def prepareData():
    raw_data, raw_label = loadTrainingData('dataset/train.csv')
    test_data = loadTestingData('dataset/test.csv')

    #nomalize the train & test data
    raw_data = nnet.normalization(raw_data)
    test_data = nnet.normalization(test_data)
    
    #split training data to training set & validating set
    training_data = raw_data[:37000,:]
    training_label = raw_label[:37000,]
    validating_data = raw_data[37000:,:]
    validating_label = raw_label[37000:,]
    
    return training_data.transpose(),training_label,validating_data.transpose(),validating_label,test_data.transpose()
Exemple #2
0
def prepareData():
    raw_data, raw_label = loadTrainingData('dataset/train.csv')
    test_data = loadTestingData('dataset/test.csv')

    #nomalize the train & test data
    raw_data = nnet.normalization(raw_data)
    test_data = nnet.normalization(test_data)

    #split training data to training set & validating set
    training_data = raw_data[:37000, :]
    training_label = raw_label[:37000, ]
    validating_data = raw_data[37000:, :]
    validating_label = raw_label[37000:, ]

    return training_data.transpose(
    ), training_label, validating_data.transpose(
    ), validating_label, test_data.transpose()
Exemple #3
0
    test_data = test_data[:,test_set[0].flatten()]

    #step2 divide training set into feature learning set & supervised training set
    labeled_set = np.nonzero(labels>=5)
    unlabeled_set = np.nonzero(labels<=4)
    
    labeled_dataset = data[:,(labels<=4).flatten()]
    labeled_labelset = labels[(labels<=4).flatten()]
    unlabeled_dataset = data[:,(labels>=5).flatten()]

    unlabeled_dataset = unlabeled_dataset[:,:unlabeled_dataset.shape[1]/3]
    print unlabeled_dataset.shape

    #step3 feature learning using sparse autoencoder
    #TODO move normalization to miscellaneous
    unlabeled_dataset = nnet.normalization(unlabeled_dataset)
    W = nnet.sparseAutoencoder(inputSize,hiddenSize,sparsityParam,lmd,beta,alpha,unlabeled_dataset)
    W1 = np.reshape(W[:hiddenSize*inputSize,], (hiddenSize, inputSize))
    b1 = np.reshape(W[2*hiddenSize*inputSize:2*hiddenSize*inputSize+hiddenSize,],(hiddenSize,1))

    #step4 plot the learned feature
    fig = plt.figure(2)

    for index in range(hiddenSize/10):
        weight = W1[index,:]
        weight = np.reshape(weight,(28,28))
        ax = fig.add_subplot(5,4,1+index)
        ax.imshow(weight,mpl.cm.gray)

    plt.show()
    label = sio.loadmat('dataset/mnist_dataset/mnist_train_labels.mat')
    data = image['mnist_train']
    training_labels = label['mnist_train_labels']
    training_data = data.transpose()
    #testing data
    image = sio.loadmat('dataset/mnist_dataset/mnist_test.mat')
    label = sio.loadmat('dataset/mnist_dataset/mnist_test_labels.mat')
    test_data = image['mnist_test']
    test_labels = label['mnist_test_labels']
    test_data = test_data.transpose()

    start = time.time()
    
    #step2 L1 feature learning using sparse autoencoder
    #TODO move normalization to miscellaneous
    training_data = nnet.normalization(training_data)
    #W = nnet.sparseAutoencoder(inputSize,hiddenSizeL1,sparsityParam,lmd,beta,alpha,training_data,iters=500)
    W = load_data('weightL1')
    W = np.array(W)
    W = W.transpose()
    #savedata(W,'weightL1')
    W11 = np.reshape(W[:hiddenSizeL1*inputSize,], (hiddenSizeL1, inputSize))
    b11 = np.reshape(W[2*hiddenSizeL1*inputSize:2*hiddenSizeL1*inputSize+hiddenSizeL1,],(hiddenSizeL1,1))

    #step3 L2 feature learning using sparse autoencoder
    training_a1 = nnet.sigmoid(W11.dot(training_data)+b11)
    #W = nnet.sparseAutoencoder(hiddenSizeL1,hiddenSizeL2,sparsityParam,lmd,beta,0.009,training_a1,iters=500)
    W = load_data('weightL2')
    W = np.array(W)
    W = W.transpose()
    #savedata(W,'weightL2')
    numClasses = 10
    hiddenSizeL1 = 5
    sparsityParam = 0.1
    beta = 3
    lmd = 0.001
    alpha = 0.07

    # step1 load training data
    # TODO move normalization to miscellaneous
    file_path = "dataset/faces.mat"

    images = sio.loadmat(file_path)
    data = images["data"]
    data = data.transpose()
    # normalize data
    data = nnet.normalization(data)

    # split data to training set and testing set
    testing_lables = []
    training_labels = []
    for index in range(10):
        train = data[:, 90 * index : 90 * index + 80]
        test = data[:, 90 * index + 80 : 90 * (index + 1)]
        if index == 0:
            training_set = train
            testing_set = test
        else:
            training_set = np.c_[training_set, train]
            testing_set = np.c_[testing_set, test]

        labels = 10 * [index]
    label = sio.loadmat('dataset/mnist_dataset/mnist_train_labels.mat')
    data = image['mnist_train']
    training_labels = label['mnist_train_labels']
    training_data = data.transpose()
    #testing data
    image = sio.loadmat('dataset/mnist_dataset/mnist_test.mat')
    label = sio.loadmat('dataset/mnist_dataset/mnist_test_labels.mat')
    test_data = image['mnist_test']
    test_labels = label['mnist_test_labels']
    test_data = test_data.transpose()

    start = time.time()

    #step2 L1 feature learning using sparse autoencoder
    #TODO move normalization to miscellaneous
    training_data = nnet.normalization(training_data)
    #W = nnet.sparseAutoencoder(inputSize,hiddenSizeL1,sparsityParam,lmd,beta,alpha,training_data,iters=500)
    W = load_data('weightL1')
    W = np.array(W)
    W = W.transpose()
    #savedata(W,'weightL1')
    W11 = np.reshape(W[:hiddenSizeL1 * inputSize, ], (hiddenSizeL1, inputSize))
    b11 = np.reshape(
        W[2 * hiddenSizeL1 * inputSize:2 * hiddenSizeL1 * inputSize +
          hiddenSizeL1, ], (hiddenSizeL1, 1))

    #step3 L2 feature learning using sparse autoencoder
    training_a1 = nnet.sigmoid(W11.dot(training_data) + b11)
    #W = nnet.sparseAutoencoder(hiddenSizeL1,hiddenSizeL2,sparsityParam,lmd,beta,0.009,training_a1,iters=500)
    W = load_data('weightL2')
    W = np.array(W)
Exemple #7
0
    numClasses = 10
    hiddenSizeL1 = 5
    sparsityParam = 0.1
    beta = 3
    lmd = 0.001
    alpha = 0.07

    #step1 load training data
    #TODO move normalization to miscellaneous
    file_path = 'dataset/faces.mat'

    images = sio.loadmat(file_path)
    data = images['data']
    data = data.transpose()
    #normalize data
    data = nnet.normalization(data)

    #split data to training set and testing set
    testing_lables = []
    training_labels = []
    for index in range(10):
        train = data[:,90*index:90*index+80]
        test = data[:,90*index+80:90*(index+1)]
        if index == 0:
            training_set = train 
            testing_set = test
        else:
            training_set = np.c_[training_set, train]
            testing_set = np.c_[testing_set, test]

        labels = 10*[index]