This commit is contained in:
Kamil Kubiak 2024-04-20 16:54:31 +02:00
parent 1f7c4d72ad
commit 9a3d25dac1
2 changed files with 60 additions and 10 deletions

View File

@ -223,17 +223,17 @@
},
{
"cell_type": "code",
"execution_count": 5,
"execution_count": 7,
"id": "secondary-wrist",
"metadata": {},
"outputs": [
{
"data": {
"text/plain": [
"2"
"3"
]
},
"execution_count": 5,
"execution_count": 8,
"metadata": {},
"output_type": "execute_result"
}
@ -344,21 +344,71 @@
},
{
"cell_type": "code",
"execution_count": 10,
"execution_count": 18,
"id": "genetic-cradle",
"metadata": {},
"outputs": [],
"source": [
"def fuzzy_lookup(sentence, threshold):\n",
" return []"
"outputs": [
{
"name": "stdout",
"output_type": "stream",
"text": [
"Wciśnij przycisk Enter -> Wciśnij przycisk Enter [ SCORE = 1.0 ]\n",
"Wciśnij przycisk Enter -> Wciśnij przycisk ENTER [ SCORE = 0.8181818181818181 ]\n",
"Wciśnij przycisk Enter -> Wciśnij przycisk Enter! [ SCORE = 0.9565217391304348 ]\n",
"Wciśnij przycisk Enter -> Wciśnij przycisk Enter [ SCORE = 1.0 ]\n"
]
}
],
"source": [
"#!pip3 install python-Levenshtein\n",
"from Levenshtein import distance as levenshtein_distance\n",
"\n",
"translation_memory = [\n",
" ('Wciśnij przycisk Enter', 'Press the ENTER button'),\n",
" ('Wciśnij przycisk ENTER', 'Press the ENTER button'), \n",
" ('Wciśnij przycisk Enter!', 'Press the ENTER button!'), \n",
" ('Wciśnij przycisk', 'Press the button'), \n",
" ('Wciśnij Enter', 'Press the ENTER'), \n",
" ('Sprawdź ustawienia sieciowe', 'Check the network settings'),\n",
" ('Drukarka jest wyłączona', 'The printer is switched off'),\n",
" ('Wymagane ponowne uruchomienie komputera', 'System restart required')\n",
" ]\n",
"\n",
"def levenshtein_similarity(x,y):\n",
" return 1 - levenshtein_distance(x,y) / max(len(x), len(y))\n",
"\n",
"def fuzzy_lookup(sentence, threshold):\n",
" for entry in translation_memory:\n",
" part1 = entry[0]\n",
" score = levenshtein_similarity(sentence, part1)\n",
" if score >= threshold:\n",
" print(sentence + ' -> ' + part1 + ' [ SCORE = ' + str(score) + ' ]')\n",
"\n",
"fuzzy_lookup('Wciśnij przycisk Enter', 0.8)"
]
},
{
"cell_type": "code",
"execution_count": null,
"id": "8b7bb6aa-5aaf-4f49-84ab-edbe6797d568",
"metadata": {},
"outputs": [],
"source": []
},
{
"cell_type": "code",
"execution_count": null,
"id": "3b095fff-71a8-44a9-b809-d872ee9a7b62",
"metadata": {},
"outputs": [],
"source": []
}
],
"metadata": {
"author": "Rafał Jaworski",
"email": "rjawor@amu.edu.pl",
"kernelspec": {
"display_name": "Python 3",
"display_name": "Python 3 (ipykernel)",
"language": "python",
"name": "python3"
},
@ -373,7 +423,7 @@
"name": "python",
"nbconvert_exporter": "python",
"pygments_lexer": "ipython3",
"version": "3.8.10"
"version": "3.9.2"
},
"subtitle": "2. Zaawansowane użycie pamięci tłumaczeń",
"title": "Komputerowe wspomaganie tłumaczenia",

View File

@ -407,7 +407,7 @@
"author": "Rafał Jaworski",
"email": "rjawor@amu.edu.pl",
"kernelspec": {
"display_name": "Python 3",
"display_name": "Python 3 (ipykernel)",
"language": "python",
"name": "python3"
},
@ -422,7 +422,7 @@
"name": "python",
"nbconvert_exporter": "python",
"pygments_lexer": "ipython3",
"version": "3.8.10"
"version": "3.9.2"
},
"subtitle": "3. Terminologia",
"title": "Komputerowe wspomaganie tłumaczenia",