Python create_same_size_sequencesの例

コード例 #1

ファイル: ngram_classifier.py プロジェクト: yallapragada/ngram_classifier

def create_data(sequences1, sequences2):
    print("size of data before same sizing", len(sequences1), len(sequences2))
    sequences1, sequences2 = util.create_same_size_sequences(sequences1, sequences2)
    print("size of data after same sizing", len(sequences1), len(sequences2))
    matrix1 = compute_frequency_matrix(3, sequences1)
    matrix2 = compute_frequency_matrix(3, sequences2)
    len1 = len(matrix1)
    len2 = len(matrix2)
    one_s = ones(len1)
    zero_s = zeros(len2)
    data=concatenate([matrix1,matrix2])
    labels=concatenate([one_s,zero_s])
    return data, labels

コード例 #2

ファイル: ngram_classifier_p.py プロジェクト: yallapragada/ngram_classifier

def create_data(sequences1, sequences2):
    print("size of data before same sizing", len(sequences1), len(sequences2))
    sequences1, sequences2 = util.create_same_size_sequences(sequences1, sequences2)
    print("size of data after same sizing", len(sequences1), len(sequences2))

    matrix1 = compute_frequency_matrix(3, sequences1)
    matrix2 = compute_frequency_matrix(3, sequences2)
    
    data = concatenate([matrix1, matrix2])
    print("size of feature vector before zero variance check", len(data), " ", len(data[0]))
    data = remove_zeros(data)
    print("size of feature vector after zero variance check", len(data), " ", len(data[0]))

    len1 = len(matrix1)
    len2 = len(matrix2)
    one_s = ones(len1)
    zero_s = zeros(len2)

    labels = concatenate([one_s, zero_s])
    return data, labels

コード例 #3

ファイル: ngram_classifier_p_shuffle_3letter_Li.py プロジェクト: yallapragada/ngram_classifier

def create_data(sequences1, sequences2):
    print("size of data before same sizing", len(sequences1), len(sequences2))
    sequences1, sequences2 = util.create_same_size_sequences(sequences1, sequences2)
    print("size of data after same sizing", len(sequences1), len(sequences2))

    sequences1 = [util.reduce_alphabet_Li(sequence) for sequence in sequences1]
    sequences2 = [util.reduce_alphabet_Li(sequence) for sequence in sequences2]    
    
    matrix1 = compute_frequency_matrix(3, sequences1)
    matrix2 = compute_frequency_matrix(3, sequences2)
    
    data = concatenate([matrix1, matrix2])
    print("size of feature vector ", len(data), " ", len(data[0]))

    len1 = len(matrix1)
    len2 = len(matrix2)
    one_s = ones(len1)
    zero_s = zeros(len2)

    labels = concatenate([one_s, zero_s])
    return data, labels