Datasets:
add notebook that was used to load the data
Browse files- load_dataset_merlin.ipynb +142 -0
load_dataset_merlin.ipynb
ADDED
@@ -0,0 +1,142 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
1 |
+
{
|
2 |
+
"cells": [
|
3 |
+
{
|
4 |
+
"cell_type": "code",
|
5 |
+
"execution_count": null,
|
6 |
+
"id": "7301572a-4803-4a16-b262-74e41e25803e",
|
7 |
+
"metadata": {},
|
8 |
+
"outputs": [],
|
9 |
+
"source": [
|
10 |
+
"from pathlib import Path\n",
|
11 |
+
"\n",
|
12 |
+
"import pandas as pd\n",
|
13 |
+
"from datasets import Dataset, DatasetDict, load_dataset\n",
|
14 |
+
"from huggingface_hub import Repository, create_repo\n",
|
15 |
+
"from selectolax.parser import HTMLParser"
|
16 |
+
]
|
17 |
+
},
|
18 |
+
{
|
19 |
+
"cell_type": "code",
|
20 |
+
"execution_count": null,
|
21 |
+
"id": "a898baed-3640-4ca8-9d3d-b88f6c85a428",
|
22 |
+
"metadata": {},
|
23 |
+
"outputs": [],
|
24 |
+
"source": [
|
25 |
+
"def _parse_start_end(node):\n",
|
26 |
+
" return int(node.attrs[\"start\"][1:]), int(node.attrs[\"end\"][1:])\n",
|
27 |
+
"\n",
|
28 |
+
"\n",
|
29 |
+
"def get_original_text(sent_toks) -> str:\n",
|
30 |
+
" empty_tokens = [i for i, t in enumerate(sent_toks) if not t.text().strip()]\n",
|
31 |
+
" org_sent_toks = [t.text() for i, t in enumerate(sent_toks) if not i in empty_tokens]\n",
|
32 |
+
" return \" \".join(org_sent_toks)\n",
|
33 |
+
"\n",
|
34 |
+
"\n",
|
35 |
+
"def get_corrected_text(toks_cor, last_end, sent_end) -> str:\n",
|
36 |
+
" cor_toks = []\n",
|
37 |
+
" for tok in toks_cor:\n",
|
38 |
+
" tok_start, tok_end = _parse_start_end(tok)\n",
|
39 |
+
" if tok_start >= last_end and tok_end <= sent_end:\n",
|
40 |
+
" cor_toks.append(tok.text())\n",
|
41 |
+
" last_end = tok_end\n",
|
42 |
+
" return last_end, \" \".join(cor_toks)\n",
|
43 |
+
"\n",
|
44 |
+
"\n",
|
45 |
+
"def process_doc(doc, path):\n",
|
46 |
+
" toks = doc.select('tier[category=\"tok\"] event').matches\n",
|
47 |
+
" toks_cor = doc.select('tier[category=\"TH1\"] event').matches\n",
|
48 |
+
" sents = doc.select('tier[category=\"sentence\"] event').matches\n",
|
49 |
+
"\n",
|
50 |
+
" last_end = 0\n",
|
51 |
+
" for sent_no, org_sent in enumerate(sents):\n",
|
52 |
+
" sent_start, sent_end = _parse_start_end(org_sent)\n",
|
53 |
+
" sent_toks = toks[sent_start:sent_end]\n",
|
54 |
+
" original_text = get_original_text(sent_toks)\n",
|
55 |
+
" last_end, corrected_text = get_corrected_text(toks_cor, last_end, sent_end)\n",
|
56 |
+
"\n",
|
57 |
+
" yield (\n",
|
58 |
+
" {\n",
|
59 |
+
" \"original\": original_text,\n",
|
60 |
+
" \"corrected\": corrected_text,\n",
|
61 |
+
" \"id\": f\"{path.stem}-{sent_no}\",\n",
|
62 |
+
" }\n",
|
63 |
+
" )"
|
64 |
+
]
|
65 |
+
},
|
66 |
+
{
|
67 |
+
"cell_type": "code",
|
68 |
+
"execution_count": null,
|
69 |
+
"id": "191fe2e3-8a4e-47e2-9316-5b6028662c02",
|
70 |
+
"metadata": {},
|
71 |
+
"outputs": [],
|
72 |
+
"source": [
|
73 |
+
"DATASET_NAME = \"merlin\"\n",
|
74 |
+
"dataset_path = Path.home() / DATASET_NAME\n",
|
75 |
+
"if not Path(dataset_path).exists():\n",
|
76 |
+
" repo_url = create_repo(name=DATASET_NAME, repo_type=\"dataset\")\n",
|
77 |
+
" repo = Repository(local_dir=str(dataset_path), clone_from=repo_url)\n",
|
78 |
+
" repo.lfs_track(\"*.jsonl\")"
|
79 |
+
]
|
80 |
+
},
|
81 |
+
{
|
82 |
+
"cell_type": "code",
|
83 |
+
"execution_count": null,
|
84 |
+
"id": "005d059f-5de4-4f14-bde3-0cc9ff2435c0",
|
85 |
+
"metadata": {},
|
86 |
+
"outputs": [],
|
87 |
+
"source": [
|
88 |
+
"MERLN_EXMARALDA_BASE = Path.home() / Path(\n",
|
89 |
+
" \"Downloads/MERLIN Written Learner Corpus for Czech, German, Italian 1.1/merlin-exmaralda-v1.1/\"\n",
|
90 |
+
")\n",
|
91 |
+
"\n",
|
92 |
+
"for lang in (\"german\", \"czech\", \"italian\"):\n",
|
93 |
+
" lang_docs = []\n",
|
94 |
+
" for path in (MERLN_EXMARALDA_BASE / lang).glob(\"*.exb\"):\n",
|
95 |
+
" with open(path) as fp:\n",
|
96 |
+
" xml = HTMLParser(fp.read())\n",
|
97 |
+
" docs = list(process_doc(xml, path))\n",
|
98 |
+
" lang_docs.extend(docs)\n",
|
99 |
+
" Dataset.from_dict(pd.DataFrame(lang_docs)).to_json(dataset_path / f\"{lang}.jsonl\")"
|
100 |
+
]
|
101 |
+
},
|
102 |
+
{
|
103 |
+
"cell_type": "code",
|
104 |
+
"execution_count": null,
|
105 |
+
"id": "f9d396b2-98dc-4c04-950f-0332a3a6d751",
|
106 |
+
"metadata": {},
|
107 |
+
"outputs": [],
|
108 |
+
"source": [
|
109 |
+
"repo.push_to_hub()"
|
110 |
+
]
|
111 |
+
},
|
112 |
+
{
|
113 |
+
"cell_type": "code",
|
114 |
+
"execution_count": null,
|
115 |
+
"id": "91377eaf-ffac-4df0-9c85-4f6ee979f99f",
|
116 |
+
"metadata": {},
|
117 |
+
"outputs": [],
|
118 |
+
"source": []
|
119 |
+
}
|
120 |
+
],
|
121 |
+
"metadata": {
|
122 |
+
"kernelspec": {
|
123 |
+
"display_name": "Python 3 (ipykernel)",
|
124 |
+
"language": "python",
|
125 |
+
"name": "python3"
|
126 |
+
},
|
127 |
+
"language_info": {
|
128 |
+
"codemirror_mode": {
|
129 |
+
"name": "ipython",
|
130 |
+
"version": 3
|
131 |
+
},
|
132 |
+
"file_extension": ".py",
|
133 |
+
"mimetype": "text/x-python",
|
134 |
+
"name": "python",
|
135 |
+
"nbconvert_exporter": "python",
|
136 |
+
"pygments_lexer": "ipython3",
|
137 |
+
"version": "3.9.7"
|
138 |
+
}
|
139 |
+
},
|
140 |
+
"nbformat": 4,
|
141 |
+
"nbformat_minor": 5
|
142 |
+
}
|