DataPreparation.ipynb
6.2 KB
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
61
62
63
64
65
66
67
68
69
70
71
72
73
74
75
76
77
78
79
80
81
82
83
84
85
86
87
88
89
90
91
92
93
94
95
96
97
98
99
100
101
102
103
104
105
106
107
108
109
110
111
112
113
114
115
116
117
118
119
120
121
122
123
124
125
126
127
128
129
130
131
132
133
134
135
136
137
138
139
140
141
142
143
144
145
146
147
148
149
150
151
152
153
154
155
156
157
158
159
160
161
162
163
164
165
166
167
168
169
170
171
172
173
174
175
176
177
178
179
180
181
182
183
184
185
186
187
188
189
190
191
192
193
194
195
196
197
198
199
200
201
202
203
204
205
206
207
208
209
210
211
212
213
{
"cells": [
{
"cell_type": "code",
"execution_count": 46,
"id": "5cd26f6f",
"metadata": {},
"outputs": [],
"source": [
"import os\n",
"\n",
"from datasets import load_dataset\n",
"\n",
"from IPython.display import display"
]
},
{
"cell_type": "code",
"execution_count": 47,
"id": "fecef4af",
"metadata": {},
"outputs": [
{
"name": "stderr",
"output_type": "stream",
"text": [
"Found cached dataset pdb_c_beta (/home/kkrasnowska/.cache/huggingface/datasets/pdb_c_beta/pdb_c_beta/0.2.0/d9c6dc764ae2a3483fa112c6159db4a0342dba8083bdb3b5981c45435b0692e1)\n"
]
},
{
"data": {
"application/vnd.jupyter.widget-view+json": {
"model_id": "1c89c7103bba4347a3fa7d23cac42cfe",
"version_major": 2,
"version_minor": 0
},
"text/plain": [
" 0%| | 0/3 [00:00<?, ?it/s]"
]
},
"metadata": {},
"output_type": "display_data"
}
],
"source": [
"pdbc_dataset = load_dataset('../pdb_c_beta')"
]
},
{
"cell_type": "code",
"execution_count": 48,
"id": "23da801f",
"metadata": {},
"outputs": [],
"source": [
"CONLLU_DIR = 'connlu'\n",
"! rm -r {CONLLU_DIR}\n",
"! mkdir {CONLLU_DIR}"
]
},
{
"cell_type": "code",
"execution_count": 50,
"id": "91fb3bf3",
"metadata": {},
"outputs": [],
"source": [
"import sys\n",
"sys.path.append('../')\n",
"from neural_parser.hybrid_tree_utils import tree_from_dataset_instance"
]
},
{
"cell_type": "code",
"execution_count": 60,
"id": "c105feff",
"metadata": {},
"outputs": [
{
"name": "stdout",
"output_type": "stream",
"text": [
"train\n",
" connlu/pdbc-train.conllu\n",
" 17659\n",
" connlu/pdbc-cont-train.conllu\n",
" 15903\n",
"validation\n",
" connlu/pdbc-validation.conllu\n",
" 2211\n",
" connlu/pdbc-cont-validation.conllu\n",
" 1980\n",
"test\n",
" connlu/pdbc-test.conllu\n",
" 2205\n",
" connlu/pdbc-cont-test.conllu\n",
" 1990\n"
]
}
],
"source": [
"features = pdbc_dataset['train'].features\n",
"\n",
"for part, dataset in pdbc_dataset.items():\n",
" print(part)\n",
" s_cont, s_all = [], [] \n",
" for sentence in dataset:\n",
" # TODO! check if discont\n",
" tokens = sentence['tokens']\n",
" lemmas = sentence['lemmas']\n",
" heads = sentence['heads']\n",
" heads = [h + 1 if h is not None else 0 for i, h in enumerate(heads)]\n",
" deprels = [features['deprels'].feature.int2str(d) for d in sentence['deprels']]\n",
" deprels = ['root' if deprel == 'ROOT' else deprel for deprel in deprels]\n",
" rows = [f'# text = {\" \".join(tokens)}'] + [\n",
" f'{i + 1}\\t{t}\\t{l}\\t_\\t_\\t_\\t{h}\\t{d}\\t{h}:{d}\\t_'\n",
" for i, (t, l, h, d) in enumerate(zip(tokens, lemmas, heads, deprels))\n",
" ]\n",
" s_all.append(rows)\n",
" if tree_from_dataset_instance(sentence, features).is_continuous():\n",
" s_cont.append(rows)\n",
" f_all = os.path.join(CONLLU_DIR, f'pdbc-{part}.conllu')\n",
" f_cont = os.path.join(CONLLU_DIR, f'pdbc-cont-{part}.conllu')\n",
" with open(f_all, 'w') as f:\n",
" print(' ', f_all)\n",
" print(' ', len(s_all))\n",
" for rows in s_all:\n",
" print('\\n'.join(rows), end='\\n\\n', file=f)\n",
" with open(f_cont, 'w') as f:\n",
" print(' ', f_cont)\n",
" print(' ', len(s_cont))\n",
" for rows in s_cont:\n",
" print('\\n'.join(rows), end='\\n\\n', file=f)"
]
},
{
"cell_type": "code",
"execution_count": 61,
"id": "c849233c",
"metadata": {},
"outputs": [
{
"name": "stdout",
"output_type": "stream",
"text": [
" 32509 319813 1398303 connlu/pdbc-cont-test.conllu\n",
" 32509 319813 1198902 connlu/pdbc-cont-test-pred.conllu\n",
" 271337 2682725 11781617 connlu/pdbc-cont-train.conllu\n",
" 33491 330792 1452373 connlu/pdbc-cont-validation.conllu\n",
" 33491 330792 1244192 connlu/pdbc-cont-validation-pred.conllu\n",
" 37754 373431 1639937 connlu/pdbc-test.conllu\n",
" 37754 373431 1406776 connlu/pdbc-test-pred.conllu\n",
" 315364 3133712 13808053 connlu/pdbc-train.conllu\n",
" 38987 386865 1704685 connlu/pdbc-validation.conllu\n",
" 38987 386865 1461922 connlu/pdbc-validation-pred.conllu\n",
" 872183 8638239 37096760 total\n"
]
}
],
"source": [
"! wc {CONLLU_DIR}/*.conllu"
]
},
{
"cell_type": "code",
"execution_count": 62,
"id": "6b571716",
"metadata": {},
"outputs": [
{
"name": "stdout",
"output_type": "stream",
"text": [
"# text = Skośnooka dziewczynka trzyma w rękach drewniane pałeczki , a przed nią znajdują się naczynia kuchenne .\r\n",
"1\tSkośnooka\tskośnooki\t_\t_\t_\t2\tadjunct\t2:adjunct\t_\r\n",
"2\tdziewczynka\tdziewczynka\t_\t_\t_\t3\tsubj\t3:subj\t_\r\n",
"3\ttrzyma\ttrzymać\t_\t_\t_\t9\tconjunct\t9:conjunct\t_\r\n",
"4\tw\tw\t_\t_\t_\t3\tadjunct_locat\t3:adjunct_locat\t_\r\n",
"5\trękach\tręka\t_\t_\t_\t4\tcomp\t4:comp\t_\r\n",
"6\tdrewniane\tdrewniany\t_\t_\t_\t7\tadjunct\t7:adjunct\t_\r\n",
"7\tpałeczki\tpałeczka\t_\t_\t_\t3\tobj\t3:obj\t_\r\n",
"8\t,\t,\t_\t_\t_\t9\tpunct\t9:punct\t_\r\n",
"9\ta\ta\t_\t_\t_\t0\troot\t0:root\t_\r\n"
]
}
],
"source": [
"! head {CONLLU_DIR}/pdbc-train.conllu"
]
}
],
"metadata": {
"kernelspec": {
"display_name": "TF_zajecia",
"language": "python",
"name": "tf_zajecia"
},
"language_info": {
"codemirror_mode": {
"name": "ipython",
"version": 3
},
"file_extension": ".py",
"mimetype": "text/x-python",
"name": "python",
"nbconvert_exporter": "python",
"pygments_lexer": "ipython3",
"version": "3.10.6"
}
},
"nbformat": 4,
"nbformat_minor": 5
}