utils.py

import pickle
import itertools
import codecs
import numpy as np

NONE_TAG = "<NONE>"
POS_KEY = "POS"

class CSVLogger:

    def __init__(self, filename, columns):
        self.file = open(filename, "w")
        self.columns = columns
        self.file.write(','.join(columns) + "\n")

    def add_column(self, data):
        self.file.write(','.join([str(d) for d in data]) + "\n")
        self.file.flush()

    def close(self):
        self.file.close()


def read_pretrained_embeddings(filename, w2i):
    word_to_embed = {}
    with codecs.open(filename, "r", "utf-8") as f:
        for line in f:
            split = line.split()
            if len(split) > 2:
                word = split[0]
                vec = split[1:]
                word_to_embed[word] = vec
    embedding_dim = len(word_to_embed[list(word_to_embed.keys())[0]])
    out = np.random.uniform(-0.8, 0.8, (len(w2i), embedding_dim))
    for word, embed in list(word_to_embed.items()):
        embed_arr = np.array(embed)
        if np.linalg.norm(embed_arr) < 15.0 and word in w2i:
            # Theres a reason for this if condition.  Some tokens in ptb
            # cause numerical problems because they are long strings of the same punctuation, e.g
            # !!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!! which end up having huge norms, since Morfessor will
            # segment it as a ton of ! and then the sum of these morpheme vectors is huge.
            out[w2i[word]] = np.array(embed)
    return out


def read_text_embs(files):
    word_embs = dict()
    for filename in files:
        with codecs.open(filename, "r", "utf-8") as f:
            for line in f:
                split = line.split()
                if len(split) > 2:
                    word_embs[split[0]] = np.array([float(s) for s in split[1:]])
    return list(zip(*iter(word_embs.items())))

def read_pickle_embs(files):
    word_embs = dict()
    for filename in files:
        print(filename)
        words, embs = pickle.load(open(filename, "r"))
        word_embs.update(list(zip(words, embs)))
    return list(zip(*iter(word_embs.items())))


def split_tagstring(s, uni_key=False, has_pos=False):
    '''
    Returns attribute-value mapping from UD-type CONLL field
    :param uni_key: if toggled, returns attribute-value pairs as joined strings (with the '=')
    :param has_pos: input line segment includes POS tag label
    '''
    if has_pos:
        s = s.split("\t")[1]
    ret = [] if uni_key else {}
    if "=" not in s: # incorrect format
        return ret
    for attval in s.split('|'):
        attval = attval.strip()
        if not uni_key:
            a,v = attval.split('=')
            ret[a] = v
        else:
            ret.append(attval)
    return ret


def morphotag_strings(i2ts, tag_mapping, pos_separate_col=True):
    senlen = len(list(tag_mapping.values())[0])
    key_value_strs = []

    # j iterates along sentence, as we're building the string representations
    # in the opposite orientation as the mapping
    for j in range(senlen):
        place_strs = []
        for att, seq in list(tag_mapping.items()):
            val = i2ts[att][seq[j]]
            if pos_separate_col and att == POS_KEY:
                pos_str = val
            elif val != NONE_TAG:
                place_strs.append(att + "=" + val)
        morpho_str = "|".join(sorted(place_strs))
        if pos_separate_col:
            key_value_strs.append(pos_str + "\t" + morpho_str)
        else:
            key_value_strs.append(morpho_str)
    return key_value_strs

def sortvals(dct):
    return [v for (k,v) in sorted(dct.items())]