-
Notifications
You must be signed in to change notification settings - Fork 0
/
run.bash
247 lines (173 loc) · 7.1 KB
/
run.bash
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
214
215
216
217
218
219
220
221
222
223
224
225
226
227
228
229
230
231
232
233
234
235
236
237
238
239
240
241
242
243
244
245
246
#!/bin/bash
# Version: 2
# Written by: Tirtharaj Dash, Ph.D. Student, BITS Pilani, Goa Campus, India
# Date: During December 2019
# E-mail: tirtharaj@goa.bits-pilani.ac.in
# Work: GNN, Vertex-Enriched GNN (VEGNN), VEGNN_HS (VEGNN')
# What are these?
# GNN: Graph Neural Network (simple graph strutured data): Nodes contain only atom features
# VEGNN: GNN with Relational Features at Nodes
# VEGNN_HS (VEGNN'): GNN with Relational Features and ILP Features at Nodes
# Purpose: bash script to run graph classification methods on NCI data (73 problems).
# Learning Type: Classification (Binary)
# Result storage mapping: [ Methods: GNN, VEGNN, VEGNN_HS | ID: 1 (networks1.py), 2 (networks2.py), 3 (networks3.py), 4 (networks4.py), 5 (networks5.py) ]
#path settings
prefixdir="/home/dell5810/tdash/DataForVEGNN/Datasets"
relfeatsdir="/home/dell5810/tdash/DataForVEGNN/ILPFeats"
trntstsplitdir="/home/dell5810/tdash/DataForVEGNN/TrainTestSplit"
#create the directory where the Results will be stored
resultdir="Result_GNN5_Czech"
mkdir $resultdir
#for each dataset in the list
for dataset in `cat datasets`
do
echo "Working on: $dataset"
#copy the input: train and test files to run dir
rm -rf ./data/DS/*
mkdir ./data/DS/raw
cp $prefixdir/$dataset/DS_*.txt ./data/DS/raw/.
##based on your choices, you need to un-comment the following(1-3)
#1: without background knowledge (AB features)
cat ./data/DS/raw/DS_node_attributes_bin.txt | cut -d, -f1-6 > ./data/DS/raw/DS_node_attributes.txt
#2: with background knowledge (ABFR features)
#mv ./data/DS/raw/DS_node_attributes_bin.txt ./data/DS/raw/DS_node_attributes.txt
#3: with background knowledge (ABFR features) + concatenated Relational Features (HS)
#paste -d "," ./data/DS/raw/DS_node_attributes_bin.txt $relfeatsdir/$dataset/HS_n1000/RelFeats.csv > ./data/DS/raw/DS_node_attributes.txt
#copy the train_test split info
cp $trntstsplitdir/$dataset/*_split ./data/DS/.
#run the python program
python main.py --dataset DS
#store the results
if [ -d ./$resultdir/$dataset ]
then
rm -rf ./$resultdir/$dataset
fi
mkdir ./$resultdir/$dataset
#mv ./data/DS/* $resultdir/$dataset/.
mv score.txt $resultdir/$dataset/.
mv latest.pth $resultdir/$dataset/.
done
resultdir="Result_VEGNN5_Czech"
mkdir $resultdir
#for each dataset in the list
for dataset in `cat datasets`
do
echo "Working on: $dataset"
#copy the input: train and test files to run dir
rm -rf ./data/DS/*
mkdir ./data/DS/raw
cp $prefixdir/$dataset/DS_*.txt ./data/DS/raw/.
##based on your choices, you need to un-comment the following(1-3)
#1: without background knowledge (AB features)
#cat ./data/DS/raw/DS_node_attributes_bin.txt | cut -d, -f1-6 > ./data/DS/raw/DS_node_attributes.txt
#2: with background knowledge (ABFR features)
mv ./data/DS/raw/DS_node_attributes_bin.txt ./data/DS/raw/DS_node_attributes.txt
#3: with background knowledge (ABFR features) + concatenated Relational Features (HS)
#paste -d "," ./data/DS/raw/DS_node_attributes_bin.txt $relfeatsdir/$dataset/HS_n1000/RelFeats.csv > ./data/DS/raw/DS_node_attributes.txt
#copy the train_test split info
cp $trntstsplitdir/$dataset/*_split ./data/DS/.
#run the python program
python main.py --dataset DS
#store the results
if [ -d ./$resultdir/$dataset ]
then
rm -rf ./$resultdir/$dataset
fi
mkdir ./$resultdir/$dataset
#mv ./data/DS/* $resultdir/$dataset/.
mv score.txt $resultdir/$dataset/.
mv latest.pth $resultdir/$dataset/.
done
resultdir="Result_VEGNN5_HS100_Czech"
mkdir $resultdir
#for each dataset in the list
for dataset in `cat datasets`
do
echo "Working on: $dataset"
#copy the input: train and test files to run dir
rm -rf ./data/DS/*
mkdir ./data/DS/raw
cp $prefixdir/$dataset/DS_*.txt ./data/DS/raw/.
##based on your choices, you need to un-comment the following(1-3)
#1: without background knowledge (AB features)
#cat ./data/DS/raw/DS_node_attributes_bin.txt | cut -d, -f1-6 > ./data/DS/raw/DS_node_attributes.txt
#2: with background knowledge (ABFR features)
#mv ./data/DS/raw/DS_node_attributes_bin.txt ./data/DS/raw/DS_node_attributes.txt
#3: with background knowledge (ABFR features) + concatenated Relational Features (HS)
paste -d "," ./data/DS/raw/DS_node_attributes_bin.txt $relfeatsdir/$dataset/HS_n100/RelFeats.csv > ./data/DS/raw/DS_node_attributes.txt
#copy the train_test split info
cp $trntstsplitdir/$dataset/*_split ./data/DS/.
#run the python program
python main.py --dataset DS
#store the results
if [ -d ./$resultdir/$dataset ]
then
rm -rf ./$resultdir/$dataset
fi
mkdir ./$resultdir/$dataset
#mv ./data/DS/* $resultdir/$dataset/.
mv score.txt $resultdir/$dataset/.
mv latest.pth $resultdir/$dataset/.
done
resultdir="Result_VEGNN5_HS500_Czech"
mkdir $resultdir
#for each dataset in the list
for dataset in `cat datasets`
do
echo "Working on: $dataset"
#copy the input: train and test files to run dir
rm -rf ./data/DS/*
mkdir ./data/DS/raw
cp $prefixdir/$dataset/DS_*.txt ./data/DS/raw/.
##based on your choices, you need to un-comment the following(1-3)
#1: without background knowledge (AB features)
#cat ./data/DS/raw/DS_node_attributes_bin.txt | cut -d, -f1-6 > ./data/DS/raw/DS_node_attributes.txt
#2: with background knowledge (ABFR features)
#mv ./data/DS/raw/DS_node_attributes_bin.txt ./data/DS/raw/DS_node_attributes.txt
#3: with background knowledge (ABFR features) + concatenated Relational Features (HS)
paste -d "," ./data/DS/raw/DS_node_attributes_bin.txt $relfeatsdir/$dataset/HS_n500/RelFeats.csv > ./data/DS/raw/DS_node_attributes.txt
#copy the train_test split info
cp $trntstsplitdir/$dataset/*_split ./data/DS/.
#run the python program
python main.py --dataset DS
#store the results
if [ -d ./$resultdir/$dataset ]
then
rm -rf ./$resultdir/$dataset
fi
mkdir ./$resultdir/$dataset
#mv ./data/DS/* $resultdir/$dataset/.
mv score.txt $resultdir/$dataset/.
mv latest.pth $resultdir/$dataset/.
done
resultdir="Result_VEGNN5_HS1000_Czech"
mkdir $resultdir
#for each dataset in the list
for dataset in `cat datasets`
do
echo "Working on: $dataset"
#copy the input: train and test files to run dir
rm -rf ./data/DS/*
mkdir ./data/DS/raw
cp $prefixdir/$dataset/DS_*.txt ./data/DS/raw/.
##based on your choices, you need to un-comment the following(1-3)
#1: without background knowledge (AB features)
#cat ./data/DS/raw/DS_node_attributes_bin.txt | cut -d, -f1-6 > ./data/DS/raw/DS_node_attributes.txt
#2: with background knowledge (ABFR features)
#mv ./data/DS/raw/DS_node_attributes_bin.txt ./data/DS/raw/DS_node_attributes.txt
#3: with background knowledge (ABFR features) + concatenated Relational Features (HS)
paste -d "," ./data/DS/raw/DS_node_attributes_bin.txt $relfeatsdir/$dataset/HS_n1000/RelFeats.csv > ./data/DS/raw/DS_node_attributes.txt
#copy the train_test split info
cp $trntstsplitdir/$dataset/*_split ./data/DS/.
#run the python program
python main.py --dataset DS
#store the results
if [ -d ./$resultdir/$dataset ]
then
rm -rf ./$resultdir/$dataset
fi
mkdir ./$resultdir/$dataset
#mv ./data/DS/* $resultdir/$dataset/.
mv score.txt $resultdir/$dataset/.
mv latest.pth $resultdir/$dataset/.
done