Mercurial > hg > nsaunier > traffic-intelligence

comparison scripts/train-object-classification.py @ 961:ec1682ed999f

Find changesets by keywords (author, files, the commit message), revision number or hash, or revset expression.
added computation of confusion matrix and improved default parameter for block normalization for SVM classification
author Nicolas Saunier <nicolas.saunier@polymtl.ca>
date Sun, 05 Nov 2017 23:45:47 -0500
parents 238008f81c16
children 2757efeabbb4
comparison
equal deleted inserted replaced
960:0c1d1eeef544 961:ec1682ed999f
16 parser.add_argument('--coef0', dest = 'coef0', help = 'SVM coef0', default = 0, type = int) 16 parser.add_argument('--coef0', dest = 'coef0', help = 'SVM coef0', default = 0, type = int)
17 parser.add_argument('--cvalue', dest = 'cvalue', help = 'SVM Cvalue', default = 1, type = int) 17 parser.add_argument('--cvalue', dest = 'cvalue', help = 'SVM Cvalue', default = 1, type = int)
18 parser.add_argument('--nu', dest = 'nu', help = 'SVM nu', default = 0, type = int) 18 parser.add_argument('--nu', dest = 'nu', help = 'SVM nu', default = 0, type = int)
19 parser.add_argument('--svmp', dest = 'svmP', help = 'SVM p', default = 0, type = int) 19 parser.add_argument('--svmp', dest = 'svmP', help = 'SVM p', default = 0, type = int)
20 parser.add_argument('--cfg', dest = 'configFilename', help = 'name of the classifier configuration file', required = True) 20 parser.add_argument('--cfg', dest = 'configFilename', help = 'name of the classifier configuration file', required = True)
21 # parser.add_argument('-s', dest = 'rescaleSize', help = 'rescale size of image samples', default = 64, type = int) 21 parser.add_argument('--compute-classifications', dest = 'computeClassifications', help = 'compute the confusion matrix on the training data', action = 'store_true')
22 # parser.add_argument('-o', dest = 'nOrientations', help = 'number of orientations in HoG', default = 9, type = int)
23 # parser.add_argument('-p', dest = 'nPixelsPerCell', help = 'number of pixels per cell', default = 8, type = int)
24 # parser.add_argument('-c', dest = 'nCellsPerBlock', help = 'number of cells per block', default = 2, type = int)
25 22
26 args = parser.parse_args() 23 args = parser.parse_args()
27 classifierParams = storage.ClassifierParameters(args.configFilename) 24 classifierParams = storage.ClassifierParameters(args.configFilename)
28
29 # rescaleSize = (args.rescaleSize, args.rescaleSize)
30 # nPixelsPerCell = (args.nPixelsPerCell, args.nPixelsPerCell)
31 # nCellsPerBlock = (args.nCellsPerBlock, args.nCellsPerBlock)
32 25
33 imageDirectories = {'pedestrian': args.directoryName + "/Pedestrians/", 26 imageDirectories = {'pedestrian': args.directoryName + "/Pedestrians/",
34 'bicycle': args.directoryName + "/Cyclists/", 27 'bicycle': args.directoryName + "/Cyclists/",
35 'car': args.directoryName + "/Vehicles/"} 28 'car': args.directoryName + "/Vehicles/"}
36 29
55 trainingSamplesPV[k], trainingLabelsPV[k] = trainingSamples, trainingLabels 48 trainingSamplesPV[k], trainingLabelsPV[k] = trainingSamples, trainingLabels
56 49
57 # Training the Support Vector Machine 50 # Training the Support Vector Machine
58 print "Training Pedestrian-Cyclist-Vehicle Model" 51 print "Training Pedestrian-Cyclist-Vehicle Model"
59 model = ml.SVM(args.svmType, args.kernelType, args.degree, args.gamma, args.coef0, args.cvalue, args.nu, args.svmP) 52 model = ml.SVM(args.svmType, args.kernelType, args.degree, args.gamma, args.coef0, args.cvalue, args.nu, args.svmP)
60 model.train(np.concatenate(trainingSamplesPBV.values()), np.concatenate(trainingLabelsPBV.values())) 53 classifications = model.train(np.concatenate(trainingSamplesPBV.values()), np.concatenate(trainingLabelsPBV.values()), True)
54 if args.computeClassifications:
55 print(classifications)
61 model.save(args.directoryName + "/modelPBV.xml") 56 model.save(args.directoryName + "/modelPBV.xml")
62 57
63 print "Training Cyclist-Vehicle Model" 58 print "Training Cyclist-Vehicle Model"
64 model = ml.SVM(args.svmType, args.kernelType, args.degree, args.gamma, args.coef0, args.cvalue, args.nu, args.svmP) 59 model = ml.SVM(args.svmType, args.kernelType, args.degree, args.gamma, args.coef0, args.cvalue, args.nu, args.svmP)
65 model.train(np.concatenate(trainingSamplesBV.values()), np.concatenate(trainingLabelsBV.values())) 60 classifications = model.train(np.concatenate(trainingSamplesBV.values()), np.concatenate(trainingLabelsBV.values()), True)
61 if args.computeClassifications:
62 print(classifications)
66 model.save(args.directoryName + "/modelBV.xml") 63 model.save(args.directoryName + "/modelBV.xml")
67 64
68 print "Training Pedestrian-Cyclist Model" 65 print "Training Pedestrian-Cyclist Model"
69 model = ml.SVM(args.svmType, args.kernelType, args.degree, args.gamma, args.coef0, args.cvalue, args.nu, args.svmP) 66 model = ml.SVM(args.svmType, args.kernelType, args.degree, args.gamma, args.coef0, args.cvalue, args.nu, args.svmP)
70 model.train(np.concatenate(trainingSamplesPB.values()), np.concatenate(trainingLabelsPB.values())) 67 classifications = model.train(np.concatenate(trainingSamplesPB.values()), np.concatenate(trainingLabelsPB.values()), True)
68 if args.computeClassifications:
69 print(classifications)
71 model.save(args.directoryName + "/modelPB.xml") 70 model.save(args.directoryName + "/modelPB.xml")
72 71
73 print "Training Pedestrian-Vehicle Model" 72 print "Training Pedestrian-Vehicle Model"
74 model = ml.SVM(args.svmType, args.kernelType, args.degree, args.gamma, args.coef0, args.cvalue, args.nu, args.svmP) 73 model = ml.SVM(args.svmType, args.kernelType, args.degree, args.gamma, args.coef0, args.cvalue, args.nu, args.svmP)
75 model.train(np.concatenate(trainingSamplesPV.values()), np.concatenate(trainingLabelsPV.values())) 74 classifications = model.train(np.concatenate(trainingSamplesPV.values()), np.concatenate(trainingLabelsPV.values()), True)
75 if args.computeClassifications:
76 print(classifications)
76 model.save(args.directoryName + "/modelPV.xml") 77 model.save(args.directoryName + "/modelPV.xml")