-
Notifications
You must be signed in to change notification settings - Fork 0
/
Copy pathutil.py
241 lines (225 loc) · 7.35 KB
/
util.py
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
import pickle
import os
import argparse
import random
import warnings
with warnings.catch_warnings():
import numpy as np
import sklearn
import csv
import json
def onehot(x, hx, cat_col, category):
new_x = []
new_hx = []
prev_c = None
for c in cat_col:
if prev_c is None:
tmp = x[:, :c]
else:
tmp = x[:, prev_c + 1 : c]
prev_c = c
tmp = x[:, c + 1 : c]
return x, y, hx
def flatten(l):
ret = []
for a in l:
if hasattr(a, "__iter__"):
ret += flatten(a)
else:
ret.append(a)
return ret
def load_data_xsv(filename, header, ignore_col, ans_col, sep, cat_col=[], option={}):
x = []
y = []
index = []
hx = None
hy = None
col_num = 0
categorical = {}
option_data = {k: [] for k in option.keys()}
option_vals = flatten(option.values())
with open(filename) as fp:
tsv = csv.reader(fp, delimiter=sep, quotechar='"')
if header:
row = next(tsv)
hx = []
hy = []
if col_num == 0:
col_num = len(row)
print("the number of columns=", col_num)
for i in range(col_num):
if i in ignore_col:
pass
elif i in option_vals:
pass
elif i == ans_col:
hy.append(row[i])
else:
hx.append(row[i])
for line_no, row in enumerate(tsv):
if header:
line_no+=1
x_vec = []
y_vec = []
opt_vec = []
valid_line = True
if col_num == 0:
col_num = len(row)
for i in range(col_num):
if i in ignore_col:
pass
elif i == ans_col:
try:
y_vec.append(float(row[i]))
except:
valid_line = False
print("[SKIP] could not convert string to float:", row[i])
break
elif i in option_vals:
opt_vec.append((i,row[i]))
elif i in cat_col:
if row[i] not in categorical:
categorical[row[i]] = len(categorical)
x_vec.append(categorical[row[i]])
else:
if i >= len(row):
x_vec.append(np.nan)
print(
"[WARN] Line",
line_no,
"is length",
len(row),
" and shorter than its expectation",
col_num,
)
elif row[i] == "":
x_vec.append(np.nan)
else:
try:
cell=float(row[i])
except:
cell=np.nan
x_vec.append(cell)
if valid_line:
x.append(x_vec)
index.append(line_no)
if len(y_vec) > 0:
y.append(y_vec[0])
option_flag = {k: True for k in option.keys()}
for i,cell in opt_vec:
for key, value in option.items():
if (hasattr(value, "__iter__") and i in value) or (i == value):
if option_flag[key]:
option_data[key].append(cell)
option_flag[key] = False
else:
s = option_data[key][-1]
option_data[key][-1] = s + "-" + cell
return np.array(x), np.array(y), option_data, hx,np.array(index)
def load_data(filename, header=False, ignore_col=[], ans_col=[], cat_col=[], option={}):
print(filename)
if "," in filename:
pair = filename.split(",")
print("[LOAD]", pair[0])
print("[LOAD]", pair[1])
print("[LOAD]", pair[2])
x = np.load(pair[0])
y = np.load(pair[1])
opt = {}
opt["group"] = np.load(pair[2])
opt["group_type"] = "int"
return x, y, opt, None, None
_, ext = os.path.splitext(filename)
if ext == ".csv":
return load_data_xsv(
filename, header, ignore_col, ans_col, ",", cat_col, option
)
elif ext == ".tsv":
return load_data_xsv(
filename, header, ignore_col, ans_col, "\t", cat_col, option
)
elif ext == ".txt":
return load_data_xsv(
filename, header, ignore_col, ans_col, "\t", cat_col, option
)
else:
print("[ERROR] unknown file format")
return None, None, None, None, None
def extract_data(
filename, save_filename, support, header=False, ignore_col=[], ans_col=[]
):
_, ext_in = os.path.splitext(filename)
_, ext_out = os.path.splitext(save_filename)
if ext_in == ".csv":
sep_in = ","
elif ext_in == ".tsv":
sep_in = "\t"
elif ext_in == ".txt":
sep_in = "\t"
else:
print("[ERROR] unknown file format")
if ext_out == ".csv":
sep_out = ","
elif ext_out == ".tsv":
sep_out = "\t"
elif ext_out == ".txt":
sep_out = "\t"
else:
print("[ERROR] unknown file format")
extract_data_xsv(
filename, save_filename, support, header, ignore_col, ans_col, sep_in, sep_out
)
def extract_data_xsv(
filename, save_filename, support, header, ignore_col, ans_col, sep_in, sep_out
):
col_num = 0
ofp = open(save_filename, "w")
with open(filename) as fp:
tsv = csv.reader(fp, delimiter=sep_in)
if header:
row = next(tsv)
if col_num == 0:
col_num = len(row)
line = []
line_count = 0
for i in range(col_num):
if i in ignore_col:
line.append(row[i])
elif i == ans_col:
line.append(row[i])
else:
if support[line_count]:
line.append(row[i])
line_count += 1
ofp.write(sep_out.join(line))
ofp.write("\n")
for row in tsv:
if col_num == 0:
col_num = len(row)
line = []
line_count = 0
for i in range(col_num):
if i in ignore_col:
line.append(row[i])
elif i == ans_col:
line.append(row[i])
else:
if support[line_count]:
line.append(row[i])
line_count += 1
ofp.write(sep_out.join(line))
ofp.write("\n")
return
class NumPyArangeEncoder(json.JSONEncoder):
def default(self, obj):
if isinstance(obj, np.int64):
return int(obj)
if isinstance(obj, np.float64):
return float(obj)
if isinstance(obj, np.int32):
return int(obj)
if isinstance(obj, np.float32):
return float(obj)
if isinstance(obj, np.ndarray):
return obj.tolist() # or map(int, obj)
return json.JSONEncoder.default(self, obj)