main.py

import argparse

from trees2vectors import trees2vectors
from sparse_transform import sparse_transform
from collect_statistic import collect_statistic
from normalize import normalize
from vectors2matrix import vectors2matrix
from matrix2csv import matrix2csv

parser = argparse.ArgumentParser()
parser.add_argument('--input_folder', '-i', nargs=1, type=str, help='folder with trees')
parser.add_argument('--output_folder', '-o', nargs=1, type=str,
                    help='output folder with files, which will contain tree_features and feature values as JSON')
parser.add_argument('--features_file', '-f', nargs=1, type=str, help='path to file with features')
parser.add_argument('--sparse_format', default='list', choices=['list', 'map'])
parser.add_argument('--all_features_file', nargs=1, type=str,
                    help='path to all_features file (in JSON) generated by trees2vectors stage')
parser.add_argument('--output_file', nargs=1, type=str,
                    help='path to output matrix file (dataset.json (vectors2matrix stage)'
                         ', dataset.csv (matrix2csv stage), for example)')
parser.add_argument('--input_file', nargs=1, type=str,
                    help='path to input matrix file (dataset.json, for example)')
parser.add_argument('-n', nargs='*', default=[1, 2, 3], help='n for collect n-grams statistic')
parser.add_argument('--stage', '-s',
                    choices=['trees2vectors', 'sparse_transformation',
                             'normalize', 'collect_statistic', 'vectors2matrix', 'matrix2csv'])
args = parser.parse_args()
stage = args.stage

if stage == 'trees2vectors':
    output_folder = args.output_folder[0]
    input_folder = args.input_folder[0]
    features_file = args.features_file[0]
    trees2vectors(input_folder, output_folder, features_file)
elif stage == 'sparse_transformation':
    output_folder = args.output_folder[0]
    input_folder = args.input_folder[0]
    sparse_format = args.sparse_format
    all_features_file = args.all_features_file[0]
    sparse_transform(input_folder, output_folder, all_features_file, sparse_format)
elif stage == 'normalize':
    output_folder = args.output_folder[0]
    input_folder = args.input_folder[0]
    all_features_file = args.all_features_file[0]
    normalize(input_folder, output_folder, all_features_file)
elif stage == 'collect_statistic':
    output_folder = args.output_folder[0]
    all_features_file = args.all_features_file[0]
    n = args.n
    collect_statistic(output_folder, all_features_file, n)
elif stage == 'vectors2matrix':
    input_folder = args.input_folder[0]
    output_file = args.output_file[0]
    vectors2matrix(input_folder, output_file)
elif stage == 'matrix2csv':
    input_file = args.input_file[0]
    output_file = args.output_file[0]
    matrix2csv(input_file, output_file)