-
Notifications
You must be signed in to change notification settings - Fork 3
/
Copy pathtrain_and_test_split_2.py
184 lines (151 loc) · 6.45 KB
/
train_and_test_split_2.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
import os
import random
from shutil import copyfile
import pandas as pd
import numpy as np
import csv
file_name = 'shuffled_datafile.csv'
csv_file = open('shuffled_datafile.csv','r')
csv_file.readline()
img_source_dir = './images'
train_size = 0.8
# def img_train_test_split(img_source_dir, df, train_size):
if not (isinstance(img_source_dir, str)):
raise AttributeError('img_source_dir must be a string')
if not os.path.exists(img_source_dir):
raise OSError('img_source_dir does not exist')
if not (isinstance(train_size, float)):
raise AttributeError('train_size must be a float')
# Set up empty folder structure if not exists
if not os.path.exists('data1'):
os.makedirs('data1')
else:
if not os.path.exists('data1/train'):
os.makedirs('data1/train')
# if not os.path.exists('data1/train/broken'):
# os.makedirs('data1/train/broken')
# if not os.path.exists('data1/train/silkcut'):
# os.makedirs('data1/train/silkcut')
# if not os.path.exists('data1/train/pure'):
# os.makedirs('data1/train/pure')
# if not os.path.exists('data1/train/discolored'):
# os.makedirs('data1/train/discolored')
if not os.path.exists('data1/validation'):
os.makedirs('data1/validation')
# if not os.path.exists('data1/validation/broken'):
# os.makedirs('data1/validation/broken')
# if not os.path.exists('data1/validation/silkcut'):
# os.makedirs('data1/validation/silkcut')
# if not os.path.exists('data1/validation/pure'):
# os.makedirs('data1/validation/pure')
# if not os.path.exists('data1/validation/discolored'):
# os.makedirs('data1/validation/discolored')
train_subdir = os.path.join('data1/train')
validation_subdir = os.path.join('data1/validation')
# Create subdirectories in train and validation folders
if not os.path.exists(train_subdir):
os.makedirs(train_subdir)
if not os.path.exists(validation_subdir):
os.makedirs(validation_subdir)
train_counter = 0
validation_counter = 0
df_test= pd.DataFrame(columns=['name', 'class'])
df_train = pd.DataFrame(columns=['name', 'class'])
i_test = 0
i_train = 0
# Randomly assign an image to train or validation folder
for filename, label in csv.reader(csv_file, delimiter=','):
# print(filename)
if filename.endswith(".jpg") or filename.endswith(".png"):
fileparts = filename.split('.')
if random.uniform(0, 1) <= train_size:
if label=='0':
copyfile(os.path.join(img_source_dir, filename), os.path.join(train_subdir,'discolored', filename ))
train_counter += 1
df_train.loc[i_train] = filename , label
i_train += 1
if label=='3':
copyfile(os.path.join(img_source_dir, filename), os.path.join(train_subdir,'silkcut', filename ))
train_counter += 1
df_train.loc[i_train] = filename , label
i_train += 1
if label=='2':
copyfile(os.path.join(img_source_dir, filename), os.path.join(train_subdir,'broken', filename ))
train_counter += 1
df_train.loc[i_train] = filename , label
i_train += 1
if label=='1':
copyfile(os.path.join(img_source_dir, filename), os.path.join(train_subdir,'pure', filename ))
train_counter += 1
df_train.loc[i_train] = filename , label
i_train += 1
else:
if label=='0':
copyfile(os.path.join(img_source_dir, filename), os.path.join(validation_subdir,'discolored', filename))
validation_counter += 1
df_test.loc[i_test] = filename , label
i_test += 1
if label=='3':
copyfile(os.path.join(img_source_dir, filename), os.path.join(validation_subdir,'silkcut', filename))
validation_counter += 1
df_test.loc[i_test] = filename , label
i_test += 1
if label=='2':
copyfile(os.path.join(img_source_dir, filename), os.path.join(validation_subdir,'broken', filename))
validation_counter += 1
df_test.loc[i_test] = filename , label
i_test += 1
if label=='1':
copyfile(os.path.join(img_source_dir, filename), os.path.join(validation_subdir,'pure', filename))
validation_counter += 1
df_test.loc[i_test] = filename , label
i_test += 1
df_test.to_csv('test_data_file.csv')
df_train.to_csv('train_data_file.csv')
print('Copied ' + str(train_counter) + ' images to data/train/' )
print('Copied ' + str(validation_counter) + ' images to data/validation/')
# data_info = pd.read_csv(file_name)
# df = pd.DataFrame.from_csv(file_name)
# data_info = data_info.to_list()
# print(data_info.shape)
# print(data_info.size)
# random.shuffle(data_info)
# print(data_info)
# img_names = pd.read_csv(file_name)
# labels = np.asarray(data_info.iloc[:, 1])
# img_names = np.asarray(data_info.iloc[:, 0])
# test = img_train_test_split(img_source_dir, df, train_size)
# seeds_dataset_labels_file.csv
# titanic_data.head()
# city = pd.DataFrame([['Sacramento', 'California'], ['Miami', 'Florida']], columns=['City', 'State'])
# city.to_csv('city.csv')
## Code to generate DataFrame:
# name_dict = {
# 'Name': ['a','b','c','d'],
# 'Score': [90,80,95,20]
# }
# df = pd.DataFrame(name_dict)
# list of name, degree, score
# name = ["aparna", "pankaj", "sudhir", "Geeku"]
# label = [90, 40, 80, 98]
# dictionary of lists
# dict = {'name': name, 'label': label}
# df = pd.DataFrame(dict)
# saving the dataframe
# df.to_csv('train_.csv')
# random.shuffle(number_list)
# A continuous index value will be maintained
# across the rows in the new appended data frame.
# df1.append(df2, ignore_index = True)
# df = pd.DataFrame(columns=['name', 'class'])
# df = pd.DataFrame(columns=['lib', 'qty1', 'qty2'])
# >>> for i in range(5):
# >>> df.loc[i] = ['name' + str(i)] + list(randint(10, size=2))
# >>> df
# lib qty1 qty2
# 0 name0 3 3
# 1 name1 2 4
# 2 name2 2 8
# 3 name3 2 1
# 4 name4 9 6
# copyfile(os.path.join(img_source_dir, filename), os.path.join(validation_subdir, filename + '.' + fileparts[1]))