index = 0 partitionOfData = 0.85 for label in data: currentTargets = labels[index] sizeOfLabel = np.shape(label)[0] trainingPortion = int(np.ceil(sizeOfLabel*partitionOfData)) testPortion = int(np.ceil(sizeOfLabel*partitionOfData+1)) if len(Xtraining) == 0: Xtraining = label[0:trainingPortion] Ytraining = currentTargets[0:trainingPortion,:] else: Xtraining = np.concatenate((Xtraining,label[0:trainingPortion]),axis=0) Ytraining = np.concatenate((Ytraining,currentTargets[0:trainingPortion,:]),axis=0) index += 1 gaussiansPyramids = dataAugmentation.getGaussianPyramidOfList(Xtraining,3) Xtraining1 = gaussiansPyramids[0]; Xtraining2 = gaussiansPyramids[1]; Xtraining3 = gaussiansPyramids[2]; Xtraining = np.expand_dims(Xtraining,1) Xtraining = Xtraining.astype("float32") Xtraining1 = np.expand_dims(Xtraining1,1) Xtraining1 = Xtraining1.astype("float32") Xtraining2 = np.expand_dims(Xtraining2,1) Xtraining2 = Xtraining2.astype("float32") Xtraining3 = np.expand_dims(Xtraining3,1) Xtraining3 = Xtraining3.astype("float32") ###############START GENERATING TEST###################### laplaceImages = 0 data = list()
import numpy as np from skimage import data, io, filters, transform import sklearn import scipy as sc from PIL import Image as imLib1 from dataAugmentation import getGaussianPyramidOfList setOfImages = list() imageTest = io.imread('./13NaturalSceneDataset/bedroom/image_0216.jpg') imageTest = transform.resize(imageTest,(200,200)) setOfImages.append(imageTest) imageTest = io.imread('./13NaturalSceneDataset/bedroom/image_0200.jpg') imageTest = transform.resize(imageTest,(200,200)) setOfImages.append(imageTest) test = getGaussianPyramidOfList(setOfImages,3)