kzinmr

df = df.set_index('date')
for date, new_df in df.groupby(level=0):
    print(new_df)

kzinmr

import pickle
import subprocess
from pathlib import Path
import sys

from lxml import etree


def extract_textbox_slide(page):
    """

kzinmr

import contextlib
import time

@contextlib.contextmanager
def time_measure(ident):
    tstart = time.time()
    yield
    elapsed = time.time() - tstart
    print(f"{ident}: {elapsed} sec")

kzinmr

"""
Usage: python remove_output.py notebook.ipynb [ > without_output.ipynb ]
Modified from remove_output by Minrk

"""
import sys
import io
import os
from IPython.nbformat.current import read, write

kzinmr

{
    'NAME':
    {
        'Name_Other': {},
        'Person': {},
        'God': {},
        'Organization': 
        {
            'Organization_Other': {},
            'International_Organization': {},

kzinmr

def katakana_to_hiragana(string):
    result = ''
    for character in string:
        code = ord(character)
        if ord('ァ') <= code <= ord('ヶ'):
            result += chr(code - ord('ァ') + ord('ぁ'))
        else:
            result += character
    return result

kzinmr

from collections import defaultdict

def dict_of_list(keys, values):
    assert(len(keys) == len(values))
    key2values = defaultdict(list)
    for k, v in zip(keys, values):
        key2values[k].append(v)
    return key2values

kzinmr

from itertools import tee, zip_longest


def ngrams(iterable, n=3):
    """
    >>> list(ngrams(range(5), 3))
    [(0, 1, 2), (1, 2, 3), (2, 3, 4)]
    """
    ts = tee(iterable, n)
    for i, t in enumerate(ts[1:]):

kzinmr

import numpy as np

def calculate_cosine(query, X, k=10, threshold=0.5):
    """
    Arguments: 
      query: (dim,) or (dim, 1)-array
      X: (dim, N)-array of N-vectors
    Returns:
      topk_score_indices: sorted top-k-scored indices of N-vectors
      topk_score: and its scores

kzinmr

from functools import reduce

def compose(*functions):
    return reduce(lambda f, g: lambda x: f(g(x)), functions, lambda x: x)