{koniec napisu}}
+ \bitboxes{1}{00000000}
+ \end{leftwordgroup}
+ \end{bytefield}
+\end{document}
diff --git a/wyk/02_Jezyki.ipynb b/wyk/02_Jezyki.ipynb
new file mode 100644
index 0000000..10ca574
--- /dev/null
+++ b/wyk/02_Jezyki.ipynb
@@ -0,0 +1,533 @@
+{
+ "cells": [
+ {
+ "cell_type": "markdown",
+ "metadata": {},
+ "source": [
+ "![Logo 1](https://git.wmi.amu.edu.pl/AITech/Szablon/raw/branch/master/Logotyp_AITech1.jpg)\n",
+ "\n",
+ "
Modelowanie języka \n",
+ " 02. Języki i ich prawa statystyczne [wykład] \n",
+ " Filip Graliński (2022) \n",
+ "\n",
+ "\n",
+ "![Logo 2](https://git.wmi.amu.edu.pl/AITech/Szablon/raw/branch/master/Logotyp_AITech2.jpg)\n",
+ "\n"
+ ]
+ },
+ {
+ "cell_type": "markdown",
+ "metadata": {},
+ "source": [
+ "## Języki i ich prawa statystyczne\n",
+ "\n"
+ ]
+ },
+ {
+ "cell_type": "markdown",
+ "metadata": {},
+ "source": [
+ "Jakim rozkładom statystycznym podlegają języki?\n",
+ "\n"
+ ]
+ },
+ {
+ "cell_type": "markdown",
+ "metadata": {},
+ "source": [
+ "### Język naturalny albo „Pan Tadeusz” w liczbach\n",
+ "\n"
+ ]
+ },
+ {
+ "cell_type": "markdown",
+ "metadata": {},
+ "source": [
+ "Przygotujmy najpierw „infrastrukturę” do *segmentacji* tekstu na różnego rodzaju jednostki.\n",
+ "Używać będziemy generatorów.\n",
+ "\n",
+ "**Pytanie** Dlaczego generatory zamiast list?\n",
+ "\n"
+ ]
+ },
+ {
+ "cell_type": "code",
+ "execution_count": 1,
+ "metadata": {},
+ "outputs": [
+ {
+ "name": "stdout",
+ "output_type": "stream",
+ "text": [
+ "Księga pierwsza\n",
+ "\n",
+ "\n",
+ "\n",
+ "Gospodarstwo\n",
+ "\n",
+ "Powrót pani"
+ ]
+ }
+ ],
+ "source": [
+ "import requests\n",
+ "\n",
+ "url = 'https://wolnelektury.pl/media/book/txt/pan-tadeusz.txt'\n",
+ "pan_tadeusz = requests.get(url).content.decode('utf-8')\n",
+ "\n",
+ "pan_tadeusz[100:150]"
+ ]
+ },
+ {
+ "cell_type": "markdown",
+ "metadata": {},
+ "source": [
+ "#### Znaki\n",
+ "\n"
+ ]
+ },
+ {
+ "cell_type": "code",
+ "execution_count": 1,
+ "metadata": {},
+ "outputs": [
+ {
+ "name": "stdout",
+ "output_type": "stream",
+ "text": [
+ "['K', 's', 'i', 'ę', 'g', 'a', ' ', 'p', 'i', 'e', 'r', 'w', 's', 'z', 'a', '\\r', '\\n', '\\r', '\\n', '\\r', '\\n', '\\r', '\\n', 'G', 'o', 's', 'p', 'o', 'd', 'a', 'r', 's', 't', 'w', 'o', '\\r', '\\n', '\\r', '\\n', 'P', 'o', 'w', 'r', 'ó', 't', ' ', 'p', 'a', 'n', 'i']"
+ ]
+ }
+ ],
+ "source": [
+ "from itertools import islice\n",
+ "\n",
+ "def get_characters(t):\n",
+ " yield from t\n",
+ "\n",
+ "list(islice(get_characters(pan_tadeusz), 100, 150))"
+ ]
+ },
+ {
+ "cell_type": "code",
+ "execution_count": 1,
+ "metadata": {},
+ "outputs": [
+ {
+ "name": "stdout",
+ "output_type": "stream",
+ "text": [
+ "Counter({' ': 63444, 'a': 30979, 'i': 29353, 'e': 25343, 'o': 23050, 'z': 22741, 'n': 15505, 'r': 15328, 's': 15255, 'w': 14625, 'c': 14153, 'y': 13732, 'k': 12362, 'd': 11465, '\\r': 10851, '\\n': 10851, 't': 10757, 'm': 10269, 'ł': 10059, ',': 9130, 'p': 8031, 'u': 7699, 'l': 6677, 'j': 6586, 'b': 5753, 'ę': 5534, 'ą': 4794, 'g': 4775, 'h': 3915, 'ż': 3334, 'ó': 3097, 'ś': 2524, '.': 2380, 'ć': 1956, ';': 1445, 'P': 1265, 'W': 1258, ':': 1152, '!': 1083, 'S': 1045, 'T': 971, 'I': 795, 'N': 793, 'Z': 785, 'J': 729, '—': 720, 'A': 698, 'K': 683, 'ń': 651, 'M': 585, 'B': 567, 'O': 567, 'C': 556, 'D': 552, '«': 540, '»': 538, 'R': 489, '?': 441, 'ź': 414, 'f': 386, 'G': 358, 'L': 316, 'H': 309, 'Ż': 219, 'U': 184, '…': 157, '*': 150, '(': 76, ')': 76, 'Ś': 71, 'F': 47, 'é': 43, '-': 33, 'Ł': 24, 'E': 23, '/': 19, 'Ó': 13, '8': 10, '9': 8, '2': 6, 'v': 5, 'Ź': 4, '1': 4, '3': 3, 'x': 3, 'V': 3, '7': 2, '4': 2, '5': 2, 'q': 2, 'æ': 2, 'à': 1, 'Ć': 1, '6': 1, '0': 1})"
+ ]
+ }
+ ],
+ "source": [
+ "from collections import Counter\n",
+ "\n",
+ "c = Counter(get_characters(pan_tadeusz))\n",
+ "\n",
+ "c"
+ ]
+ },
+ {
+ "cell_type": "markdown",
+ "metadata": {},
+ "source": [
+ "Napiszmy pomocniczą funkcję, która zwraca **listę frekwencyjną**.\n",
+ "\n",
+ "Counter({' ': 63444, 'a': 30979, 'i': 29353, 'e': 25343, 'o': 23050, 'z': 22741, 'n': 15505, 'r': 15328, 's': 15255, 'w': 14625, 'c': 14153, 'y': 13732, 'k': 12362, 'd': 11465, '\\r': 10851, '\\n': 10851, 't': 10757, 'm': 10269, 'ł': 10059, ',': 9130, 'p': 8031, 'u': 7699, 'l': 6677, 'j': 6586, 'b': 5753, 'ę': 5534, 'ą': 4794, 'g': 4775, 'h': 3915, 'ż': 3334, 'ó': 3097, 'ś': 2524, '.': 2380, 'ć': 1956, ';': 1445, 'P': 1265, 'W': 1258, ':': 1152, '!': 1083, 'S': 1045, 'T': 971, 'I': 795, 'N': 793, 'Z': 785, 'J': 729, '—': 720, 'A': 698, 'K': 683, 'ń': 651, 'M': 585, 'B': 567, 'O': 567, 'C': 556, 'D': 552, '«': 540, '»': 538, 'R': 489, '?': 441, 'ź': 414, 'f': 386, 'G': 358, 'L': 316, 'H': 309, 'Ż': 219, 'U': 184, '…': 157, '\\*': 150, '(': 76, ')': 76, 'Ś': 71, 'F': 47, 'é': 43, '-': 33, 'Ł': 24, 'E': 23, '/': 19, 'Ó': 13, '8': 10, '9': 8, '2': 6, 'v': 5, 'Ź': 4, '1': 4, '3': 3, 'x': 3, 'V': 3, '7': 2, '4': 2, '5': 2, 'q': 2, 'æ': 2, 'à': 1, 'Ć': 1, '6': 1, '0': 1})\n",
+ "\n"
+ ]
+ },
+ {
+ "cell_type": "code",
+ "execution_count": 1,
+ "metadata": {},
+ "outputs": [
+ {
+ "name": "stdout",
+ "output_type": "stream",
+ "text": [
+ "OrderedDict([(' ', 63444), ('a', 30979), ('i', 29353), ('e', 25343), ('o', 23050), ('z', 22741), ('n', 15505), ('r', 15328)])"
+ ]
+ }
+ ],
+ "source": [
+ "from collections import Counter\n",
+ "from collections import OrderedDict\n",
+ "\n",
+ "def freq_list(g, top=None):\n",
+ " c = Counter(g)\n",
+ "\n",
+ " if top is None:\n",
+ " items = c.items()\n",
+ " else:\n",
+ " items = c.most_common(top)\n",
+ "\n",
+ " return OrderedDict(sorted(items, key=lambda t: -t[1]))\n",
+ "\n",
+ "freq_list(get_characters(pan_tadeusz), top=8)"
+ ]
+ },
+ {
+ "cell_type": "code",
+ "execution_count": 1,
+ "metadata": {},
+ "outputs": [
+ {
+ "data": {
+ "image/png": "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",
+ "text/plain": [
+ ""
+ ]
+ },
+ "metadata": {},
+ "output_type": "display_data"
+ }
+ ],
+ "source": [
+ "import matplotlib.pyplot as plt\n",
+ "from collections import OrderedDict\n",
+ "\n",
+ "def rang_freq_with_labels(name, g, top=None):\n",
+ " freq = freq_list(g, top)\n",
+ "\n",
+ " plt.figure(figsize=(12, 3))\n",
+ " plt.ylabel('liczba wystąpień')\n",
+ "\n",
+ " plt.bar(freq.keys(), freq.values())\n",
+ "\n",
+ " fname = f'02_Jezyki/{name}.png'\n",
+ "\n",
+ " plt.savefig(fname)\n",
+ "\n",
+ " return fname\n",
+ "\n",
+ "rang_freq_with_labels('pt-chars', get_characters(pan_tadeusz))"
+ ]
+ },
+ {
+ "cell_type": "markdown",
+ "metadata": {},
+ "source": [
+ "#### Słowa\n",
+ "\n"
+ ]
+ },
+ {
+ "cell_type": "markdown",
+ "metadata": {},
+ "source": [
+ "Co rozumiemy pod pojęciem słowa czy wyrazu, nie jest oczywiste. W praktyce zależy to od wyboru **tokenizatora**.\n",
+ "\n",
+ "Załóżmy, że przez wyraz rozumieć będziemy nieprzerwany ciąg liter bądź cyfr (oraz gwiazdek\n",
+ "— to za chwilę ułatwi nam analizę pewnego tekstu…).\n",
+ "\n"
+ ]
+ },
+ {
+ "cell_type": "code",
+ "execution_count": 1,
+ "metadata": {},
+ "outputs": [
+ {
+ "name": "stdout",
+ "output_type": "stream",
+ "text": [
+ "['Ty', 'co', 'gród', 'zamkowy', 'Nowogródzki', 'ochraniasz', 'z', 'jego', 'wiernym', 'ludem', 'Jak', 'mnie', 'dziecko', 'do', 'zdrowia', 'powróciłaś', 'cudem', 'Gdy', 'od', 'płaczącej', 'matki', 'pod', 'Twoją', 'opiekę', 'Ofiarowany', 'martwą', 'podniosłem', 'powiekę', 'I', 'zaraz']"
+ ]
+ }
+ ],
+ "source": [
+ "from itertools import islice\n",
+ "import regex as re\n",
+ "\n",
+ "def get_words(t):\n",
+ " for m in re.finditer(r'[\\p{L}0-9\\*]+', t):\n",
+ " yield m.group(0)\n",
+ "\n",
+ "list(islice(get_words(pan_tadeusz), 100, 130))"
+ ]
+ },
+ {
+ "cell_type": "markdown",
+ "metadata": {},
+ "source": [
+ "Zobaczmy 20 najczęstszych wyrazów.\n",
+ "\n"
+ ]
+ },
+ {
+ "cell_type": "code",
+ "execution_count": 1,
+ "metadata": {},
+ "outputs": [
+ {
+ "data": {
+ "image/png": "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",
+ "text/plain": [
+ ""
+ ]
+ },
+ "metadata": {},
+ "output_type": "display_data"
+ }
+ ],
+ "source": [
+ "rang_freq_with_labels('pt-words-20', get_words(pan_tadeusz), top=20)"
+ ]
+ },
+ {
+ "cell_type": "markdown",
+ "metadata": {},
+ "source": [
+ "Zobaczmy pełny obraz, już bez etykiet.\n",
+ "\n"
+ ]
+ },
+ {
+ "cell_type": "code",
+ "execution_count": 1,
+ "metadata": {},
+ "outputs": [
+ {
+ "data": {
+ "image/png": "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",
+ "text/plain": [
+ ""
+ ]
+ },
+ "metadata": {},
+ "output_type": "display_data"
+ }
+ ],
+ "source": [
+ "import matplotlib.pyplot as plt\n",
+ "from math import log\n",
+ "\n",
+ "def rang_freq(name, g):\n",
+ " freq = freq_list(g)\n",
+ "\n",
+ " plt.figure().clear()\n",
+ " plt.plot(range(1, len(freq.values())+1), freq.values())\n",
+ "\n",
+ " fname = f'02_Jezyki/{name}.png'\n",
+ "\n",
+ " plt.savefig(fname)\n",
+ "\n",
+ " return fname\n",
+ "\n",
+ "rang_freq('pt-words', get_words(pan_tadeusz))"
+ ]
+ },
+ {
+ "cell_type": "markdown",
+ "metadata": {},
+ "source": [
+ "Widać, jak różne skale obejmuje ten wykres. Zastosujemy logarytm,\n",
+ "najpierw tylko do współrzędnej $y$.\n",
+ "\n"
+ ]
+ },
+ {
+ "cell_type": "code",
+ "execution_count": 1,
+ "metadata": {},
+ "outputs": [
+ {
+ "data": {
+ "image/png": "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",
+ "text/plain": [
+ ""
+ ]
+ },
+ "metadata": {},
+ "output_type": "display_data"
+ }
+ ],
+ "source": [
+ "import matplotlib.pyplot as plt\n",
+ "from math import log\n",
+ "\n",
+ "def rang_log_freq(name, g):\n",
+ " freq = freq_list(g)\n",
+ "\n",
+ " plt.figure().clear()\n",
+ " plt.plot(range(1, len(freq.values())+1), [log(y) for y in freq.values()])\n",
+ "\n",
+ " fname = f'02_Jezyki/{name}.png'\n",
+ "\n",
+ " plt.savefig(fname)\n",
+ "\n",
+ " return fname\n",
+ "\n",
+ "rang_log_freq('pt-words-log', get_words(pan_tadeusz))"
+ ]
+ },
+ {
+ "cell_type": "markdown",
+ "metadata": {},
+ "source": [
+ "****Pytanie**** Dlaczego widzimy coraz dłuższe „schodki”?\n",
+ "\n"
+ ]
+ },
+ {
+ "cell_type": "markdown",
+ "metadata": {},
+ "source": [
+ "#### Hapax legomena\n",
+ "\n"
+ ]
+ },
+ {
+ "cell_type": "markdown",
+ "metadata": {},
+ "source": [
+ "Z poprzedniego wykresu możemy odczytać, że ok. 2/3 wyrazów wystąpiło\n",
+ "dokładnie 1 raz. Słowa występujące jeden raz w danym korpusie noszą\n",
+ "nazwę *hapax legomena* (w liczbie pojedynczej *hapax legomenon*, ἅπαξ\n",
+ "λεγόμενον, „raz powiedziane”, żargonowo: „hapaks”).\n",
+ "\n",
+ "„Prawdziwe” hapax legomena, słowa, które wystąpiły tylko raz w *całym*\n",
+ "korpusie tekstów danego języka (np. starożytnego) rzecz jasna\n",
+ "sprawiają olbrzymie trudności w tłumaczeniu. Przykładem jest greckie\n",
+ "słowo ἐπιούσιος, przydawka odnosząca się do chleba w modlitwie „Ojcze\n",
+ "nasz”. Jest to jedyne poświadczenie tego słowa w całym znanym korpusie\n",
+ "greki (nie tylko z Pisma Świętego). W języku polskim tłumaczymy je na\n",
+ "„powszedni”, ale na przykład w rosyjskim przyjął się odpowiednik\n",
+ "„насущный” — o przeciwstawnym do polskiego znaczeniu!\n",
+ "\n",
+ "W sumie podobne problemy hapaksy mogą sprawiać metodom statystycznym\n",
+ "przy przetwarzaniu jakiekolwiek korpusu.\n",
+ "\n"
+ ]
+ },
+ {
+ "cell_type": "markdown",
+ "metadata": {},
+ "source": [
+ "#### Wykres log-log\n",
+ "\n"
+ ]
+ },
+ {
+ "cell_type": "markdown",
+ "metadata": {},
+ "source": [
+ "Jeśli wspomniany wcześniej wykres narysujemy używając skali\n",
+ "logarytmicznej dla ****obu**** osi, otrzymamy kształt zbliżony do linii prostej.\n",
+ "\n",
+ "Tę własność tekstów nazywamy ****prawem Zipfa****.\n",
+ "\n"
+ ]
+ },
+ {
+ "cell_type": "code",
+ "execution_count": 1,
+ "metadata": {},
+ "outputs": [
+ {
+ "data": {
+ "image/png": "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",
+ "text/plain": [
+ ""
+ ]
+ },
+ "metadata": {},
+ "output_type": "display_data"
+ }
+ ],
+ "source": [
+ "import matplotlib.pyplot as plt\n",
+ "from math import log\n",
+ "\n",
+ "def log_rang_log_freq(name, g):\n",
+ " freq = freq_list(g)\n",
+ "\n",
+ " plt.figure().clear()\n",
+ " plt.plot([log(x) for x in range(1, len(freq.values())+1)], [log(y) for y in freq.values()])\n",
+ "\n",
+ " fname = f'02_Jezyki/{name}.png'\n",
+ "\n",
+ " plt.savefig(fname)\n",
+ "\n",
+ " return fname\n",
+ "\n",
+ "log_rang_log_freq('pt-words-log-log', get_words(pan_tadeusz))"
+ ]
+ },
+ {
+ "cell_type": "markdown",
+ "metadata": {},
+ "source": [
+ "#### Związek między frekwencją a długością\n",
+ "\n"
+ ]
+ },
+ {
+ "cell_type": "markdown",
+ "metadata": {},
+ "source": [
+ "Powiązane z prawem Zipfa prawo językowe opisuje zależność między\n",
+ "częstością użycia słowa a jego długością. Generalnie im krótsze słowo, tym częstsze.\n",
+ "\n"
+ ]
+ },
+ {
+ "cell_type": "code",
+ "execution_count": 1,
+ "metadata": {},
+ "outputs": [
+ {
+ "data": {
+ "image/png": "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",
+ "text/plain": [
+ ""
+ ]
+ },
+ "metadata": {},
+ "output_type": "display_data"
+ }
+ ],
+ "source": [
+ "def freq_vs_length(name, g, top=None):\n",
+ " freq = freq_list(g)\n",
+ "\n",
+ " plt.figure().clear()\n",
+ " plt.scatter([len(x) for x in freq.keys()], [log(y) for y in freq.values()],\n",
+ " facecolors='none', edgecolors='r')\n",
+ "\n",
+ " fname = f'02_Jezyki/{name}.png'\n",
+ "\n",
+ " plt.savefig(fname)\n",
+ "\n",
+ " return fname\n",
+ "\n",
+ "freq_vs_length('pt-lengths', get_words(pan_tadeusz))"
+ ]
+ }
+ ],
+ "metadata": {
+ "kernelspec": {
+ "display_name": "Python 3 (ipykernel)",
+ "language": "python",
+ "name": "python3"
+ },
+ "language_info": {
+ "codemirror_mode": {
+ "name": "ipython",
+ "version": 3
+ },
+ "file_extension": ".py",
+ "mimetype": "text/x-python",
+ "name": "python",
+ "nbconvert_exporter": "python",
+ "pygments_lexer": "ipython3",
+ "version": "3.10.9"
+ },
+ "org": null
+ },
+ "nbformat": 4,
+ "nbformat_minor": 1
+}
diff --git a/wyk/02_Jezyki.org b/wyk/02_Jezyki.org
new file mode 100644
index 0000000..2d278ab
--- /dev/null
+++ b/wyk/02_Jezyki.org
@@ -0,0 +1,270 @@
+* Języki i ich prawa statystyczne
+
+Jakim rozkładom statystycznym podlegają języki?
+
+** Język naturalny albo „Pan Tadeusz” w liczbach
+
+Przygotujmy najpierw „infrastrukturę” do /segmentacji/ tekstu na różnego rodzaju jednostki.
+Używać będziemy generatorów.
+
+*Pytanie* Dlaczego generatory zamiast list?
+
+#+BEGIN_SRC ipython :session mysession :exports both :results raw drawer
+ import requests
+
+ url = 'https://wolnelektury.pl/media/book/txt/pan-tadeusz.txt'
+ pan_tadeusz = requests.get(url).content.decode('utf-8')
+
+ pan_tadeusz[100:150]
+#+END_SRC
+
+#+RESULTS:
+:results:
+Księga pierwsza
+
+
+
+Gospodarstwo
+
+Powrót pani
+:end:
+
+*** Znaki
+
+#+BEGIN_SRC ipython :session mysession :exports both :results raw drawer
+ from itertools import islice
+
+ def get_characters(t):
+ yield from t
+
+ list(islice(get_characters(pan_tadeusz), 100, 150))
+#+END_SRC
+
+#+RESULTS:
+:results:
+['K', 's', 'i', 'ę', 'g', 'a', ' ', 'p', 'i', 'e', 'r', 'w', 's', 'z', 'a', '\r', '\n', '\r', '\n', '\r', '\n', '\r', '\n', 'G', 'o', 's', 'p', 'o', 'd', 'a', 'r', 's', 't', 'w', 'o', '\r', '\n', '\r', '\n', 'P', 'o', 'w', 'r', 'ó', 't', ' ', 'p', 'a', 'n', 'i']
+:end:
+
+#+BEGIN_SRC ipython :session mysession :exports both :results raw drawer
+ from collections import Counter
+
+ c = Counter(get_characters(pan_tadeusz))
+
+ c
+#+END_SRC
+
+#+RESULTS:
+:results:
+Counter({' ': 63444, 'a': 30979, 'i': 29353, 'e': 25343, 'o': 23050, 'z': 22741, 'n': 15505, 'r': 15328, 's': 15255, 'w': 14625, 'c': 14153, 'y': 13732, 'k': 12362, 'd': 11465, '\r': 10851, '\n': 10851, 't': 10757, 'm': 10269, 'ł': 10059, ',': 9130, 'p': 8031, 'u': 7699, 'l': 6677, 'j': 6586, 'b': 5753, 'ę': 5534, 'ą': 4794, 'g': 4775, 'h': 3915, 'ż': 3334, 'ó': 3097, 'ś': 2524, '.': 2380, 'ć': 1956, ';': 1445, 'P': 1265, 'W': 1258, ':': 1152, '!': 1083, 'S': 1045, 'T': 971, 'I': 795, 'N': 793, 'Z': 785, 'J': 729, '—': 720, 'A': 698, 'K': 683, 'ń': 651, 'M': 585, 'B': 567, 'O': 567, 'C': 556, 'D': 552, '«': 540, '»': 538, 'R': 489, '?': 441, 'ź': 414, 'f': 386, 'G': 358, 'L': 316, 'H': 309, 'Ż': 219, 'U': 184, '…': 157, '*': 150, '(': 76, ')': 76, 'Ś': 71, 'F': 47, 'é': 43, '-': 33, 'Ł': 24, 'E': 23, '/': 19, 'Ó': 13, '8': 10, '9': 8, '2': 6, 'v': 5, 'Ź': 4, '1': 4, '3': 3, 'x': 3, 'V': 3, '7': 2, '4': 2, '5': 2, 'q': 2, 'æ': 2, 'à': 1, 'Ć': 1, '6': 1, '0': 1})
+:end:
+
+Napiszmy pomocniczą funkcję, która zwraca *listę frekwencyjną*.
+
+#+RESULTS:
+:results:
+Counter({' ': 63444, 'a': 30979, 'i': 29353, 'e': 25343, 'o': 23050, 'z': 22741, 'n': 15505, 'r': 15328, 's': 15255, 'w': 14625, 'c': 14153, 'y': 13732, 'k': 12362, 'd': 11465, '\r': 10851, '\n': 10851, 't': 10757, 'm': 10269, 'ł': 10059, ',': 9130, 'p': 8031, 'u': 7699, 'l': 6677, 'j': 6586, 'b': 5753, 'ę': 5534, 'ą': 4794, 'g': 4775, 'h': 3915, 'ż': 3334, 'ó': 3097, 'ś': 2524, '.': 2380, 'ć': 1956, ';': 1445, 'P': 1265, 'W': 1258, ':': 1152, '!': 1083, 'S': 1045, 'T': 971, 'I': 795, 'N': 793, 'Z': 785, 'J': 729, '—': 720, 'A': 698, 'K': 683, 'ń': 651, 'M': 585, 'B': 567, 'O': 567, 'C': 556, 'D': 552, '«': 540, '»': 538, 'R': 489, '?': 441, 'ź': 414, 'f': 386, 'G': 358, 'L': 316, 'H': 309, 'Ż': 219, 'U': 184, '…': 157, '*': 150, '(': 76, ')': 76, 'Ś': 71, 'F': 47, 'é': 43, '-': 33, 'Ł': 24, 'E': 23, '/': 19, 'Ó': 13, '8': 10, '9': 8, '2': 6, 'v': 5, 'Ź': 4, '1': 4, '3': 3, 'x': 3, 'V': 3, '7': 2, '4': 2, '5': 2, 'q': 2, 'æ': 2, 'à': 1, 'Ć': 1, '6': 1, '0': 1})
+:end:
+
+#+BEGIN_SRC ipython :session mysession :exports both :results raw drawer
+ from collections import Counter
+ from collections import OrderedDict
+
+ def freq_list(g, top=None):
+ c = Counter(g)
+
+ if top is None:
+ items = c.items()
+ else:
+ items = c.most_common(top)
+
+ return OrderedDict(sorted(items, key=lambda t: -t[1]))
+
+ freq_list(get_characters(pan_tadeusz), top=8)
+#+END_SRC
+
+#+RESULTS:
+:results:
+OrderedDict([(' ', 63444), ('a', 30979), ('i', 29353), ('e', 25343), ('o', 23050), ('z', 22741), ('n', 15505), ('r', 15328)])
+:end:
+
+
+#+BEGIN_SRC ipython :session mysession :results file
+ import matplotlib.pyplot as plt
+ from collections import OrderedDict
+
+ def rang_freq_with_labels(name, g, top=None):
+ freq = freq_list(g, top)
+
+ plt.figure(figsize=(12, 3))
+ plt.ylabel('liczba wystąpień')
+
+ plt.bar(freq.keys(), freq.values())
+
+ fname = f'02_Jezyki/{name}.png'
+
+ plt.savefig(fname)
+
+ return fname
+
+ rang_freq_with_labels('pt-chars', get_characters(pan_tadeusz))
+#+END_SRC
+
+#+RESULTS:
+[[file:02_Jezyki/pt-chars.png]]
+
+*** Słowa
+
+Co rozumiemy pod pojęciem słowa czy wyrazu, nie jest oczywiste. W praktyce zależy to od wyboru *tokenizatora*.
+
+Załóżmy, że przez wyraz rozumieć będziemy nieprzerwany ciąg liter bądź cyfr (oraz gwiazdek
+— to za chwilę ułatwi nam analizę pewnego tekstu…).
+
+#+BEGIN_SRC ipython :session mysession :exports both :results raw drawer
+ from itertools import islice
+ import regex as re
+
+ def get_words(t):
+ for m in re.finditer(r'[\p{L}0-9\*]+', t):
+ yield m.group(0)
+
+ list(islice(get_words(pan_tadeusz), 100, 130))
+#+END_SRC
+
+#+RESULTS:
+:results:
+['Ty', 'co', 'gród', 'zamkowy', 'Nowogródzki', 'ochraniasz', 'z', 'jego', 'wiernym', 'ludem', 'Jak', 'mnie', 'dziecko', 'do', 'zdrowia', 'powróciłaś', 'cudem', 'Gdy', 'od', 'płaczącej', 'matki', 'pod', 'Twoją', 'opiekę', 'Ofiarowany', 'martwą', 'podniosłem', 'powiekę', 'I', 'zaraz']
+:end:
+
+Zobaczmy 20 najczęstszych wyrazów.
+
+
+#+BEGIN_SRC ipython :session mysession :results file
+ rang_freq_with_labels('pt-words-20', get_words(pan_tadeusz), top=20)
+#+END_SRC
+
+#+RESULTS:
+[[file:02_Jezyki/pt-words-20.png]]
+
+Zobaczmy pełny obraz, już bez etykiet.
+
+#+BEGIN_SRC ipython :session mysession :results file
+ import matplotlib.pyplot as plt
+ from math import log
+
+ def rang_freq(name, g):
+ freq = freq_list(g)
+
+ plt.figure().clear()
+ plt.plot(range(1, len(freq.values())+1), freq.values())
+
+ fname = f'02_Jezyki/{name}.png'
+
+ plt.savefig(fname)
+
+ return fname
+
+ rang_freq('pt-words', get_words(pan_tadeusz))
+#+END_SRC
+
+#+RESULTS:
+[[file:02_Jezyki/pt-words.png]]
+
+Widać, jak różne skale obejmuje ten wykres. Zastosujemy logarytm,
+najpierw tylko do współrzędnej $y$.
+
+#+BEGIN_SRC ipython :session mysession :results file
+ import matplotlib.pyplot as plt
+ from math import log
+
+ def rang_log_freq(name, g):
+ freq = freq_list(g)
+
+ plt.figure().clear()
+ plt.plot(range(1, len(freq.values())+1), [log(y) for y in freq.values()])
+
+ fname = f'02_Jezyki/{name}.png'
+
+ plt.savefig(fname)
+
+ return fname
+
+ rang_log_freq('pt-words-log', get_words(pan_tadeusz))
+#+END_SRC
+
+#+RESULTS:
+[[file:02_Jezyki/pt-words-log.png]]
+
+**Pytanie** Dlaczego widzimy coraz dłuższe „schodki”?
+
+*** Hapax legomena
+
+Z poprzedniego wykresu możemy odczytać, że ok. 2/3 wyrazów wystąpiło
+dokładnie 1 raz. Słowa występujące jeden raz w danym korpusie noszą
+nazwę /hapax legomena/ (w liczbie pojedynczej /hapax legomenon/, ἅπαξ
+λεγόμενον, „raz powiedziane”, żargonowo: „hapaks”).
+
+„Prawdziwe” hapax legomena, słowa, które wystąpiły tylko raz w /całym/
+korpusie tekstów danego języka (np. starożytnego) rzecz jasna
+sprawiają olbrzymie trudności w tłumaczeniu. Przykładem jest greckie
+słowo ἐπιούσιος, przydawka odnosząca się do chleba w modlitwie „Ojcze
+nasz”. Jest to jedyne poświadczenie tego słowa w całym znanym korpusie
+greki (nie tylko z Pisma Świętego). W języku polskim tłumaczymy je na
+„powszedni”, ale na przykład w rosyjskim przyjął się odpowiednik
+„насущный” — o przeciwstawnym do polskiego znaczeniu!
+
+W sumie podobne problemy hapaksy mogą sprawiać metodom statystycznym
+przy przetwarzaniu jakiekolwiek korpusu.
+
+*** Wykres log-log
+
+Jeśli wspomniany wcześniej wykres narysujemy używając skali
+logarytmicznej dla **obu** osi, otrzymamy kształt zbliżony do linii prostej.
+
+Tę własność tekstów nazywamy **prawem Zipfa**.
+
+#+BEGIN_SRC ipython :session mysession :results file
+ import matplotlib.pyplot as plt
+ from math import log
+
+ def log_rang_log_freq(name, g):
+ freq = freq_list(g)
+
+ plt.figure().clear()
+ plt.plot([log(x) for x in range(1, len(freq.values())+1)], [log(y) for y in freq.values()])
+
+ fname = f'02_Jezyki/{name}.png'
+
+ plt.savefig(fname)
+
+ return fname
+
+ log_rang_log_freq('pt-words-log-log', get_words(pan_tadeusz))
+#+END_SRC
+
+#+RESULTS:
+[[file:02_Jezyki/pt-words-log-log.png]]
+
+*** Związek między frekwencją a długością
+
+Powiązane z prawem Zipfa prawo językowe opisuje zależność między
+częstością użycia słowa a jego długością. Generalnie im krótsze słowo, tym częstsze.
+
+#+BEGIN_SRC ipython :session mysession :results file
+ def freq_vs_length(name, g, top=None):
+ freq = freq_list(g)
+
+ plt.figure().clear()
+ plt.scatter([len(x) for x in freq.keys()], [log(y) for y in freq.values()],
+ facecolors='none', edgecolors='r')
+
+ fname = f'02_Jezyki/{name}.png'
+
+ plt.savefig(fname)
+
+ return fname
+
+ freq_vs_length('pt-lengths', get_words(pan_tadeusz))
+#+END_SRC
+
+#+RESULTS:
+[[file:02_Jezyki/pt-lengths.png]]
diff --git a/wyk/02_Jezyki/pt-chars.png b/wyk/02_Jezyki/pt-chars.png
new file mode 100644
index 0000000..71cd92b
Binary files /dev/null and b/wyk/02_Jezyki/pt-chars.png differ
diff --git a/wyk/02_Jezyki/pt-lengths.png b/wyk/02_Jezyki/pt-lengths.png
new file mode 100644
index 0000000..b786bcb
Binary files /dev/null and b/wyk/02_Jezyki/pt-lengths.png differ
diff --git a/wyk/02_Jezyki/pt-words-20.png b/wyk/02_Jezyki/pt-words-20.png
new file mode 100644
index 0000000..b295bc1
Binary files /dev/null and b/wyk/02_Jezyki/pt-words-20.png differ
diff --git a/wyk/02_Jezyki/pt-words-log-log.png b/wyk/02_Jezyki/pt-words-log-log.png
new file mode 100644
index 0000000..b91fb94
Binary files /dev/null and b/wyk/02_Jezyki/pt-words-log-log.png differ
diff --git a/wyk/02_Jezyki/pt-words-log.png b/wyk/02_Jezyki/pt-words-log.png
new file mode 100644
index 0000000..131459d
Binary files /dev/null and b/wyk/02_Jezyki/pt-words-log.png differ
diff --git a/wyk/02_Jezyki/pt-words.png b/wyk/02_Jezyki/pt-words.png
new file mode 100644
index 0000000..e1d0f04
Binary files /dev/null and b/wyk/02_Jezyki/pt-words.png differ
diff --git a/wyk/helpers/add-intro.sh b/wyk/helpers/add-intro.sh
new file mode 100755
index 0000000..cb12e8a
--- /dev/null
+++ b/wyk/helpers/add-intro.sh
@@ -0,0 +1,5 @@
+#!/bin/bash
+
+./helpers/fix-intro.pl $1
+
+perl -pne 's/^{"cells":\[//' $1 | cat helpers/intro - | sponge $1
diff --git a/wyk/helpers/fix-intro.pl b/wyk/helpers/fix-intro.pl
new file mode 100755
index 0000000..8493e5f
--- /dev/null
+++ b/wyk/helpers/fix-intro.pl
@@ -0,0 +1,61 @@
+#!/usr/bin/perl
+
+use strict;
+
+binmode(STDERR, ':utf8');
+
+my $intro_file = 'helpers/intro';
+my $intro_template = 'helpers/intro-template';
+
+my $notebook_file = $ARGV[0];
+
+my $number;
+
+if (($number) = ($notebook_file =~ m{^(\d+)_})) {
+
+} else {
+ die "????";
+}
+
+my $org_file = $notebook_file;
+$org_file =~ s{\.ipynb}{\.org};
+
+my $title = undef;
+
+open(my $org_fh, '<', $org_file);
+binmode($org_fh, ':utf8');
+
+LOOP:
+while (my $line=<$org_fh>) {
+ chomp $line;
+
+ if ($line =~ m{^\* (.*)}) {
+ $title = $1;
+ last LOOP;
+ } elsif ($line =~ /\S/) {
+ last LOOP;
+ }
+}
+
+if (not defined($title)) {
+ die '???? TITLE';
+}
+
+print STDERR "NUMBER: $number\n";
+print STDERR "TITLE: $title\n";
+
+my $intro = '';
+
+open(my $intro_ih, '<', $intro_template);
+binmode($intro_ih, ':utf8');
+while(my $line=<$intro_ih>) {
+ $line =~ s{NUMBER}{$number};
+ $line =~ s{TITLE}{$title};
+ $intro .= $line;
+}
+close($intro_ih);
+
+open(my $intro_oh, '>', $intro_file);
+binmode($intro_oh, ':utf8');
+print $intro_oh $intro;
+close($intro_oh);
diff --git a/wyk/helpers/intro b/wyk/helpers/intro
new file mode 100644
index 0000000..7b96dcd
--- /dev/null
+++ b/wyk/helpers/intro
@@ -0,0 +1,18 @@
+
+{
+ "cells": [
+ {
+ "cell_type": "markdown",
+ "metadata": {},
+ "source": [
+ "![Logo 1](https://git.wmi.amu.edu.pl/AITech/Szablon/raw/branch/master/Logotyp_AITech1.jpg)\n",
+ "\n",
+ "
Modelowanie języka \n",
+ " 02. Języki i ich prawa statystyczne [wykład] \n",
+ " Filip Graliński (2022) \n",
+ "\n",
+ "\n",
+ "![Logo 2](https://git.wmi.amu.edu.pl/AITech/Szablon/raw/branch/master/Logotyp_AITech2.jpg)\n",
+ "\n"
+ ]
+ },
diff --git a/wyk/helpers/intro-template b/wyk/helpers/intro-template
new file mode 100644
index 0000000..9b9cafa
--- /dev/null
+++ b/wyk/helpers/intro-template
@@ -0,0 +1,18 @@
+
+{
+ "cells": [
+ {
+ "cell_type": "markdown",
+ "metadata": {},
+ "source": [
+ "![Logo 1](https://git.wmi.amu.edu.pl/AITech/Szablon/raw/branch/master/Logotyp_AITech1.jpg)\n",
+ "\n",
+ "
Modelowanie języka \n",
+ " NUMBER. TITLE [wykład] \n",
+ " Filip Graliński (2022) \n",
+ "\n",
+ "\n",
+ "![Logo 2](https://git.wmi.amu.edu.pl/AITech/Szablon/raw/branch/master/Logotyp_AITech2.jpg)\n",
+ "\n"
+ ]
+ },