How to use the nltools.misc function in nltools

import time
import xml.etree.ElementTree as ET
import numpy as np

from optparse               import OptionParser
from gensim.models          import word2vec
from nltools                import misc, tokenizer
from align_model            import AlignModel

OUTPUT_DIR  = 'out'

#
# init, cmdline
#

misc.init_app('csv_align')

parser = OptionParser("usage: %prog [options] foo.csv")

parser.add_option ("-v", "--verbose", action="store_true", dest="verbose",
                   help="verbose output")

(options, args) = parser.parse_args()

if options.verbose:
    logging.basicConfig(level=logging.DEBUG)
else:
    logging.basicConfig(level=logging.INFO)

if len(args) != 1:
    parser.print_usage()
    sys.exit(1)

promptsfn = '%s/etc/prompts-original' % dstdir
            transfn = '%s/%s/%s/%s/%s-%s.trans.txt' % (srcdir, subset, speaker, book_id, speaker, book_id)

            with codecs.open (promptsfn, 'w', 'utf8') as promptsf:
                with codecs.open(transfn, 'r', 'utf8') as transf:
                    for line in transf:
                        parts = line.split()
                        promptsf.write(line)

                        flac_src = '%s/%s/%s/%s/%s.flac' % (srcdir, subset, speaker, book_id, parts[0])
                        flac_dst = '%s/flac/%s.flac' % (dstdir, parts[0])

                        logging.debug (' %s -> %s' % (flac_src, flac_dst))

                        misc.symlink(flac_src, flac_dst)


            logging.debug ('%s written.' % promptsfn)

nwords = 0

        for word in words:
            if word in index2word_set:
                nwords = nwords+1
                featureVec = np.add(featureVec, model[word])

        if nwords>0:
            featureVec = np.divide(featureVec, nwords)
        return featureVec

#
# init, cmdline
#

misc.init_app('aim2topics')

parser = OptionParser("usage: %prog [options] foo.aiml")

parser.add_option ("-l", "--lang", dest="lang", type = "string", default='en',
                   help="language, default: en")
parser.add_option ("-v", "--verbose", action="store_true", dest="verbose",
                   help="verbose output")

(options, args) = parser.parse_args()

if options.verbose:
    logging.basicConfig(level=logging.DEBUG)
else:
    logging.basicConfig(level=logging.INFO)

if len(args) != 1:

VOICE_PATH      = 'data/model/%s'
# VOICE           = 'voice-karlsson-latest'
DEFAULT_VOICE   = 'voice-linda-latest'
BATCH_SIZE      = 32

# x_13036.npy (1, 240) these figures suggest that the agents of the secret service are substantially overworked .

X_FN  = NPDIR + 'x_13036.npy'
X_TXT = 'These figures suggest that the agents of the secret service are substantially overworked.'
X_L   = len(X_TXT)

#
# init
#

misc.init_app(PROC_TITLE)

#
# command line
#

parser = OptionParser("usage: %prog [options] [

misc.mkdirs('%s/valid' % data_dir)
misc.mkdirs('%s/train' % data_dir)

#
# load dict
#

logging.info("loading lexicon...")
lex = Lexicon(file_name=dictionary)
logging.info("loading lexicon...done.")

#
# language model
#

misc.copy_file('%s/lm.arpa' % language_model_dir, '%s/lm.arpa' % data_dir)


#
# scripts
#

misc.render_template('data/src/speech/w2l_run_train.sh.template', '%s/run_train.sh' % work_dir, w2l_env_activate=w2l_env_activate, w2l_train=w2l_train)
misc.render_template('data/src/speech/w2l_run_decode.sh.template', '%s/run_decode.sh' % work_dir, w2l_env_activate=w2l_env_activate, w2l_train=w2l_train)
misc.mkdirs('%s/config/conv_glu' % work_dir)
misc.render_template('data/src/speech/w2l_config_conv_glu_train.cfg.template', '%s/config/conv_glu/train.cfg' % work_dir, runname=model_name)
misc.copy_file('data/src/speech/w2l_config_conv_glu_network.arch', '%s/config/conv_glu/network.arch' % work_dir)

#
# export audio
#

if not pws in ps:
                    ps[pws] = set([p])
                else:
                    ps[pws].add(p)

logging.info ( "%s written." % dictfn2 )

logging.info ( "Exporting dictionary ... done." )

#
# copy phoneme sets from original model
#

misc.copy_file ('%s/data/local/dict/nonsilence_phones.txt' % src_model, '%s/data/local/dict/nonsilence_phones.txt' % dst_dir)
misc.copy_file ('%s/data/local/dict/silence_phones.txt' % src_model,    '%s/data/local/dict/silence_phones.txt' % dst_dir)
misc.copy_file ('%s/data/local/dict/optional_silence.txt' % src_model,  '%s/data/local/dict/optional_silence.txt' % dst_dir)
misc.copy_file ('%s/data/local/dict/extra_questions.txt' % src_model,   '%s/data/local/dict/extra_questions.txt' % dst_dir)

#
# language model / grammar
#

if lm_name.endswith('arpa'):
    misc.copy_file (lm_name, '%s/lm.arpa' % dst_dir)
elif lm_name.endswith('jsgf'):
    misc.copy_file (lm_name, '%s/G.jsgf' % dst_dir)
else:
    misc.copy_file (lm_name, '%s/G.src.fst' % dst_dir)


#

def main(verbose=False):
    """Convert gspv2 corpus to the VoxForge corpus format

    The variable `speech_arc` in ~/.speechrc must point to a folder
    gspv2 which is used as the source containing the original gspv2 corpus, 
    i.e. containing the subfolders dev, test, and train.

    The variable `speech_corpora` in ~/.speechrc must point to a folder
    where the resulting corpus should be written. The script will create
    a subfolder gspv2 here for the resulting voxforge-formatted data.
    """
    misc.init_app('speech_audio_scan')
    config = misc.load_config('.speechrc')

    if verbose:
        logging.basicConfig(level=logging.DEBUG)
    else:
        logging.basicConfig(level=logging.INFO)

    speech_arc_dir     = Path(config.get("speech", "speech_arc"))
    speech_corpora_dir = Path(config.get("speech", "speech_corpora"))
    src_root_dir = speech_arc_dir / "gspv2"
    dst_root_dir = speech_corpora_dir / "gspv2"

    exit_if_dst_root_dir_exists(dst_root_dir)

    speakers = set()
    speaker_gender = {}