kanekomasahiro

import numpy as np


def calculate_matrix_cosine_similarity(matrix1, matrix2):
    return np.dot(matrix1, matrix2.T) / (np.linalg.norm(matrix1, axis=1) * np.linalg.norm(matrix2, axis=1)).reshape(-1, 1)

kanekomasahiro

import itertools


def flat_list_of_lists(inputs):
    return itertools.chain.from_iterable(inputs)

kanekomasahiro

import numpy as np


def calculate_mean_vector(embedding):
    return np.mean(embedding[list(embedding.vocab)], axis=0)

kanekomasahiro

import numpy as np


def calculate_vector_cosine_similarity(vector1, vector2):
    return np.dot(vector1, vector2) / (np.linalg.norm(vector1) * np.linalg.norm(vector2))

kanekomasahiro

import linecache

from gensim.models import KeyedVectors


def load_embedding_with_gensim(embedding_name):
    '''
    Load embeddings with gensim.
    '''
    if embedding_name.endswith('bin'):

kanekomasahiro

import argparse


parser = argparse.ArgumentParser()

args_list = lambda x:list(map(str, x.split(',')))
parser.add_argument('--inputs', type=args_list)

args = parser.parse_args()

kanekomasahiro

import argparse


def parse_args():
    parser = argparse.ArgumentParser()

    parser.add_argument('--input', type=str, required=True)

    args = parser.parse_args()

kanekomasahiro

import gensim
import argparse

import numpy as np


def parse_args():
    parser = argparse.ArgumentParser()
    parser.add_argument('--input', type=str, required=True)
    parser.add_argument('--output', type=str, required=True)

kanekomasahiro

import regex as re


def split_sentence_to_words(sent):
    pat = re.compile(r"'s|'t|'re|'ve|'m|'ll|'d| ?\p{L}+| ?\p{N}+| ?[^\s\p{L}\p{N}]+|\s+(?!\S)|\s+")
    
    return re.findall(pat, sent)

kanekomasahiro

import sys
import linecache

from gensim.models import KeyedVectors


def save_word_embedding_text_to_binary(input, output):
    if linecache.getline(input, 1).split() == 2:
        no_header = False
    else: