Introduction

Datasets

Visit link http://deeplearning.net/datasets/ for a list of datasets of different types.

Preprocessing image datasets



In [1]:

    
import numpy as np
import matplotlib.pyplot as plt



In [2]:

    
plt.style.use('ggplot')



In [14]:

    
from libs import utils



In [35]:

    
#files is the list of files in the img_align_celeba directory
files = utils.get_celeb_files()
image = plt.imread(files[50])
plt.imshow(image)
plt.show()



In [17]:

    
def get_celeb_imgs(files):
    return [plt.imread(file) for file in files]



In [43]:

    
imgs = get_celeb_imgs(files[0:100])



In [49]:

    
# Give a tensor of all images
data = np.array(imgs)
print type(data[0, 0, 0, 0])









    



<type 'numpy.uint8'>



In [47]:

    
plt.imshow(np.mean(data, axis = 0).astype(np.uint8))
plt.show()



In [ ]: