{"cells":[{"metadata":{"trusted":true},"cell_type":"code","source":"import tensorflow\n\nimport cv2\nimport glob\nimport os\n\nimport imageio\nimport matplotlib.pyplot as plt\nimport numpy as np\nimport pandas as pd\nimport random\n\nfrom tqdm import tqdm\nfrom tensorflow.keras.models import load_model\nfrom imageio import imread\nfrom skimage.transform import resize as imresize\n\nfrom tensorflow.keras.utils import plot_model\nfrom IPython.display import Image","execution_count":null,"outputs":[]},{"metadata":{"trusted":true},"cell_type":"code","source":"#data_dir ='../input/pneumoniag/PNEUMONIA/*.*'\ndata_dir ='../input/cassava-leaf-disease-classification/train_images/*.*'\n\nos.mkdir('./out_GAN')","execution_count":null,"outputs":[]},{"metadata":{"trusted":true},"cell_type":"code","source":"import tensorflow as tf\nfrom PIL import Image\ndef save_all_images1( generated_image, path):  \n    # save low-resolution, high-resolution(original) and generated high-resolution images into one picture\n    \n    path2=path +'.jpg'\n    #generated_image=generated_image*255.0\n    #generated_image=Image.fromarray(tf.cast(generated_image, tf.uint32).numpy())\n    plt.imsave(path2,generated_image)#[:, :, 0]* 127.5) + 127.5 ,cmap='gray'tf.squeeze() [:, :, 2]*255.0","execution_count":null,"outputs":[]},{"metadata":{"trusted":true},"cell_type":"code","source":"path = \"../input/cassava-leaf-disease-classification/train.csv\"\ndf = pd.read_csv(path)\ndf.head(10)","execution_count":null,"outputs":[]},{"metadata":{"trusted":true},"cell_type":"code","source":"len(df)","execution_count":null,"outputs":[]},{"metadata":{"trusted":true},"cell_type":"code","source":"df['label'].value_counts()","execution_count":null,"outputs":[]},{"metadata":{"trusted":true},"cell_type":"code","source":"img_id = df.image_id[3]\nprint(img_id)\nidx = df.label[3]\nprint(idx)","execution_count":null,"outputs":[]},{"metadata":{"trusted":true},"cell_type":"code","source":"#for_Covid\n#files = glob.glob(pathname= str( '../input/vinbigdata-chest-xray-3ch-jpg/vinbigdata/train'+'/*.*'))\ndata = df\ns = 224\nn = 0\nfor file in tqdm(data.index[0:100]): \n    img_id = data.image_id[file]\n    idx = data.label[file]\n    if idx == 0 :\n        image = cv2.imread('../input/cassava-leaf-disease-classification/train_images/'+img_id)\n        #image = cv2.cvtColor(image, cv2.COLOR_BGR2RGB)\n        image_array = cv2.resize(image , (s,s), interpolation = cv2.INTER_LINEAR)#INTER_LINEAR  INTER_AREA\n        save_all_images1(image_array, path='./out_GAN/CBB_{}'.format(n))\n        n = n + 1","execution_count":null,"outputs":[]},{"metadata":{"trusted":true},"cell_type":"code","source":"import os\nimport zipfile\n \nfantasy_zip = zipfile.ZipFile('./CBB.zip', 'w')\n \nfor folder, subfolders, files in os.walk('./out_GAN'):\n \n    for file in files:\n        if file.endswith('.jpg'):\n            fantasy_zip.write(os.path.join(folder, file), os.path.relpath(os.path.join(folder,file), './out_GAN'), compress_type = zipfile.ZIP_DEFLATED)\n \nfantasy_zip.close()","execution_count":null,"outputs":[]},{"metadata":{"trusted":true},"cell_type":"code","source":"all_images = glob.glob(data_dir)\nlow_resolution_shape_r=(224,224)\n#3836\nlow_resolution_images = []\nfor img in tqdm(all_images[0:21397]):#all_images[0:4300]\n    # take the numpy ndarray from the current image\n    img1 = imread(img, as_gray=False, pilmode='RGB')#False\n    img1 = img1.astype(np.float32)\n    # change the size\n    img1_low_resolution=cv2.resize(img1 , low_resolution_shape_r, interpolation = cv2.INTER_AREA)#INTER_LINEAR, INTER_AREA\n    low_resolution_images.append(img1_low_resolution)\n    \n","execution_count":null,"outputs":[]},{"metadata":{"trusted":true},"cell_type":"code","source":"len(low_resolution_images)","execution_count":null,"outputs":[]},{"metadata":{"trusted":true},"cell_type":"code","source":"low_resolution_images=np.array(low_resolution_images)\n    \nlow_resolution_images = low_resolution_images / 255.","execution_count":null,"outputs":[]},{"metadata":{"trusted":true},"cell_type":"code","source":"from skimage.util import random_noise\nlow_resolution_images = random_noise(low_resolution_images, mode='gaussian', mean=0.0, var=0.15, seed=None, clip=True)","execution_count":null,"outputs":[]},{"metadata":{"trusted":true},"cell_type":"code","source":"len(low_resolution_images)\n#print(low_resolution_images[0])","execution_count":null,"outputs":[]},{"metadata":{"trusted":true},"cell_type":"code","source":"'''\ngenerator1 = load_model('../input/generator-model-1/generator_model.h5')\ngenerated_images1 = generator1.predict_on_batch(low_resolution_images)\n'''","execution_count":null,"outputs":[]},{"metadata":{"trusted":true},"cell_type":"code","source":"encoder = load_model('../input/denoise-model-var-015/encoded.h5')\ndecoder = load_model('../input/denoise-model-var-015/decoder.h5')\nj=0\nn=1\ngenerated=[]\n\nfor i in tqdm(range(1920)):#range(4300)\n    if (i==20) or (i==40) or (i==60) or (i==80) or (i==100) or (i==120) or (i==140)or (i==160)or (i==180)or (i==200)or(i==220) or (i==240)or (i==260)or (i==280)or (i==300)or (i==320) or (i==340)or (i==360)or (i==380)or (i==400)or (i==420) or (i==440)or (i==460)or (i==480)or (i==500)or (i==520) or (i==540)or (i==560)or (i==580)or (i==600)or (i==620) or (i==640)or (i==660)or (i==680)or (i==700)or (i==720) or (i==740)or (i==760)or (i==780)or (i==800)or (i==820) or (i==840)or (i==860)or (i==880)or (i==900)or (i==920) or (i==940)or (i==960)or (i==980)or (i==1000)or (i==1020) or (i==1040)or (i==1060)or (i==1080)or (i==1100)or (i==1120) or (i==1140)or (i==1160) or (i==1180)or (i==1200)or (i==1220) or (i==1240)or (i==1260)or (i==1280)or (i==1300)or (i==1320)or (i==1340)or (i==1360)or (i==1380)or (i==1400)or (i==1420)or (i==1440)or (i==1460)or (i==1480)or (i==1500)or (i==1520)or (i==1540)or (i==1560)or (i==1580)or (i==1600)or (i==1620)or (i==1640)or (i==1660)or (i==1680)or (i==1700)or (i==1720)or (i==1740)or (i==1760)or (i==1780)or (i==1800)or (i==1820)or (i==1840)or (i==1860)or (i==1880)or (i==1900)or (i==1920):#or (i==1940)or (i==1960)or (i==1980)or (i==2000)or (i==2020)or (i==2040)or (i==2060)or (i==2080)or (i==2100)or (i==2120)or (i==2140)or (i==2160)or (i==2180)or (i==2200)or (i==2220)or (i==2240)or (i==2260)or (i==2280)or (i==2300)or (i==2320)or (i==2340)or (i==2360)or (i==2380)or (i==2400)or (i==2420)or (i==2440)or (i==2460)or (i==2480)or (i==2500)or (i==2520)or (i==2540)or (i==2560)or (i==2580)or (i==2600)or (i==2620)or (i==2640)or (i==2660)or (i==2680)or (i==2700)or (i==2720)or (i==2740)or (i==2760)or (i==2780)or (i==2800)or (i==2820)or (i==2840)or (i==2860)or (i==2880)or (i==2900)or (i==2920)or (i==2940)or (i==2960)or (i==2980)or (i==3000)or (i==3020)or (i==3040)or (i==3060)or (i==3080)or (i==3100)or (i==3120)or (i==3140)or (i==3160)or (i==3180)or (i==3200)or (i==3220)or (i==3240)or (i==3260)or (i==3280)or (i==3300)or (i==3320)or (i==3340)or (i==3360)or (i==3380)or (i==3400)or (i==3420)or (i==3440)or (i==3460)or (i==3480)or (i==3500)or (i==3520)or (i==3540)or (i==3560)or (i==3580)or (i==3600)or (i==3620)or (i==3640)or (i==3660)or (i==3680)or (i==3700)or (i==3720)or (i==3740)or (i==3760)or (i==3780)or (i==3800)or (i==3820):\n        generated_images1 = encoder.predict(low_resolution_images[j:i])\n        generated_images1 = decoder.predict(generated_images1)\n        for img in generated_images1:\n            save_all_images1(img, path='./out_GAN/Generated_{}'.format(n))\n            n=n+1\n        j=i\n'''\n\n\ngenerated_images1 = generator1.predict(low_resolution_images)\nsave_all_images1(generated_images1, path='./out_GAN/Generated')\n'''\n            \n        \n#(i==1180)or (i==1200)or (i==1220) or (i==1240)or (i==1260)or (i==1280)or (i==1300)or (i==1320)or (i==1340)or (i==1360)or (i==1380)or (i==1400)or (i==1420)or (i==1440)or (i==1460)or (i==1480)or (i==1500)or (i==1520)or (i==1540)or (i==1560)or (i==1580)or (i==1600)or (i==1620)or (i==1640)or (i==1660)or (i==1680)or (i==1700)or (i==1720)or (i==1740)or (i==1760)or (i==1780)or (i==1800)or (i==1820)or (i==1840)or (i==1860)or (i==1880)or (i==1900)or (i==1920)or (i==1940)or (i==1960)or (i==1980)or (i==2000)or (i==2020)or (i==2040)or (i==2060)or (i==2080)or (i==2100)or (i==2120)or (i==2140)or (i==2160)or (i==2180)or (i==2200)or (i==2220)or (i==2240)or (i==2260)or (i==2280)or (i==2300)or (i==2320)or (i==2340)or (i==2360)or (i==2380)or (i==2400)or (i==2420)or (i==2440)or (i==2460)or (i==2480)or (i==2500)or (i==2520)or (i==2540)or (i==2560)or (i==2580)or (i==2600)or (i==2620)or (i==2640)or (i==2660)or (i==2680)or (i==2700)or (i==2720)or (i==2740)or (i==2760)or (i==2780)or (i==2800)or (i==2820)or (i==2840)or (i==2860)or (i==2880)or (i==2900)or (i==2920)or (i==2940)or (i==2960)or (i==2980)or (i==3000)or (i==3020)or (i==3040)or (i==3060)or (i==3080)or (i==3100)or (i==3120)or (i==3140)or (i==3160)or (i==3180)or (i==3200)or (i==3220)or (i==3240)or (i==3260)or (i==3280)or (i==3300)or (i==3320)or (i==3340)or (i==3360)or (i==3380)or (i==3400)or (i==3420)or (i==3440)or (i==3460)or (i==3480)or (i==3500)or (i==3520)or (i==3540)or (i==3560)or (i==3580)or (i==3600)or (i==3620)or (i==3640)or (i==3660)or (i==3680)or (i==3700)or (i==3720)or (i==3740)or (i==3760)or (i==3780)or (i==3800)or (i==3820)or (i==3840)or (i==3860)or (i==3880)or (i==3900)or (i==3920)or (i==3940)or (i==3960)or (i==3980)or (i==4000)or (i==4020)or (i==4040)or (i==4060)or (i==4080)or (i==4100)or (i==4120)or (i==4140)or (i==4160)or (i==4180)or (i==4200)or (i==4220)or (i==4240):\n        ","execution_count":null,"outputs":[]},{"metadata":{"trusted":true},"cell_type":"code","source":"len(generated_images1)","execution_count":null,"outputs":[]},{"metadata":{"trusted":true},"cell_type":"code","source":"import os\nimport zipfile\n \nfantasy_zip = zipfile.ZipFile('./X_Ray_Normal_part_2.zip', 'w')\n \nfor folder, subfolders, files in os.walk('./out_GAN'):\n \n    for file in files:\n        if file.endswith('.jpg'):\n            fantasy_zip.write(os.path.join(folder, file), os.path.relpath(os.path.join(folder,file), './out_GAN'), compress_type = zipfile.ZIP_DEFLATED)\n \nfantasy_zip.close()","execution_count":null,"outputs":[]},{"metadata":{"trusted":true},"cell_type":"code","source":"","execution_count":null,"outputs":[]}],"metadata":{"kernelspec":{"language":"python","display_name":"Python 3","name":"python3"},"language_info":{"pygments_lexer":"ipython3","nbconvert_exporter":"python","version":"3.6.4","file_extension":".py","codemirror_mode":{"name":"ipython","version":3},"name":"python","mimetype":"text/x-python"}},"nbformat":4,"nbformat_minor":4}