def main(files):
list(map(z.comp(convert_and_update, z.do(print)), files))
key in `ks`"""
lowtxt = txt.lower()
find_icase = partial(findall_ignore_case, low_txt=lowtxt, txt=txt)
casedata = (DataFrame([(k, len(find_all(txt, k)), len(find_icase(k)))
for k in ks], columns=['Word', 'Case', 'Nocase'])
.query('Case != Nocase').assign(Ratio=lambda x: x.eval('Case / Nocase'))
.sort_values('Ratio', ascending=True).reset_index(drop=1)
)
return {diffcase: phrase2wd(k) for k in casedata.query('Ratio < @thresh').Word
for diffcase in set(find_icase(k))}
# Gensim
def to_gensim_params(cnf, **kw):
gparams = dict(
size=cnf.N, # 80, #
alpha=cnf.eta,
min_alpha=cnf.min_eta,
window=cnf.C / 2,
sample=0,
negative=cnf.K, #[5, 7, 10, 12, 15, 17], 0
sg=1,
# iter=4,
)
gparams.update( **kw)
return gparams
phrase2wd = lambda x: '_'.join(re.split(r'[ -]', x))
take = z.comp(list, islice)
ilen = lambda xs: sum(1 for _ in xs)
# import itertools as it
from os.path import basename, splitext, join, abspath
import re
import subprocess
import toolz.curried as z
import argparse
import yaml
fst = z.operator.itemgetter(0)
snd = z.operator.itemgetter(1)
NOTEBOOK_DIR = '_notebooks'
# NOTEBOOK_HTML_DIR = 'ipy_html'
NOTEBOOK_HTML_DIR = '_includes'
base = z.comp(fst, splitext, basename)
htmlname = z.comp(z.curry(join, NOTEBOOK_HTML_DIR), '{}.html'.format, base)
pat = re.compile(r'''^---\n
(.+?)\n
---
''', re.VERBOSE + re.DOTALL) # # ^---
# print(pat.findall(txt)[0])
def post_todict(posttxt, ret_yaml=False):
m = pat.findall(posttxt)
if not m:
if ret_yaml:
return (None, None)
return
import operator
import toolz.curried as toolz
import pyramda as R
reg_typename = "{}_reg_t".format
rin_name = "{}_rin".format
r_name = "{}_r".format
v_name = "{}_v".format
architecture_id = "two_process_{}".format
indent = R.curry_n(
2,
toolz.comp(
"\n".join,
R.apply(map),
R.unapply(toolz.juxt([
toolz.comp(R.add, R.multiply(" "), toolz.first),
toolz.comp(operator.methodcaller("split", "\n"), toolz.second)]))))
import re
from bs4 import BeautifulSoup
from ratelimit import limits, sleep_and_retry
from requests import get
from tqdm import tqdm
from yaml import dump
from toolz.curried import comp
data_path = 'data.yaml'
country_list_url = 'https://simple.wikipedia.org/wiki/List_of_European_countries'
border_list_url = 'https://en.wikipedia.org/wiki/List_of_countries_and_territories_by_land_borders'
base_url = 'https://en.wikipedia.org/wiki/'
wikilimiter = comp(sleep_and_retry, limits(1, 1))
@wikilimiter
def download_countries():
country_list_page = BeautifulSoup(get(country_list_url).text, features='html.parser')
country_table = [row.find_all('a', href=re.compile(r'^/wiki/'))[1:]
for row in country_list_page.find('tbody').find_all('tr')[1:]]
return {str(a[0].string): str(a[-1]['href']).replace('/wiki/', '') for a in country_table}
@wikilimiter
def download_borders():
border_list_page = BeautifulSoup(get(border_list_url).text, features='html.parser')
return dict(((q:=[str(a.string) for a in row.find_all('a', href=re.compile(r'^/wiki/')) if str(a.string)[0].isupper()])[0], q[1:])
for row in border_list_page.find('tbody').find_all('tr')[2:] if not ('overseas' in str(row) and 'excluding' not in str(row)))
@wikilimiter
def download_coords(capital):
capital_page = BeautifulSoup(get(base_url + capital).text, features='html.parser')
from itertools import repeat, islice, count
import numpy as np
import numpy.random as nr
from numpy.linalg import norm
from operator import itemgetter as itg
from pandas import Series, DataFrame, Index
from numba import jit
import toolz.curried as z
from voluptuous import Any, Invalid, Schema, ALLOW_EXTRA
import numba_utils as nbu
import utils as ut
nopython = jit(nopython=True)
map = z.comp(list, builtins.map)
UNKNOWN = '<UNK>'
# Matrix weight representation
class Cat(object):
"Join and split W matrices for passing as single arg"
@staticmethod
def join(w1, w2):
return np.hstack([w1, w2.T])
@staticmethod
def split(Wall):
n = Wall.shape[1] / 2
W1, W2_ = Wall[:, :n], Wall[:, n:]
return W1, W2_.T
import operator
import toolz.curried as toolz
import pyramda as R
reg_typename = "{}_reg_t".format
rin_name = "{}_rin".format
r_name = "{}_r".format
v_name = "{}_v".format
architecture_id = "two_process_{}".format
indent = R.curry_n(
2,
toolz.comp(
"\n".join, R.apply(map),
R.unapply(
toolz.juxt([
toolz.comp(R.add, R.multiply(" "), toolz.first),
toolz.comp(operator.methodcaller("split", "\n"), toolz.second)
]))))
casedata = (DataFrame(
[(k, len(find_all(txt, k)), len(find_icase(k))) for k in ks],
columns=['Word', 'Case', 'Nocase']).query('Case != Nocase').assign(
Ratio=lambda x: x.eval('Case / Nocase')).sort_values(
'Ratio', ascending=True).reset_index(drop=1))
return {
diffcase: phrase2wd(k)
for k in casedata.query('Ratio < @thresh').Word
for diffcase in set(find_icase(k))
}
# Gensim
def to_gensim_params(cnf, **kw):
gparams = dict(
size=cnf.N, # 80, #
alpha=cnf.eta,
min_alpha=cnf.min_eta,
window=cnf.C / 2,
sample=0,
negative=cnf.K, #[5, 7, 10, 12, 15, 17], 0
sg=1,
# iter=4,
)
gparams.update(**kw)
return gparams
phrase2wd = lambda x: '_'.join(re.split(r'[ -]', x))
take = z.comp(list, islice)
ilen = lambda xs: sum(1 for _ in xs)
def main(files):
list(map(z.comp(convert_and_update, z.do(print)), files))
# import itertools as it
from os.path import basename, splitext, join, abspath
import re
import subprocess
import toolz.curried as z
import argparse
import yaml
fst = z.operator.itemgetter(0)
snd = z.operator.itemgetter(1)
NOTEBOOK_DIR = '_notebooks'
# NOTEBOOK_HTML_DIR = 'ipy_html'
NOTEBOOK_HTML_DIR = '_includes'
base = z.comp(fst, splitext, basename)
htmlname = z.comp(z.curry(join, NOTEBOOK_HTML_DIR), '{}.html'.format, base)
pat = re.compile(r'''^---\n
(.+?)\n
---
''', re.VERBOSE + re.DOTALL) # # ^---
# print(pat.findall(txt)[0])
def post_todict(posttxt, ret_yaml=False):
m = pat.findall(posttxt)
if not m:
if ret_yaml:
return (None, None)
return
import toolz.curried as toolz
import pyramda as R
get_sigtype = R.if_else(
toolz.comp(R.equals(1), len), R.always("std_ulogic"),
toolz.comp("std_logic_vector({} downto 0)".format, R.dec, len))
import toolz.curried as toolz
import pyramda as R
get_sigtype = R.if_else(
toolz.comp(R.equals(1), len),
R.always("std_ulogic"),
toolz.comp("std_logic_vector({} downto 0)".format, R.dec, len))
