Data_preprocessing

# install kaggle and download breast histopathology images

!pip install kaggle 

from google.colab import files
files.upload()

!mkdir -p ~/.kaggle
!cp kaggle.json ~/.kaggle/
!chmod 600 ~/.kaggle/kaggle.json


!kaggle datasets download -d paultimothymooney/breast-histopathology-images
!unzip breast-histopathology-images.zip -d breast_cancer_data

# create two folders for 0:IDC_negative and 1:IDC_positive

import os 
import shutil 
from glob import glob

# current directory 
current_dir = '/content/breast_cancer_data'

os.makedirs('/content/breast_cancer_data_sorted/IDC_negative', exist_ok=True)
os.makedirs('/content/breast_cancer_data_sorted/IDC_positive', exist_ok=True)

# relocate labeled images to these two directories 
for filepath in glob(current_dir + '/**/*.png', recursive=True):
  filename = os.path.basename(filepath)
  if filename.endswith('class0.png'):
    shutil.copy(filepath, '/content/breast_cancer_data_sorted/IDC_negative')
  elif filename.endswith('class1.png'):
    shutil.copy(filepath, '/content/breast_cancer_data_sorted/IDC_positive')

# inspect the total number of files 

print('Total number of IDC_negative images: ', len(os.listdir('/content/breast_cancer_data_sorted/IDC_negative')))
print('Total number of IDC_positive images: ', len(os.listdir('/content/breast_cancer_data_sorted/IDC_positive')))

from IPython.display import Image, display
import glob

# Paths to directories
positive_dir = '/content/breast_cancer_data_sorted/IDC_positive'
negative_dir = '/content/breast_cancer_data_sorted/IDC_negative'

#  IDC_positive
positive_images = glob.glob(f"{positive_dir}/*.png")
print("IDC Positive Example:")
display(Image(filename=positive_images[2]))


# IDC_negative
negative_images = glob.glob(f"{negative_dir}/*.png")
print("IDC Negative Example:")
display(Image(filename=negative_images[2]))

# IDC_negative:198738
# IDC_positive: 78786