utils.py

import html
import tensorflow as tf


def find_trainable_variables(key):
    return tf.get_collection(
        tf.GraphKeys.TRAINABLE_VARIABLES, ".*{}.*".format(key))


def preprocess(text, front_pad='\n ', end_pad=' '):
    text = html.unescape(text)
    text = text.replace('\n', ' ').strip()
    text = front_pad+text+end_pad
    text = text.encode()
    return text


def iter_data(*data, **kwargs):
    size = kwargs.get('size', 128)
    try:
        n = len(data[0])
    except:
        n = data[0].shape[0]
    batches = n // size
    if n % size != 0:
        batches += 1

    for b in range(batches):
        start = b * size
        end = (b + 1) * size
        if end > n:
            end = n
        if len(data) == 1:
            yield data[0][start:end]
        else:
            yield tuple([d[start:end] for d in data])


class HParams(object):

    def __init__(self, **kwargs):
        for k, v in kwargs.items():
            setattr(self, k, v)