You can not select more than 25 topics
Topics must start with a letter or number, can include dashes ('-') and can be up to 35 characters long.
227 lines
7.3 KiB
227 lines
7.3 KiB
{
|
|
"cells": [
|
|
{
|
|
"cell_type": "code",
|
|
"execution_count": 1,
|
|
"metadata": {},
|
|
"outputs": [],
|
|
"source": [
|
|
"from FremdWB import *\n",
|
|
"\n",
|
|
"\n",
|
|
"\n",
|
|
"fwb = FremdWB(None,None)"
|
|
]
|
|
},
|
|
{
|
|
"cell_type": "code",
|
|
"execution_count": 2,
|
|
"metadata": {},
|
|
"outputs": [
|
|
{
|
|
"name": "stdout",
|
|
"output_type": "stream",
|
|
"text": [
|
|
"1000\n",
|
|
"2000\n",
|
|
"creating the hkl dump of Fremd_WBDBAll\n",
|
|
"done..\n",
|
|
"Creating the hkl dump of Fremd_WBDB 1\n",
|
|
"Creating the hkl dump of Fremd_WBDB 2\n"
|
|
]
|
|
},
|
|
{
|
|
"data": {
|
|
"text/plain": [
|
|
"'done'"
|
|
]
|
|
},
|
|
"execution_count": 2,
|
|
"metadata": {},
|
|
"output_type": "execute_result"
|
|
}
|
|
],
|
|
"source": [
|
|
"fwb.create_hklDB_from_csv('HurrakiWoerterbuch_nodoubles.txt', 'None')"
|
|
]
|
|
},
|
|
{
|
|
"cell_type": "code",
|
|
"execution_count": 3,
|
|
"metadata": {},
|
|
"outputs": [
|
|
{
|
|
"name": "stdout",
|
|
"output_type": "stream",
|
|
"text": [
|
|
"Creating the bag of words...\n",
|
|
"\n",
|
|
"dumping the data to hkl format..\n",
|
|
"done\n",
|
|
"Creating the bag of words...\n",
|
|
"\n",
|
|
"dumping the data to hkl format..\n",
|
|
"done\n"
|
|
]
|
|
}
|
|
],
|
|
"source": [
|
|
"fwb.load_DB_into_FASTsearch()"
|
|
]
|
|
},
|
|
{
|
|
"cell_type": "code",
|
|
"execution_count": 4,
|
|
"metadata": {},
|
|
"outputs": [],
|
|
"source": [
|
|
"sentences = [['das', 'ist', 'Abfall'],['er', 'ging', 'über', 'die', 'Straße'], ['halt', 'ab', 'hier']]\n",
|
|
"punctuations = ['.', '!', '.']"
|
|
]
|
|
},
|
|
{
|
|
"cell_type": "code",
|
|
"execution_count": 5,
|
|
"metadata": {},
|
|
"outputs": [],
|
|
"source": [
|
|
"outsentences, punctuations = fwb.fremdEintragAppend(sentences, punctuations)"
|
|
]
|
|
},
|
|
{
|
|
"cell_type": "code",
|
|
"execution_count": 6,
|
|
"metadata": {},
|
|
"outputs": [
|
|
{
|
|
"name": "stdout",
|
|
"output_type": "stream",
|
|
"text": [
|
|
"[['das', 'ist', 'Abfall'], ['Abfall', 'sind', 'Sachen', 'die', 'wir', 'nicht', 'mehr', 'brauchen'], ['er', 'ging', 'über', 'die', 'Straße'], ['halt', 'ab', 'hier']] ['.', '.', '!', '.']\n"
|
|
]
|
|
}
|
|
],
|
|
"source": [
|
|
"print(outsentences, punctuations)"
|
|
]
|
|
},
|
|
{
|
|
"cell_type": "code",
|
|
"execution_count": 1,
|
|
"metadata": {},
|
|
"outputs": [
|
|
{
|
|
"name": "stdout",
|
|
"output_type": "stream",
|
|
"text": [
|
|
"oi\n"
|
|
]
|
|
}
|
|
],
|
|
"source": [
|
|
"print('oi')\n"
|
|
]
|
|
},
|
|
{
|
|
"cell_type": "code",
|
|
"execution_count": 2,
|
|
"metadata": {},
|
|
"outputs": [
|
|
{
|
|
"name": "stdout",
|
|
"output_type": "stream",
|
|
"text": [
|
|
"Aktiv.txt hkldb_AllAktiv_old.hkl\r\n",
|
|
"atest1 hkldb_AllVorgangspassiv.hkl\r\n",
|
|
"atest1out hkldb_AllVorgangspassiv_old.hkl\r\n",
|
|
"Babelfish.jpg hkldb_AllZustandspassiv.hkl\r\n",
|
|
"Babelfish.png hkldb_AllZustandspassiv_old.hkl\r\n",
|
|
"bagofwordshkldb1Aktiv.pkl hkldbFremd_WB1.hkl\r\n",
|
|
"bagofwordshkldb1Vorgangspassiv.pkl hkldbFremd_WB2.hkl\r\n",
|
|
"bagofwordshkldb1Zustandspassiv.pkl hkldbFremd_WB_All.hkl\r\n",
|
|
"bagofwordshkldb2Aktiv.pkl hkldbIndi_Conju1.hkl\r\n",
|
|
"bagofwordshkldb2Vorgangspassiv.pkl hkldbIndi_Conju2.hkl\r\n",
|
|
"bagofwordshkldb2Zustandspassiv.pkl hkldbIndi_Conju_All.hkl\r\n",
|
|
"bagofwordshkldbFremd_WB1.pkl hkldbMedio1.hkl\r\n",
|
|
"bagofwordshkldbFremd_WB2.pkl hkldbMedio2.hkl\r\n",
|
|
"bagofwordshkldbIndi_Conju1.pkl hkldbMedio_All.hkl\r\n",
|
|
"bagofwordshkldbIndi_Conju2.pkl hkldbOpposites1.hkl\r\n",
|
|
"bagofwordshkldbMedio1.pkl hkldbOpposites2.hkl\r\n",
|
|
"bagofwordshkldbMedio2.pkl hkldbOpposites_All.hkl\r\n",
|
|
"bagofwordshkldbOpposites1.pkl hkldbShorts_All.hkl\r\n",
|
|
"bagofwordshkldbOpposites2.pkl hkldbShorts.hkl\r\n",
|
|
"bagofwordshkldbShorts.pkl \u001b[0m\u001b[01;34mhtml\u001b[0m/\r\n",
|
|
"bagofwords.pkl HurrakiWoerterbuch_nodoubles.txt\r\n",
|
|
"bagofwordstwolabel.pkl Indikativ_Conjunktiv.txt\r\n",
|
|
"brainBasaBuuka5.png Medio.py\r\n",
|
|
"ConjunctSolve.py mediowords.txt\r\n",
|
|
"DataBaseOneZeroshkldb1Aktiv.hkl oi.py\r\n",
|
|
"DataBaseOneZeroshkldb1Vorgangspassiv.hkl Passiv2Aktiv.py\r\n",
|
|
"DataBaseOneZeroshkldb1Zustandspassiv.hkl Prototype.ipynb\r\n",
|
|
"DataBaseOneZeroshkldb2Aktiv.hkl \u001b[01;34m__pycache__\u001b[0m/\r\n",
|
|
"DataBaseOneZeroshkldb2Vorgangspassiv.hkl requirements.txt\r\n",
|
|
"DataBaseOneZeroshkldb2Zustandspassiv.hkl SayYes.py\r\n",
|
|
"DataBaseOneZeroshkldbFremd_WB1.hkl SentGlue.py\r\n",
|
|
"DataBaseOneZeroshkldbFremd_WB2.hkl SentSeg.py\r\n",
|
|
"DataBaseOneZeroshkldbIndi_Conju1.hkl SolveShorts.py\r\n",
|
|
"DataBaseOneZeroshkldbIndi_Conju2.hkl svg.py\r\n",
|
|
"DataBaseOneZeroshkldbMedio1.hkl test1out\r\n",
|
|
"DataBaseOneZeroshkldbMedio2.hkl trainedSGD.pkl\r\n",
|
|
"DataBaseOneZeroshkldbOpposites1.hkl trainedSGD_twolabel.pkl\r\n",
|
|
"DataBaseOneZeroshkldbOpposites2.hkl untitled.txt\r\n",
|
|
"DataBaseOneZeroshkldbShorts.hkl UseConjunctiveSolve.ipynb\r\n",
|
|
"FASTsearch.py Use_ConjunctSolve.py\r\n",
|
|
"FremdWB.py Use_Fremd_WB.ipynb\r\n",
|
|
"Gegenwoerter.csv UseGenitivSolve.ipynb\r\n",
|
|
"GenitivSolve.py Use_GenitivSolve.py\r\n",
|
|
"GS_Utils.py Use_Medio.ipynb\r\n",
|
|
"hkldb1Aktiv.hkl UsePassiv2Aktiv.ipynb\r\n",
|
|
"hkldb1Aktiv_old.hkl Use_Passiv2Aktiv.py\r\n",
|
|
"hkldb1Vorgangspassiv.hkl UseSayYes.ipynb\r\n",
|
|
"hkldb1Vorgangspassiv_old.hkl Use_SayYes.py\r\n",
|
|
"hkldb1Zustandspassiv.hkl UseSentSeg.ipynb\r\n",
|
|
"hkldb1Zustandspassiv_old.hkl Use_SentSeg.py\r\n",
|
|
"hkldb2Aktiv.hkl UseSolveShorts.ipynb\r\n",
|
|
"hkldb2Aktiv_old.hkl Use_SolveShorts.py\r\n",
|
|
"hkldb2Vorgangspassiv.hkl \u001b[01;34mVerbesserungen\u001b[0m/\r\n",
|
|
"hkldb2Vorgangspassiv_old.hkl voilastate.db\r\n",
|
|
"hkldb2Zustandspassiv.hkl Vorgangspassiv.txt\r\n",
|
|
"hkldb2Zustandspassiv_old.hkl Zustandspassiv.txt\r\n",
|
|
"hkldb_AllAktiv.hkl\r\n"
|
|
]
|
|
}
|
|
],
|
|
"source": [
|
|
"ls\n"
|
|
]
|
|
},
|
|
{
|
|
"cell_type": "code",
|
|
"execution_count": null,
|
|
"metadata": {},
|
|
"outputs": [],
|
|
"source": []
|
|
}
|
|
],
|
|
"metadata": {
|
|
"kernelspec": {
|
|
"display_name": "Python 3",
|
|
"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": "3.5.3"
|
|
}
|
|
},
|
|
"nbformat": 4,
|
|
"nbformat_minor": 2
|
|
}
|