TreeGen/trans_train.py at master · zysszy/TreeGen · GitHub

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
f = open("train.txt", "r")
lines = f.readlines()
f.close()

f = open("train_trans.txt", "w")

rule = []

def find_father(words, i):
    index = i - 1
    count = 0
    while index >= 0:
        #print ("fa", words[i])
        if "^" in words[index]:
            count -= 1
        else:
            count += 1
        if count == 1:
#            exit()
            print (words[index])
            return rule[index]
        index -= 1
    return -1

def find_father_index(words, i):
    index = i - 1
    count = 0
    while index >= 0:
        #print ("fa", words[i])
        if "^" in words[index]:
            count -= 1
        else:
            count += 1
        if count == 1:
#            exit()
#            print (words[index])
            #return words[index]
            return index
        index -= 1
    return -1

def get_deepth(line, rule, index_rule):
    words = line.strip().split()
    #depth = len(words)
    #if words[0] == "arguments":
    #    return "a" + str(depth)
    #return depth
    deepth = -1
    bf = ""
    for i in range(len(words)):
        word = words[i]
        if "^" in word:
            deepth -= 1
        else:
            if "node_gen" == words[i]:
            #    if bf == "arguments":
            #        return "a" + str(deepth)
                rule.append(-1)
                print ("-----")
                print (words[i])
                index = find_father_index(words, i)
                rule[index] = index_rule
                return find_father(words, index)
                #return father
            deepth += 1
        bf = word
        if i >= len(rule):
            rule.append(-1)
            #if "node_gen" == word:
            #    return deepth
    return -1


def write_one(lines, index):
    line_now = ""
    ast = ""
    deepth = []
    numbers = 1
    for i in range(9):
        if i == 5:
            line_now = lines[index + i].strip() + " " + str(1 + int(lines[index + i + 1].strip())) + "\n"
            numbers = line_now.strip().split()
            for tt in range(len(numbers)):
                numbers[tt] = str(int(numbers[tt]))
            ll = " ".join(numbers)
            f.write(ll + "\n")
        elif i == 6:
            rule_r = line_now.strip().split()
            l = []
            for r in rule_r:
                l.append(int(r))
                f.write(str(int(r) - 1))
                f.write(" ")
            f.write("\n")
        elif i == 7:
            continue
        else:
            f.write(lines[index + i].strip() + "\n")
            if i == 7:
                length = len(numbers)
                global rule
                rule = []
                index_rule = 0
                #for site in range(length)[::-1]:
                # if i == 7:
                #    deepth.append(get_deepth(lines[index + i - 6 - 8 * site], rule, index_rule))
                #    index_rule += 1
                #print (deepth)
#                exit()
            if i == 8:
                length = len(numbers)
                for site in range(length)[::-1]:
                    if i == 8:
                        deepth.append(int(lines[index + i - 9 * site]))
                        #index_rule += 1
                #deepth.append(int(lines[index + i]))
    #deepth = deepth[::-1]
    for i in deepth:
        f.write(str(i) + " ")
    f.write("\n")


for i in range(10, len(lines) + 9):
    t = i % 9
    if t == 5: # predicted rules lines
       if i >= len(lines):
           rules = []
       else:
           rules = lines[i].strip().split()
       if len(rules) == 0: #
           write_one(lines, i - 9 - 5)
           #rules = lines[i - 8].strip().split()

f.close()

f = open("dev.txt", "r")
lines = f.readlines()
f.close()

f = open("dev_trans.txt", "w")


for i in range(10, len(lines) + 9):
    t = i % 9
    if t == 5: # predicted rules lines
       if i >= len(lines):
           rules = []
       else:
           rules = lines[i].strip().split()
       if len(rules) == 0: #
           print (i)
           write_one(lines, i - 9 - 5)
           #rules = lines[i - 8].strip().split()

f.close()

f = open("test.txt", "r")
lines = f.readlines()
f.close()

f = open("test_trans.txt", "w")


for i in range(10, len(lines) + 9):
    t = i % 9
    if t == 5: # predicted rules lines
       if i >= len(lines):
           rules = []
       else:
           rules = lines[i].strip().split()
       if len(rules) == 0: #
           print (i)
           write_one(lines, i - 9 - 5)
           #rules = lines[i - 8].strip().split()

f.close()