-
Notifications
You must be signed in to change notification settings - Fork 11
/
NoisyDataSet.py
124 lines (103 loc) · 4.5 KB
/
NoisyDataSet.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
'''''''''''''''''''''''''''''''''''''''''''''''''''''''''''''''''''''''''''''
\file NoisyDataSet.py
\brief NoisyDataSet class.
\copyright Copyright (c) 2019 Visual Computing group of Ulm University,
Germany. See the LICENSE file at the top-level directory of
this distribution.
\author pedro hermosilla (pedro-1.hermosilla-casajus@uni-ulm.de)
'''''''''''''''''''''''''''''''''''''''''''''''''''''''''''''''''''''''''''''
import sys
import os
from os import listdir
from os.path import isfile, join
import math
import numpy as np
class NoisyDataSet:
"""NoisyDataSet.
"""
def __init__(self, dataset, train, seed=None, fileList=None):
"""Constructor.
Args:
"""
self.train_ = train
self.datasetId_ = dataset
# Dataset folder.
if dataset == 3:
self.dataset_ = "RueMadame/"
self.instanceList_ = [f[:-4] for f in listdir(self.dataset_) if isfile(join(self.dataset_, f)) and f.endswith(".txt")]
self.modelList_ = self.instanceList_
else:
if dataset == 0:
self.dataset_ = "NoisyDataSets/NoisyColoredGaussianDataSet/"
elif dataset == 1:
self.dataset_ = "NoisyDataSets/NoisyColoredGaussianDataSet/"
elif dataset == 2:
self.dataset_ = "NoisyDataSets/BlensorColored/"
# Clean/Noisy dataset.
if train:
DSFilePath = "NoisyDataSets/noisy_dataset.txt"
else:
DSFilePath = "NoisyDataSets/test_dataset.txt"
if not(fileList is None):
DSFilePath = fileList
# Model list.
modelTypes = {}
self.modelList_ = []
self.instanceList_ = []
with open(DSFilePath, 'r') as DSFile:
for line in DSFile:
modelName = line.rstrip()
self.instanceList_.append((len(self.modelList_), "005"))
self.instanceList_.append((len(self.modelList_), "01"))
self.instanceList_.append((len(self.modelList_), "015"))
self.modelList_.append(modelName)
# Iterator.
self.randomState_ = np.random.RandomState(seed)
self.iterator_ = 0
self.randList_ = self.randomState_.permutation(self.instanceList_)
# Cache.
self.cache_ = {}
def begin_epoch(self):
self.iterator_ = 0
self.randList_ = self.randomState_.permutation(self.instanceList_)
def next(self):
self.iterator_+=1
def get_num_models(self):
return len(self.modelList_)
def get_num_instances(self):
return len(self.instanceList_)
def end_epoch(self):
return self.iterator_ >= len(self.randList_)
def get_current_model(self, clean=False):
if self.iterator_ < len(self.randList_):
if self.datasetId_ == 3:
currInstance = "0"
currModel = self.randList_[self.iterator_]
if currModel in self.cache_:
points = self.cache_[currModel]
else:
points = np.loadtxt(self.dataset_+currModel+".txt", delimiter=',')
coordMax = np.amax(points, axis=0)
coordMin = np.amin(points, axis=0)
center = (coordMax+coordMin)*0.5
points = (points - center)/5.0
self.cache_[currModel] = points
else:
currModel = self.modelList_[int(self.randList_[self.iterator_][0])]
if not clean:
currInstance = self.randList_[self.iterator_][1]
currModelPath = currModel+"_"+currInstance
if currModelPath in self.cache_:
points = self.cache_[currModelPath]
else:
points = np.loadtxt(self.dataset_+currModelPath+".txt", delimiter=',')
self.cache_[currModelPath] = points
else:
currInstance = ""
if currModel in self.cache_:
points = self.cache_[currModel]
else:
points = np.loadtxt(self.dataset_+currModel+".txt", delimiter=',')
self.cache_[currModel] = points
return points, currModel, currInstance
return -1, "", ""