SentEval/senteval/utils.py

# Copyright (c) 2017-present, Facebook, Inc.
# All rights reserved.
#
# This source code is licensed under the license found in the
# LICENSE file in the root directory of this source tree.
#

from __future__ import absolute_import, division, unicode_literals

import numpy as np
import re
import inspect
from torch import optim


def create_dictionary(sentences):
    words = {}
    for s in sentences:
        for word in s:
            if word in words:
                words[word] += 1
            else:
                words[word] = 1
    words['<s>'] = 1e9 + 4
    words['</s>'] = 1e9 + 3
    words['<p>'] = 1e9 + 2
    # words['<UNK>'] = 1e9 + 1
    sorted_words = sorted(words.items(), key=lambda x: -x[1])  # inverse sort
    id2word = []
    word2id = {}
    for i, (w, _) in enumerate(sorted_words):
        id2word.append(w)
        word2id[w] = i

    return id2word, word2id


def cosine(u, v):
    return np.dot(u, v) / (np.linalg.norm(u) * np.linalg.norm(v))


class dotdict(dict):
    """ dot.notation access to dictionary attributes """
    __getattr__ = dict.get
    __setattr__ = dict.__setitem__
    __delattr__ = dict.__delitem__


def get_optimizer(s):
    """
    Parse optimizer parameters.
    Input should be of the form:
        - "sgd,lr=0.01"
        - "adagrad,lr=0.1,lr_decay=0.05"
    """
    if "," in s:
        method = s[:s.find(',')]
        optim_params = {}
        for x in s[s.find(',') + 1:].split(','):
            split = x.split('=')
            assert len(split) == 2
            assert re.match("^[+-]?(\d+(\.\d*)?|\.\d+)$", split[1]) is not None
            optim_params[split[0]] = float(split[1])
    else:
        method = s
        optim_params = {}

    if method == 'adadelta':
        optim_fn = optim.Adadelta
    elif method == 'adagrad':
        optim_fn = optim.Adagrad
    elif method == 'adam':
        optim_fn = optim.Adam
    elif method == 'adamax':
        optim_fn = optim.Adamax
    elif method == 'asgd':
        optim_fn = optim.ASGD
    elif method == 'rmsprop':
        optim_fn = optim.RMSprop
    elif method == 'rprop':
        optim_fn = optim.Rprop
    elif method == 'sgd':
        optim_fn = optim.SGD
        assert 'lr' in optim_params
    else:
        raise Exception('Unknown optimization method: "%s"' % method)

    # check that we give good parameters to the optimizer
    expected_args = inspect.getargspec(optim_fn.__init__)[0]
    assert expected_args[:2] == ['self', 'params']
    if not all(k in expected_args[2:] for k in optim_params.keys()):
        raise Exception('Unexpected parameters: expected "%s", got "%s"' % (
            str(expected_args[2:]), str(optim_params.keys())))

    return optim_fn, optim_params