# %% [code] import os, sys import numpy as np import pandas as pd import matplotlib.pyplot as plt import skimage.io from skimage.transform import resize from imgaug import augmenters as iaa from tqdm import tqdm import PIL from PIL import Image, ImageOps import cv2 from sklearn.utils import class_weight, shuffle from keras.losses import binary_crossentropy from keras.applications.resnet50 import preprocess_input import keras.backend as K import tensorflow as tf from sklearn.metrics import f1_score, fbeta_score from keras.utils import Sequence from keras.utils import to_categorical from sklearn.model_selection import train_test_split from sklearn.utils import class_weight import keras from keras.models import Sequential from keras.layers import Dense, Dropout, Flatten from keras.layers import Conv2D, MaxPooling2D from keras.optimizers import SGD WORKERS = 2 CHANNEL = 3 import warnings warnings.filterwarnings("ignore") IMG_SIZE = 512 NUM_CLASSES = 5 SEED = 77 TRAIN_NUM = 1000 # use 1000 when you just want to explore new idea, use -1 for full train # %% [code] df_train = pd.read_csv('../input/aptos2019-blindness-detection/train.csv') df_test = pd.read_csv('../input/aptos2019-blindness-detection/test.csv') #print(df_train) print(df_train.head(10)) print(df_train.describe) print(df_train.shape) x = df_train['id_code'] y = df_train['diagnosis'] x, y = shuffle(x, y, random_state=SEED) print("x data after shuffling") print(x) print("y data after shuffling") print(y) #convert to numpy x = x.values y = y.values # %% [code] x_train, valid_x, y_train, valid_y = train_test_split(x, y, test_size=0.15, stratify=y, random_state=SEED) print(x_train.shape, y_train.shape, valid_x.shape, valid_y.shape) #train_y.hist() #valid_y.hist() # %% [markdown] # # 2. Try Ben Graham's preprocessing method. # %% [code] # %% [code] def crop_image1(img,tol=7): # img is image data # tol is tolerance mask = img>tol return img[np.ix_(mask.any(1),mask.any(0))] def crop_image_from_gray(img,tol=7): if img.ndim ==2: mask = img>tol return img[np.ix_(mask.any(1),mask.any(0))] elif img.ndim==3: gray_img = cv2.cvtColor(img, cv2.COLOR_RGB2GRAY) mask = gray_img>tol check_shape = img[:,:,0][np.ix_(mask.any(1),mask.any(0))].shape[0] if (check_shape == 0): # image is too dark so that we crop out everything, return img # return original image else: img1=img[:,:,0][np.ix_(mask.any(1),mask.any(0))] img2=img[:,:,1][np.ix_(mask.any(1),mask.any(0))] img3=img[:,:,2][np.ix_(mask.any(1),mask.any(0))] # print(img1.shape,img2.shape,img3.shape) img = np.stack([img1,img2,img3],axis=-1) # print(img.shape) return img # %% [code] # OLD version of image color cropping, use crop_image_from_gray instead # The above code work only for 1-channel. Here is my simple extension for 3-channels image def crop_image(img,tol=7): if img.ndim ==2: mask = img>tol return img[np.ix_(mask.any(1),mask.any(0))] elif img.ndim==3: h,w,_=img.shape # print(h,w) img1=cv2.resize(crop_image1(img[:,:,0]),(w,h)) img2=cv2.resize(crop_image1(img[:,:,1]),(w,h)) img3=cv2.resize(crop_image1(img[:,:,2]),(w,h)) # print(img1.shape,img2.shape,img3.shape) img[:,:,0]=img1 img[:,:,1]=img2 img[:,:,2]=img3 return img # %% [code] def load_ben_color(path, sigmaX=10): image = cv2.imread(path) image = cv2.cvtColor(image, cv2.COLOR_BGR2RGB) image = crop_image_from_gray(image) image = cv2.resize(image, (IMG_SIZE, IMG_SIZE)) image=cv2.addWeighted ( image,4, cv2.GaussianBlur( image , (0,0) , sigmaX) ,-4 ,128) return image # %% [code] # %% [code] # ## try circle crop def circle_crop(img, sigmaX=10): """ Create circular crop around image centre """ img = cv2.imread(img) img = crop_image_from_gray(img) img = cv2.cvtColor(img, cv2.COLOR_BGR2RGB) height, width, depth = img.shape x = int(width/2) y = int(height/2) r = np.amin((x,y)) circle_img = np.zeros((height, width), np.uint8) cv2.circle(circle_img, (x,y), int(r), 1, thickness=-1) img = cv2.bitwise_and(img, img, mask=circle_img) img = crop_image_from_gray(img) img=cv2.addWeighted ( img,4, cv2.GaussianBlur( img , (0,0) , sigmaX) ,-4 ,128) return img # %% [code] """ NUM_SAMP=7 fig = plt.figure(figsize=(25, 16)) for class_id in sorted(train_y.unique()): for i, (idx, row) in enumerate(df_train.loc[df_train['diagnosis'] == class_id].sample(NUM_SAMP, random_state=SEED).iterrows()): ax = fig.add_subplot(5, NUM_SAMP, class_id * NUM_SAMP + i + 1, xticks=[], yticks=[]) path=f"../input/aptos2019-blindness-detection/train_images/{row['id_code']}.png" image = load_ben_color(path,sigmaX=30) plt.imshow(image) ax.set_title('%d-%d-%s' % (class_id, idx, row['id_code']) ) """ # %% [code] dpi = 80 #inch # path=f"../input/aptos2019-blindness-detection/train_images/5c7ab966a3ee.png" # notice upper part #path=f"../input/aptos2019-blindness-detection/train_images/db3cd58aa315.png" path=f"../input/aptos2019-blindness-detection/train_images/92587e494d51.png" # lower-right, can be class3 image = circle_crop(path,sigmaX=20) #image = cv2.cvtColor(image, cv2.COLOR_RGB2GRAY) height, width = IMG_SIZE, IMG_SIZE print(height, width) SCALE=1 figsize = (width / float(dpi))/SCALE, (height / float(dpi))/SCALE fig = plt.figure(figsize=figsize) plt.imshow(image, cmap='gray') # %% [code] DATADIR=f"../input/aptos2019-blindness-detection/train_images" #convert data and fill into array train = [] i=0 for tr in x_train: path = os.path.join(DATADIR,tr + '.png') image = circle_crop(path,sigmaX=20) image = cv2.resize(image, (IMG_SIZE, IMG_SIZE)) #resize image to 512x512 #image = cv2.cvtColor(image, cv2.COLOR_RGB2GRAY) #convert to grayscale i += 1 #counter variable print(i) train.append(image) # %% [code] valid = [] i=0 for vld in valid_x: path = os.path.join(DATADIR,vld + '.png') image = circle_crop(path,sigmaX=20) image = cv2.resize(image, (IMG_SIZE, IMG_SIZE)) #resize image to 512x512 #image = cv2.cvtColor(image, cv2.COLOR_RGB2GRAY) #convert to grayscale i += 1 #counter variable print(i) valid.append(image) #convert array to numpy x_train = np.array(train) valid_x = np.array(valid) #x_train = np.array(x_train).reshape(-1, 512, 512, 1) # add channel if grayscale np.save('x_train', x_train) np.save('y_train', y_train) np.save('valid_x', valid_x) np.save('valid_y', valid_y) """ save the numpy array to a file np.save('x_train', x_train) to load the dataset x_train = np.load('x_train.npy') """ #set class weights for imbalanced data class_weights = class_weight.compute_class_weight('balanced', np.unique(y_train), y_train) class_weights = dict(enumerate(class_weights)) # %% [code] model = Sequential() # input shape 512x512 images with 1 channel. model.add(Conv2D(32, (3, 3), activation='relu', input_shape=x_train.shape[1:])) model.add(MaxPooling2D(pool_size=(2, 2))) model.add(Dropout(0.25)) model.add(Conv2D(64, (3, 3), activation='relu')) model.add(MaxPooling2D(pool_size=(2, 2))) model.add(Dropout(0.25)) model.add(Flatten()) model.add(Dense(256, activation='relu')) model.add(Dropout(0.5)) model.add(Dense(5, activation='softmax')) #sgd = SGD(lr=0.01, decay=1e-6, momentum=0.9, nesterov=True) model.compile( optimizer='adam', loss='sparse_categorical_crossentropy', metrics=['sparse_categorical_accuracy']) model.fit(x_train, y_train, batch_size=32, epochs=10, class_weight=class_weights) #evaluation = model.evaluate(valid_x, valid_y, batch_size=32) #loss, accuracy # %% [code]