53. glob-clothes || 데이터셋만들기

import numpy as np
import pandas as pd
import tensorflow as tf
import glob as glob
all_data = np.array(glob.glob('./clothes/*/*.jpg', recursive=True))
'''recursive = 폴더 아래 폴더또 있을 때, 지정한 하위폴더까지 검색설정
glob.glob : ./clothes/*/*.jpg
.:현재폴더
clothes : 현재폴더하위의 cloths 폴더
* 하위폴더
* .jpg : 모든 jpg
array로 만들어서 저장
'''

 

print(all_data[:5])

['./clothes\\black_dress\\1.jpg' './clothes\\black_dress\\10.jpg'
 './clothes\\black_dress\\100.jpg' './clothes\\black_dress\\101.jpg'
 './clothes\\black_dress\\102.jpg']

 

def check_cc(color, clothes) :
    labels = np.zeros(11,)
    # color check
    if ( color == 'black') :
        labels[0] = 1
        color_index = 0
    elif ( color == 'blue') :
        labels[1] = 1
        color_index = 1
    elif ( color == 'brown') :
        labels[2] = 1
        color_index = 2
    elif ( color == 'green') :
        labels[3] = 1
        color_index = 3
    elif ( color == 'red') :
        labels[4] = 1
        color_index = 4
    elif ( color == 'white') :
        labels[5] = 1
        color_index = 5
    # clothes check
        # color check
    if ( color == 'dress') :
        labels[6] = 1
    elif ( color == 'shirt') :
        labels[7] = 1
    elif ( color == 'pants') :
        labels[8] = 1
    elif ( color == 'shorts') :
        labels[9] = 1
    elif ( color == 'shoes') :
        labels[10] = 1
    return labels, color_index

 

print(all_data.shape)
all_labels = np.empty((all_data.shape[0], 11))
all_color_labels = np.empty((all_data.shape[0], 1))

print(all_labels.shape)
print(all_color_labels.shape)

(16170,)
(16170, 11)
(16170, 1)

 

for i, data in enumerate(all_data) :
    color_and_clothes = all_data[i].split('\\')[1].split('_')
    color = color_and_clothes[0]
    print(color_and_clothes)
    clothes = color_and_clothes[1]
    labels, color_index = check_cc(color, clothes)
    all_labels[i] = labels;
    all_color_labels[i] = color_index
print(all_labels[:10])


['black', 'dress']
['black', 'dress']
['black', 'dress']
['black', 'dress']
['black', 'dress']

 

# 훈련, 테스트, 검증데이터 분리
from sklearn.model_selection import train_test_split
train_x, test_x, train_y, test_y = train_test_split\
(all_data, all_labels, shuffle = True, test_size = 0.3, random_state = 99)
train_x, val_x, train_y, val_y = train_test_split\
(train_x, train_y, shuffle = True, test_size = 0.3, random_state = 99)
print(train_x.shape)
print(val_x.shape)
print(test_x.shape)

(7923,)
(3396,)
(4851,)

 

train_df = pd.DataFrame(
    {'image':train_x, 'black':train_y[:,0], 'blue':train_y[:,1],
    'brown':train_y[:,2], 'green':train_y[:,3], 'red':train_y[:,4],
     'white':train_y[:,5], 'dress':train_y[:,6], 'shirt':train_y[:,7],
     'pants':train_y[:,8], 'shorts':train_y[:,9], 'shoes':train_y[:,10]})
val_df = pd.DataFrame(
    {'image':val_x, 'black':val_y[:,0], 'blue':val_y[:,1],
    'brown':val_y[:,2], 'green':val_y[:,3], 'red':val_y[:,4],
     'white':val_y[:,5], 'dress':val_y[:,6], 'shirt':val_y[:,7],
     'pants':val_y[:,8], 'shorts':val_y[:,9], 'shoes':val_y[:,10]})
test_df = pd.DataFrame(
    {'image':test_x, 'black':test_y[:,0], 'blue':test_y[:,1],
    'brown':test_y[:,2], 'green':test_y[:,3], 'red':test_y[:,4],
     'white':test_y[:,5], 'dress':test_y[:,6], 'shirt':test_y[:,7],
     'pants':test_y[:,8], 'shorts':test_y[:,9], 'shoes':test_y[:,10]})

 

train_df.head()

	image	black	blue	brown	green	red	white	dress	shirt	pants	shorts	shoes
0	./clothes\silver_skirt\000095.jpg	0.0	0.0	0.0	0.0	0.0	0.0	0.0	0.0	0.0	0.0	0.0
1	./clothes\blue_shirt\82.jpg	0.0	1.0	0.0	0.0	0.0	0.0	0.0	0.0	0.0	0.0	0.0
2	./clothes\pink_hoodie\000267.jpg	0.0	0.0	0.0	0.0	0.0	0.0	0.0	0.0	0.0	0.0	0.0
3	./clothes\red_shoes\159.jpg	0.0	0.0	0.0	0.0	0.0	0.0	0.0	0.0	0.0	0.0	0.0
4	./clothes\red_shoes\601.jpg	0.0	0.0	0.0	0.0	0.0	0.0	0.0	0.0	0.0	0.0	0.0

 

train_df.to_csv('./clothes.train.csv', index = None)
val_df.to_csv('./clothes.val.csv', index = None)
test_df.to_csv('./clothes.test.csv', index = None)

 

df1 = pd.read_csv("./clothes.train.csv")
df2 = pd.read_csv("./clothes.val.csv")
df3 = pd.read_csv("./clothes.test.csv")

 

df1.info()

<class 'pandas.core.frame.DataFrame'>
RangeIndex: 7923 entries, 0 to 7922
Data columns (total 12 columns):
 #   Column  Non-Null Count  Dtype  
---  ------  --------------  -----  
 0   image   7923 non-null   object 
 1   black   7923 non-null   float64
 2   blue    7923 non-null   float64
 3   brown   7923 non-null   float64
 4   green   7923 non-null   float64
 5   red     7923 non-null   float64
 6   white   7923 non-null   float64
 7   dress   7923 non-null   float64
 8   shirt   7923 non-null   float64
 9   pants   7923 non-null   float64
 10  shorts  7923 non-null   float64
 11  shoes   7923 non-null   float64
dtypes: float64(11), object(1)
memory usage: 742.9+ KB