I want to randomly select 50 images from the Python 3 folder.

I am writing a code that reads 50 pages each from a folder containing 100 images, converts them into a numpy array, and saves the npy file twice.
After the code was executed, two npy files were saved properly, but when I looked inside, it seemed that each of the npy files was saved 100 sheets.
I tried to randomly select 50 pages from within the folder in rn.sample, but I wonder if the writing method is wrong.
I'm sorry to bother you at such a busy time, but I'd appreciate it if you could give me some advice on how to fix it.

*I am also asking questions on teratail.As there was no response from you, I posted it here as well.I will share it with you as soon as I receive your reply.

import random as rn 

img_size= (1000,500)
dir_name = './train'
file_type = 'jpeg'

img_list=glob.glob('./'+dir_name+'/*.'+file_type)

for i in range(2): 
  train_img_array_list = [ ]
  Every time rn.sample(img_list,50)#i increases by 1, it reads 50 randomly selected images.
  for img in img_list:
    train_img=load_img(img,grayscale=True,target_size=(img_size)) 
    train_img_array=img_to_array(train_img)/255
    train_img_array_list.append(train_img_array)

  train_img_array_list=np.array(train_img_array_list)
  file_name = os.path.join('.', 'image' + '{0:04d}'.format(i) + '.npy') # Save npy file with serial number
  np.save(file_name, train_img_array_list)

python python3

2022-09-30 21:40

2 Answers

I received advice from teratail and corrected it as follows (img_list=rn.sample(img_list,50).I am very sorry for confusing you because I did not mention that the module had been imported as import random as rn when I asked.Thank you to everyone who took the time to review and make suggestions.

import random as rn

img_size= (1000,500)
dir_name = './train' 
file_type = 'jpeg'

img_list=glob.glob('./'+dir_name+'/*.'+file_type)

for i in range(2): 
    train_img_array_list = [ ]
    Img_list=rn.sample(img_list,50)#i reads 50 randomly selected images each time it increases by 1
    for img in img_list:
      train_img=load_img(img,grayscale=True,target_size=(img_size)) 
      train_img_array=img_to_array(train_img)/255
      train_img_array_list.append(train_img_array)

    train_img_array_list=np.array(train_img_array_list)
    file_name = os.path.join('.', 'image' + '{0:04d}'.format(i) + '.npy')
    np.save(file_name, train_img_array_list)

2022-09-30 21:40

I don't know how rn.sample works, so I don't think everyone can answer.

In addition, is the following helpful?

train_size=x_train.shape[0]
batch_size=10#10 randomly
batch_mask=np.random.choice(train_size, batch_size)
x_batch=x_train [batch_mask]
t_batch=t_train [batch_mask]

Source:
"Deep Learning from scratch -- Theory and Implementation of Deep Learning in Python"
(First Published July 28, 2017, O'Reilly Japan, Inc.)

2022-09-30 21:40

If you have any answers or tips

Popular Tags

python x 4647

android x 1593

java x 1494

javascript x 1427

c x 927

c++ x 878

ruby-on-rails x 696

php x 692

python3 x 685

html x 656