Datasets:
Upload 2 files
Browse files- SciDMT/SciDMT.tar +3 -0
- SciDMT_demo.ipynb +235 -0
SciDMT/SciDMT.tar
ADDED
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
1 |
+
version https://git-lfs.github.com/spec/v1
|
2 |
+
oid sha256:a8579e90d1560fe33fcb9f503c964698b86431df8833ab69c5cc8f7c4b7a465d
|
3 |
+
size 3031326720
|
SciDMT_demo.ipynb
ADDED
@@ -0,0 +1,235 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
1 |
+
{
|
2 |
+
"cells": [
|
3 |
+
{
|
4 |
+
"cell_type": "code",
|
5 |
+
"execution_count": 1,
|
6 |
+
"metadata": {},
|
7 |
+
"outputs": [],
|
8 |
+
"source": [
|
9 |
+
"import json\n",
|
10 |
+
"import pickle\n",
|
11 |
+
"from const import *\n",
|
12 |
+
"import pandas as pd\n",
|
13 |
+
"import numpy as np"
|
14 |
+
]
|
15 |
+
},
|
16 |
+
{
|
17 |
+
"cell_type": "code",
|
18 |
+
"execution_count": 2,
|
19 |
+
"metadata": {},
|
20 |
+
"outputs": [],
|
21 |
+
"source": [
|
22 |
+
"SCIDMT_PATH = {\n",
|
23 |
+
" 'DICT': './SciDMT/SciDMT_dict.json',\n",
|
24 |
+
"\n",
|
25 |
+
" # machine learning inputs at sentence level\n",
|
26 |
+
" 'sent_xy': './SciDMT/SciDMT_sentences.p', \n",
|
27 |
+
" 'sent_eval': './SciDMT/SciDMT_E_sentences.json',\n",
|
28 |
+
" 'sent_split': './SciDMT/SciDMT_sentences_split.json',\n",
|
29 |
+
"\n",
|
30 |
+
" # document level inputs\n",
|
31 |
+
" 'doc_split': './SciDMT/SciDMT_split.json',\n",
|
32 |
+
" 'doc_eval': './SciDMT/SciDMT_E_human_annotations.json',\n",
|
33 |
+
" 'doc_text_and_meta': './SciDMT/SciDMT_papers.csv',\n",
|
34 |
+
"}"
|
35 |
+
]
|
36 |
+
},
|
37 |
+
{
|
38 |
+
"cell_type": "code",
|
39 |
+
"execution_count": 3,
|
40 |
+
"metadata": {},
|
41 |
+
"outputs": [],
|
42 |
+
"source": [
|
43 |
+
"def print_dict_structure(d, indent=0, indent_str=' '):\n",
|
44 |
+
" for key, value in d.items():\n",
|
45 |
+
" print(indent_str * indent + ' ' +str(key))\n",
|
46 |
+
" if isinstance(value, dict):\n",
|
47 |
+
" print_dict_structure(value, indent+1)\n",
|
48 |
+
" else:\n",
|
49 |
+
" if type(value) == list:\n",
|
50 |
+
" des = f'| len={len(value)} | first 3 entries={value[:3]}'\n",
|
51 |
+
" else:\n",
|
52 |
+
" des = ''\n",
|
53 |
+
" print(indent_str * (indent+1) + ' ' + str(type(value)) + des)"
|
54 |
+
]
|
55 |
+
},
|
56 |
+
{
|
57 |
+
"cell_type": "code",
|
58 |
+
"execution_count": 4,
|
59 |
+
"metadata": {},
|
60 |
+
"outputs": [
|
61 |
+
{
|
62 |
+
"name": "stdout",
|
63 |
+
"output_type": "stream",
|
64 |
+
"text": [
|
65 |
+
"dict_keys(['datasets', 'methods', 'tasks'])\n"
|
66 |
+
]
|
67 |
+
}
|
68 |
+
],
|
69 |
+
"source": [
|
70 |
+
"# Load SciDMT dictionary for entities\n",
|
71 |
+
"DICT = json.load(open(SCIDMT_PATH['DICT'], 'r'))\n",
|
72 |
+
"print(DICT.keys())"
|
73 |
+
]
|
74 |
+
},
|
75 |
+
{
|
76 |
+
"cell_type": "code",
|
77 |
+
"execution_count": 5,
|
78 |
+
"metadata": {},
|
79 |
+
"outputs": [],
|
80 |
+
"source": [
|
81 |
+
"# Load SciDMT evaluation set in sentence level\n",
|
82 |
+
"scidmt_e = pd.read_json(SCIDMT_PATH['sent_eval'])\n",
|
83 |
+
"\n",
|
84 |
+
"# load x_test, y_test\n",
|
85 |
+
"X_test, y_test = scidmt_e['X'].to_list(), scidmt_e['y'].to_list()"
|
86 |
+
]
|
87 |
+
},
|
88 |
+
{
|
89 |
+
"cell_type": "code",
|
90 |
+
"execution_count": 6,
|
91 |
+
"metadata": {},
|
92 |
+
"outputs": [
|
93 |
+
{
|
94 |
+
"name": "stdout",
|
95 |
+
"output_type": "stream",
|
96 |
+
"text": [
|
97 |
+
" is_contain\n",
|
98 |
+
" datasets\n",
|
99 |
+
" <class 'list'>| len=48049 | first 3 entries=[True, False, True]\n",
|
100 |
+
" tasks\n",
|
101 |
+
" <class 'list'>| len=48049 | first 3 entries=[True, True, True]\n",
|
102 |
+
" methods\n",
|
103 |
+
" <class 'list'>| len=48049 | first 3 entries=[True, True, True]\n",
|
104 |
+
" all\n",
|
105 |
+
" <class 'list'>| len=48049 | first 3 entries=[True, True, True]\n",
|
106 |
+
" is_test\n",
|
107 |
+
" datasets\n",
|
108 |
+
" <class 'list'>| len=48049 | first 3 entries=[False, False, False]\n",
|
109 |
+
" tasks\n",
|
110 |
+
" <class 'list'>| len=48049 | first 3 entries=[True, False, False]\n",
|
111 |
+
" methods\n",
|
112 |
+
" <class 'list'>| len=48049 | first 3 entries=[False, True, False]\n",
|
113 |
+
" all\n",
|
114 |
+
" <class 'list'>| len=48049 | first 3 entries=[False, False, False]\n",
|
115 |
+
" doc_pids\n",
|
116 |
+
" <class 'list'>| len=48049 | first 3 entries=[51881821, 51881855, 51881912]\n",
|
117 |
+
" is_0shot\n",
|
118 |
+
" datasets\n",
|
119 |
+
" <class 'list'>| len=48049 | first 3 entries=[False, False, False]\n",
|
120 |
+
" methods\n",
|
121 |
+
" <class 'list'>| len=48049 | first 3 entries=[False, False, False]\n",
|
122 |
+
" tasks\n",
|
123 |
+
" <class 'list'>| len=48049 | first 3 entries=[False, False, False]\n",
|
124 |
+
" all\n",
|
125 |
+
" <class 'list'>| len=48049 | first 3 entries=[False, False, False]\n",
|
126 |
+
" SHOT0\n",
|
127 |
+
" datasets\n",
|
128 |
+
" entity_indexs\n",
|
129 |
+
" <class 'list'>| len=10 | first 3 entries=[1561, 2889, 2810]\n",
|
130 |
+
" ml_inputs_idxs\n",
|
131 |
+
" <class 'list'>| len=147 | first 3 entries=[461223, 461224, 461225]\n",
|
132 |
+
" tasks\n",
|
133 |
+
" entity_indexs\n",
|
134 |
+
" <class 'list'>| len=10 | first 3 entries=[766, 1487, 1548]\n",
|
135 |
+
" ml_inputs_idxs\n",
|
136 |
+
" <class 'list'>| len=305 | first 3 entries=[646887, 646888, 646889]\n",
|
137 |
+
" methods\n",
|
138 |
+
" entity_indexs\n",
|
139 |
+
" <class 'list'>| len=10 | first 3 entries=[605, 1324, 1099]\n",
|
140 |
+
" ml_inputs_idxs\n",
|
141 |
+
" <class 'list'>| len=128 | first 3 entries=[1154530, 1154548, 1154550]\n"
|
142 |
+
]
|
143 |
+
}
|
144 |
+
],
|
145 |
+
"source": [
|
146 |
+
"# load document level split\n",
|
147 |
+
"DOC_SPLIT = json.load(open(SCIDMT_PATH['doc_split'], 'r'))\n",
|
148 |
+
"print_dict_structure(DOC_SPLIT)"
|
149 |
+
]
|
150 |
+
},
|
151 |
+
{
|
152 |
+
"cell_type": "code",
|
153 |
+
"execution_count": 14,
|
154 |
+
"metadata": {},
|
155 |
+
"outputs": [
|
156 |
+
{
|
157 |
+
"name": "stdout",
|
158 |
+
"output_type": "stream",
|
159 |
+
"text": [
|
160 |
+
" is_contain\n",
|
161 |
+
" datasets\n",
|
162 |
+
" <class 'list'>| len=1128148 | first 3 entries=[True, False, False]\n",
|
163 |
+
" tasks\n",
|
164 |
+
" <class 'list'>| len=1128148 | first 3 entries=[False, True, True]\n",
|
165 |
+
" methods\n",
|
166 |
+
" <class 'list'>| len=1128148 | first 3 entries=[False, False, False]\n",
|
167 |
+
" all\n",
|
168 |
+
" <class 'list'>| len=1128148 | first 3 entries=[True, True, True]\n",
|
169 |
+
" is_test\n",
|
170 |
+
" datasets\n",
|
171 |
+
" <class 'list'>| len=1128148 | first 3 entries=[False, False, False]\n",
|
172 |
+
" tasks\n",
|
173 |
+
" <class 'list'>| len=1128148 | first 3 entries=[True, True, True]\n",
|
174 |
+
" methods\n",
|
175 |
+
" <class 'list'>| len=1128148 | first 3 entries=[True, True, True]\n",
|
176 |
+
" all\n",
|
177 |
+
" <class 'list'>| len=1128148 | first 3 entries=[False, False, False]\n"
|
178 |
+
]
|
179 |
+
}
|
180 |
+
],
|
181 |
+
"source": [
|
182 |
+
"# load sentence level train/test split\n",
|
183 |
+
"with open(SCIDMT_PATH['sent_split'], 'r') as f:\n",
|
184 |
+
" SPLIT = json.load(f)\n",
|
185 |
+
"print_dict_structure(SPLIT)"
|
186 |
+
]
|
187 |
+
},
|
188 |
+
{
|
189 |
+
"cell_type": "code",
|
190 |
+
"execution_count": 15,
|
191 |
+
"metadata": {},
|
192 |
+
"outputs": [
|
193 |
+
{
|
194 |
+
"name": "stderr",
|
195 |
+
"output_type": "stream",
|
196 |
+
"text": [
|
197 |
+
"/tmp/ipykernel_1610900/687978490.py:4: VisibleDeprecationWarning: Creating an ndarray from ragged nested sequences (which is a list-or-tuple of lists-or-tuples-or ndarrays with different lengths or shapes) is deprecated. If you meant to do this, you must specify 'dtype=object' when creating the ndarray.\n",
|
198 |
+
" X_train = np.array(X)[is_train]#[:5000] ##debug\n",
|
199 |
+
"/tmp/ipykernel_1610900/687978490.py:5: VisibleDeprecationWarning: Creating an ndarray from ragged nested sequences (which is a list-or-tuple of lists-or-tuples-or ndarrays with different lengths or shapes) is deprecated. If you meant to do this, you must specify 'dtype=object' when creating the ndarray.\n",
|
200 |
+
" y_train = np.array(y)[is_train]#[:5000] ##debug\n"
|
201 |
+
]
|
202 |
+
}
|
203 |
+
],
|
204 |
+
"source": [
|
205 |
+
"# load x_train, y_train\n",
|
206 |
+
"# is_test = true are not all human annotated. Thus, we have a seperate test file\n",
|
207 |
+
"is_train = np.logical_and(SPLIT['is_test']['all']==False, SPLIT['is_contain']['all'] == True)\n",
|
208 |
+
"X, y, _, _, _ = pickle.load(open(SCIDMT_PATH['sent_xy'], 'rb'))\n",
|
209 |
+
"X_train = np.array(X)[is_train]#[:5000] ##debug\n",
|
210 |
+
"y_train = np.array(y)[is_train]#[:5000] ##debug"
|
211 |
+
]
|
212 |
+
}
|
213 |
+
],
|
214 |
+
"metadata": {
|
215 |
+
"kernelspec": {
|
216 |
+
"display_name": "py39",
|
217 |
+
"language": "python",
|
218 |
+
"name": "python3"
|
219 |
+
},
|
220 |
+
"language_info": {
|
221 |
+
"codemirror_mode": {
|
222 |
+
"name": "ipython",
|
223 |
+
"version": 3
|
224 |
+
},
|
225 |
+
"file_extension": ".py",
|
226 |
+
"mimetype": "text/x-python",
|
227 |
+
"name": "python",
|
228 |
+
"nbconvert_exporter": "python",
|
229 |
+
"pygments_lexer": "ipython3",
|
230 |
+
"version": "3.9.13"
|
231 |
+
}
|
232 |
+
},
|
233 |
+
"nbformat": 4,
|
234 |
+
"nbformat_minor": 2
|
235 |
+
}
|