startTime=600, stopTime=6000, label=6) a.readDataSet(equalLength=False, checkData=False) useDump = False if useDump: a.loadDumpNormParam(dumpName="MLPClassifier") clf = a.loadDumpClassifier("MLPClassifier") a.testClassifier(classifier=clf) a.setFileSink(fileSinkName="chris", fileSinkPath="../") a.startLiveClassification() else: a.initFeatNormalization(dumpName="MLPClassifier") from sklearn.neural_network import MLPClassifier clf = MLPClassifier() a.trainClassifier(classifier=clf) a.dumpClassifier(dumpName="MLPClassifier") a.testClassifier() windowedData, windowLabels = a.windowSplitSourceDataTT() index = np.linspace(0, len(windowedData) - 1, len(windowedData), dtype=int) random.shuffle(index) trainingData = [] trainingLabels = [] testData = [] testLabels = []
startTime=600, stopTime=6000, label=6) a.readDataSet(equalLength=False, checkData=False) useDump = False if useDump: a.loadDumpNormParam(dumpName="dataOnly") clf = a.loadDumpClassifier("dataOnly") a.testClassifier(classifier=clf) a.setFileSink(fileSinkName="chris", fileSinkPath="../") a.startLiveClassification() else: a.initFeatNormalization(dumpName="dataOnly") from sklearn import svm clf = svm.SVC(kernel='rbf') a.trainClassifier(classifier=clf) a.dumpClassifier(dumpName="dataOnly") a.testClassifier() windowedData, windowLabels = a.windowSplitSourceDataTT() index = np.linspace(0, len(windowedData) - 1, len(windowedData), dtype=int) random.shuffle(index) trainingData = [] trainingLabels = [] testData = [] testLabels = []
label=5) a.addDataFiles(fileSourceName="igor.txt", fileSourcePath="../", startTime=100, stopTime=2900, label=6) a.addDataFiles(fileSourceName="igor2.txt", fileSourcePath="../", startTime=600, stopTime=6000, label=6) a.readDataSet(equalLength=False, checkData=False) useDump = False if useDump: a.loadDumpNormParam(dumpName="KNeighborsClassifier") clf = a.loadDumpClassifier("KNeighborsClassifier") a.testClassifier(classifier=clf) a.setFileSink(fileSinkName="chris", fileSinkPath="../") a.startLiveClassification() else: a.initFeatNormalization(dumpName="KNeighborsClassifier") from sklearn.neighbors import KNeighborsClassifier clf = KNeighborsClassifier(n_neighbors=4, metric='euclidean') a.trainClassifier(classifier=clf) a.dumpClassifier(dumpName="KNeighborsClassifier") a.testClassifier()
stopTime=22000, label=1) # a.addDataFiles(fileSourceName="chris_c.txt", fileSourcePath="../", startTime=100, stopTime=1600, label=1) # a.addDataFiles(fileSourceName="ben.txt", fileSourcePath="../", startTime=2000, stopTime=6000, label=1) # a.addDataFiles(fileSourceName="markus.txt", fileSourcePath="../", startTime=500, stopTime=3300, label=1) # a.addDataFiles(fileSourceName="igor.txt", fileSourcePath="../", startTime=100, stopTime=2900, label=1) # a.addDataFiles(fileSourceName="igor2.txt", fileSourcePath="../", startTime=600, stopTime=6000, label=1) dataSet = a.readDataSet(equalLength=False, checkData=False) windowData, windowLabels = a.windowSplitSourceDataTT(inputData=dataSet) features = a.initFeatNormalization(inputData=windowData) features = a.featureNormalization(features=features, initDone=True) features = np.array(features) windowLabels = np.array(windowLabels) features0 = features[windowLabels == 0] features1 = features[windowLabels == 1] m0 = [] m1 = [] diff = [] for i in range(int(len(features[0, ::]))): #f = kernelDensityEstimator(x=features0[::, f0], h=0.15) # plt.scatter(i*np.ones(np.size(features0[::, f0])), features0[::, f0], c='r', marker="s", alpha=0.5) # m0.append(kernelDensityEstimator(features0[::, f0], h=0.15))
a.addDataFiles(fileSourceName="igor.txt", fileSourcePath="../", startTime=100, stopTime=2900, label=6) a.addDataFiles(fileSourceName="igor2.txt", fileSourcePath="../", startTime=600, stopTime=6000, label=6) a.readDataSet(equalLength=False, checkData=False) useDump = False if useDump: a.loadDumpNormParam(dumpName="DecisionTreeClassifier") clf = a.loadDumpClassifier("DecisionTreeClassifier") a.testClassifier(classifier=clf) a.setFileSink(fileSinkName="chris", fileSinkPath="../") a.startLiveClassification() else: a.initFeatNormalization(dumpName="DecisionTreeClassifier") from sklearn.tree import DecisionTreeClassifier clf = DecisionTreeClassifier() a.trainClassifier(classifier=clf) a.dumpClassifier(dumpName="DecisionTreeClassifier") a.testClassifier() windowedData, windowLabels = a.windowSplitSourceDataTT() index = np.linspace(0, len(windowedData) - 1, len(windowedData), dtype=int) random.shuffle(index) trainingData = [] trainingLabels = [] testData = []
startTime=600, stopTime=6000, label=6) a.readDataSet(equalLength=False, checkData=False) useDump = False if useDump: a.loadDumpNormParam(dumpName="GaussianNB") clf = a.loadDumpClassifier("GaussianNB") a.testClassifier(classifier=clf) a.setFileSink(fileSinkName="chris", fileSinkPath="../") a.startLiveClassification() else: a.initFeatNormalization(dumpName="GaussianNB") from sklearn.naive_bayes import GaussianNB clf = GaussianNB() a.trainClassifier(classifier=clf) a.dumpClassifier(dumpName="GaussianNB") a.testClassifier() windowedData, windowLabels = a.windowSplitSourceDataTT() index = np.linspace(0, len(windowedData) - 1, len(windowedData), dtype=int) random.shuffle(index) trainingData = [] trainingLabels = [] testData = [] testLabels = []
a.addDataFiles(fileSourceName="igor.txt", fileSourcePath="../", startTime=100, stopTime=2900, label=6, className="igor") a.addDataFiles(fileSourceName="igor2.txt", fileSourcePath="../", startTime=600, stopTime=6000, label=6) a.readDataSet(equalLength=False, checkData=False) windows, labels = a.windowSplitSourceDataTT() features = a.initFeatNormalization(windows) # create the RFE model and select 3 attributes features = np.array(features) labels = np.array(labels) model = XGBClassifier() model = RFE(model, 100) model.fit(features, labels) # summarize the selection of the attributes print(model.support_) print(model.ranking_) fov = a.returnFeatureIndices() print(fov) plt.scatter(fov, model.ranking_)