aniqbinlutpi's picture
download
raw
3.65 kB
{
"cells": [
{
"cell_type": "code",
"execution_count": 1,
"id": "b6755ac9",
"metadata": {},
"outputs": [],
"source": [
"from glob import glob\n",
"import json\n",
"import re"
]
},
{
"cell_type": "code",
"execution_count": 39,
"id": "1c20e2f4",
"metadata": {},
"outputs": [
{
"data": {
"text/plain": [
"['terengganu-malaysia.jsonl',\n",
" 'kelantan-malaysia-code.jsonl',\n",
" 'kelantan-malaysia.jsonl',\n",
" 'perak-malaysia-code.jsonl',\n",
" 'local-malay-malaysia.jsonl',\n",
" 'pahang-malaysia.jsonl',\n",
" 'pahang-malaysia-code.jsonl',\n",
" 'kedah-malaysia.jsonl',\n",
" 'sarawak-malaysia.jsonl',\n",
" 'sarawak-malaysia-code.jsonl',\n",
" 'sabah-malaysia-code.jsonl',\n",
" 'local-malay-malaysia-code.jsonl',\n",
" 'negeri-sembilan-malaysia-code.jsonl',\n",
" 'sabah-malaysia.jsonl',\n",
" 'melaka-malaysia.jsonl',\n",
" 'terengganu-malaysia-code.jsonl',\n",
" 'johor-malaysia-code.jsonl',\n",
" 'perak-malaysia.jsonl',\n",
" 'negeri-sembilan-malaysia.jsonl',\n",
" 'johor-malaysia.jsonl',\n",
" 'melaka-malaysia-code.jsonl',\n",
" 'kedah-malaysia-code.jsonl']"
]
},
"execution_count": 39,
"metadata": {},
"output_type": "execute_result"
}
],
"source": [
"files = glob('*.jsonl')\n",
"files"
]
},
{
"cell_type": "code",
"execution_count": 40,
"id": "1e4d73fd",
"metadata": {},
"outputs": [
{
"data": {
"text/plain": [
"[]"
]
},
"execution_count": 40,
"metadata": {},
"output_type": "execute_result"
}
],
"source": [
"tags = set()\n",
"\n",
"for f in files:\n",
" with open(f) as fopen:\n",
" for l in fopen:\n",
" l = json.loads(l)\n",
" for v in l.values():\n",
" if not isinstance(v, str):\n",
" continue\n",
" t = re.findall(r'</?[^>]+>', v)\n",
" t = [t_ for t_ in t if 'http' not in t_]\n",
" tags.update(set(t))\n",
"rejected = [t for t in tags if 'translate' in t.lower()]\n",
"rejected"
]
},
{
"cell_type": "code",
"execution_count": 38,
"id": "8cc6425e",
"metadata": {},
"outputs": [],
"source": [
"for f in files:\n",
" with open(f) as fopen:\n",
" alls = []\n",
" for l in fopen:\n",
" l = json.loads(l)\n",
" for k in l.keys():\n",
" if not isinstance(l[k], str):\n",
" continue\n",
" for r in rejected:\n",
" l[k] = l[k].replace(r, '').strip()\n",
" alls.append(l)\n",
" with open(f, 'w') as fopen:\n",
" for a in alls:\n",
" fopen.write(f'{json.dumps(a)}\\n')\n",
" fopen.flush()"
]
},
{
"cell_type": "code",
"execution_count": null,
"id": "eb771485",
"metadata": {},
"outputs": [],
"source": []
}
],
"metadata": {
"kernelspec": {
"display_name": "python3.10",
"language": "python",
"name": "python3.10"
},
"language_info": {
"codemirror_mode": {
"name": "ipython",
"version": 3
},
"file_extension": ".py",
"mimetype": "text/x-python",
"name": "python",
"nbconvert_exporter": "python",
"pygments_lexer": "ipython3",
"version": "3.10.15"
}
},
"nbformat": 4,
"nbformat_minor": 5
}

Xet Storage Details

Size:
3.65 kB
·
Xet hash:
bd968e8c557ca36b34db200f3792ae2447d62a54f5797fa545093d7fffebe3b2

Xet efficiently stores files, intelligently splitting them into unique chunks and accelerating uploads and downloads. More info.