stimmenfryslan/notebooks/To Per Word.ipynb

103 lines
2.6 KiB
Plaintext
Raw Normal View History

2018-09-28 10:35:17 +02:00
{
"cells": [
{
"cell_type": "code",
"execution_count": 6,
"metadata": {},
"outputs": [],
"source": [
"from glob import glob\n",
"\n",
"import os"
]
},
{
"cell_type": "code",
"execution_count": 16,
"metadata": {},
"outputs": [],
"source": [
"waves = [\n",
" wave\n",
" for location in os.listdir('data')\n",
" for date in os.listdir(os.path.join('data', location))\n",
" for wave in os.listdir(os.path.join('data', location, date))\n",
"]\n",
"assert len(waves) == len(set(waves)), \"Not all filenames are unique :(\""
]
},
{
"cell_type": "code",
"execution_count": 25,
"metadata": {},
"outputs": [],
"source": [
"for location in os.listdir('data'):\n",
" for date in os.listdir(os.path.join('data', location)):\n",
" for wave in os.listdir(os.path.join('data', location, date)):\n",
" source = os.path.join('data', location, date, wave)\n",
" destination = os.path.join('per_word', wave.split('_')[1])\n",
" if not os.path.isdir(destination):\n",
" os.mkdir(destination)\n",
" os.rename(source, os.path.join(destination, wave))"
]
},
{
"cell_type": "code",
"execution_count": 26,
"metadata": {},
"outputs": [],
"source": [
"import pandas\n",
"\n",
"data = pandas.read_csv('/home/herbert/picture-game-result-export.csv', delimiter=';')\n",
"data['Filename'] = [x.split('/')[-1] for x in data['Opname']]"
]
},
{
"cell_type": "code",
"execution_count": 52,
"metadata": {},
"outputs": [],
"source": [
"for word in os.listdir('per_word'):\n",
" for wave in os.listdir(os.path.join('per_word', word)):\n",
" source = os.path.join('per_word', word, wave)\n",
" if wave not in relevant:\n",
" destination = os.path.join('per_word', word, 'irrelevant_accent_' + wave)\n",
" os.rename(source, destination)"
]
},
{
"cell_type": "code",
"execution_count": 63,
"metadata": {},
"outputs": [],
"source": [
"data.to_excel('/home/herbert/picture-game-result-export-filename.xlsx')"
]
}
],
"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": "3.5.2"
}
},
"nbformat": 4,
"nbformat_minor": 2
}