index int64 0 1,000k | blob_id stringlengths 40 40 | code stringlengths 7 10.4M |
|---|---|---|
993,800 | 642d72fac44706505376fcc2ff0cc5cada6fa67e | """Views for news-here."""
import webapp2
import models
import view_utils
class MainPage(webapp2.RequestHandler):
"""Handler for landing page."""
def get(self):
"""Show dummy page for now."""
self.response.write(view_utils.render('base.html', {}))
class AddNewsItem(webapp2.RequestHandler):
"""Handler for adding links."""
def get(self):
"""Show add page."""
news_items = models.NewsItem.all()
self.response.write(view_utils.render('add.html', {
'news_items': news_items,
}))
def post(self):
headline = self.request.get('headline', '')
link = self.request.get('link', '')
if headline and link:
news_item = models.NewsItem()
news_item.author = 'abhishek'
news_item.headline = headline
news_item.link = link
news_item.put()
self.response.write('Data saved successfully.')
else:
self.response.write('Some error occured while saving data.')
|
993,801 | 21c130cd0199462a90b0a1f73cec1eb4392b7cc7 | #lesson 6 follow-along
#flow control: while loop
spam = 0
while spam <5:
print('Hello, world!')
spam = spam +1
#the "while" line checks for true/false of the statement
#while true, it will run the block
#so it stops when spam reachs 5
print()
name = ''
while name != 'your name':
print('Please type your name:')
name = input()
print('Thank you')
print()
#that loop doesn't have a counter
#but it still has a true/false check in the while line
#next is a bug example: an infinite loop:
#while True:
# print('Hi!')
#you can escape those with ctrl-c
#another way:
name = ''
while True:
print('Please type your name again:')
name = input()
if name == 'your name':
break
print('Thank you.')
print()
#In that example, the true/false statement will always be true
#so we end the loop with the "break" statement
#continue statement:
#it skips the loop when a certain condition is met
spam = 0
while spam < 5:
spam = spam +1
if spam == 3: #here's the condition. the block will skip using "continue"
continue
print('spam is ' + str(spam))
print('The end.')
|
993,802 | 32d1b59225b1a675e4cc02e1aba0943103e70550 | # Copyright The Cloud Custodian Authors.
# SPDX-License-Identifier: Apache-2.0
from bottle import Bottle, request, response, abort
import json
import logging
import os
from audit import init_audit
from controller import Controller
from utils import Encoder
log = logging.getLogger("sphere11.api")
logging.basicConfig(level=logging.INFO)
config = json.load(open(os.environ.get('SPHERE11_CONFIG', 'config.json')))
controller = Controller(config)
audit = init_audit(config.get('log-group', 'sphere11'))
app = Bottle()
@app.route("/")
def index():
return {"name": "sphere11", "version": "1.0"}
@app.route("/<account_id>/locks", method="GET")
@audit
def account_status(account_id):
result = controller.db.iter_resources(account_id)
response.content_type = "application/json"
return json.dumps(result, indent=2, cls=Encoder)
@app.route("/<account_id>/locks/<resource_id>/lock", method="POST")
@audit
def lock(account_id, resource_id):
request_data = request.json
for rp in ('region',):
if not request_data or rp not in request_data:
abort(400, "Missing required parameter %s" % rp)
return controller.lock(account_id, resource_id, request_data['region'])
@app.route("/<account_id>/locks/<resource_id>", method="GET")
def info(account_id, resource_id):
request_data = request.query
if resource_id.startswith('sg-') and 'parent_id' not in request_data:
abort(400, "Missing required parameter parent_id")
result = controller.info(
account_id, resource_id, request_data.get('parent_id', resource_id))
response.content_type = "application/json"
return json.dumps(result, indent=2, cls=Encoder)
# this set to post to restrict permissions, perhaps another url space.
@app.route("/<account_id>/locks/delta", method="POST")
@audit
def delta(account_id):
request_data = request.json
for rp in ('region',):
if not request_data or rp not in request_data:
abort(400, "Missing required parameter %s" % rp)
result = controller.get_account_delta(
account_id, request_data['region'], api_url())
response.content_type = "application/json"
return json.dumps(result, indent=2, cls=Encoder)
@app.route("/<account_id>/locks/<resource_id>/unlock", method="POST")
@audit
def unlock(account_id, resource_id):
return controller.unlock(account_id, resource_id)
def on_timer(event):
return controller.process_pending()
def on_config_message(records):
for r in records:
json.loads(r['Sns'].get('Message'))
def on_db_change(records):
pass
def api_url():
parsed = request.urlparts
url = "%s://%s%s" % (parsed.scheme, parsed.netloc, request.script_name)
return url
@app.error(500)
def error(e):
response.content_type = "application/json"
return json.dumps({
"status": e.status,
"url": repr(request.url),
"exception": repr(e.exception),
# "traceback": e.traceback and e.traceback.split('\n') or '',
"body": repr(e.body)
}, indent=2)
|
993,803 | 833bcde268ce223a1ea07e6cd666907b8bb11d79 | #!/usr/bin/env python
# -*- coding: utf-8 -*- #
import sys
import argparse
import os
from util import mkdir, image_trans
this_script_path = os.path.dirname(__file__)
sys.path.insert(1, this_script_path + '/../src')
import Parser as rp
def read_params(args):
parser = argparse.ArgumentParser(description='''diff pca | v1.0 at 2015/10/26 by liangzb ''')
parser.add_argument('-i', '--infile', dest='infile', metavar='file', type=str, required=True,
help="set the marker file in")
parser.add_argument('-o', '--outdir', dest='outdir', metavar='DIR', type=str, required=True,
help="set the work dir")
parser.add_argument('-g', '--group', dest='group', metavar='FILE', type=str, required=True,
help="set the group file")
args = parser.parse_args()
params = vars(args)
return params
if __name__ == '__main__':
params = read_params(sys.argv)
mkdir(params['outdir'])
r_job = rp.Rparser()
r_job.open(this_script_path + '/../src/template/05_diff_pca.Rtp')
r_file = params['outdir'] + '/diff_pca.R'
pdf_file = params['outdir'] + '/diff_pca.pdf'
png_file = params['outdir'] + '/diff_pca.png'
var = {
'input_file': params['infile'],
'group_file': params['group'],
'pdf_file': pdf_file
}
r_job.format(var)
r_job.write(r_file)
r_job.run()
image_trans(pdf_file, png_file)
|
993,804 | 4d1f68fca8ea3f06451a22dfd49587c6217fb277 | _file = open("input.txt", "r")
word = _file.readline()
MARKER_LENGTH = 14
for char in range(MARKER_LENGTH, len(word)):
substr = word[char - MARKER_LENGTH:char]
if len(set(substr)) == MARKER_LENGTH:
print(char)
break
|
993,805 | 3addeabedb0a08925b4e2391be8508740f8f0f85 | from django.db import models
from user.models import User
from comment_counts.models import CommentCounts
from open_date_time.models import OpenDate
from utils.base_model.base_comment_model import BaseCommentModel
from utils.base_model.base_image_model import BaseImageModel
from utils.base_model.base_tag_model import BaseTagModel
# 餐厅 -- 模型
class Restaurant(models.Model):
name = models.CharField(max_length=60,null=False,verbose_name="餐厅名")
position = models.CharField(max_length=120,null=False,verbose_name="位置")
averageConsumption = models.IntegerField(null=False,verbose_name="平均消费")
description = models.TextField(null=True,blank=True,verbose_name="描述")
phone = models.CharField(max_length=20,verbose_name="联系方式")
openDate = models.ForeignKey(OpenDate,on_delete=models.PROTECT,verbose_name="营业时间")
commentCount = models.ForeignKey(CommentCounts,on_delete=models.PROTECT,verbose_name="评论统计")
class Meta:
ordering = ["name"]
db_table = "restaurant"
verbose_name = "餐厅"
verbose_name_plural = "餐厅"
def __str__(self):
return self.name
# 餐厅配图 -- 模型
class RestaurantImage(BaseImageModel):
restaurant = models.ForeignKey(Restaurant,on_delete=models.CASCADE,verbose_name="餐厅")
class Meta:
db_table = "restaurant_images"
verbose_name_plural = "餐厅配图"
verbose_name = "餐厅配图"
def __str__(self):
return "{}的配图".format(self.restaurant.name)
# 餐厅评论 -- 模型
class RestaurantComment(BaseCommentModel):
restaurant = models.ForeignKey(Restaurant,null=False,on_delete=models.CASCADE,verbose_name="餐厅")
class Meta:
db_table = "restaurant_comment"
verbose_name_plural = "餐厅评论"
verbose_name = "餐厅评论"
def __str__(self):
return "{}-{}".format(self.restaurant,self.user)
# 餐厅评论配图 -- 模型
class RestaurantCommentImage(BaseImageModel):
restaurantComment = models.ForeignKey(RestaurantComment,on_delete=models.CASCADE,verbose_name="餐厅评论")
class Meta:
ordering=["restaurantComment"]
db_table = "restaurant_comment_image"
verbose_name = "餐厅评论配图"
verbose_name_plural="餐厅评论配图"
def __str__(self):
return "{}的配图".format(self.restaurantComment)
# 餐厅标签
class RestaurantTag(BaseTagModel):
restaurant = models.ManyToManyField(Restaurant,verbose_name="餐厅")
class Meta:
ordering=["tag"]
db_table = "restaurant_tag"
verbose_name = "餐厅标签"
verbose_name_plural = "餐厅标签"
def __str__(self):
return "{}-{}".format(self.restaurant,self.tag)
# 餐厅菜系
class RestaurantFood(models.Model):
name = models.CharField(max_length=60,null=False,blank=False,verbose_name="菜名")
restaurant = models.ManyToManyField(Restaurant,verbose_name="餐厅")
class Meta:
ordering = ["name"]
db_table = "restaurant_food"
verbose_name = "餐厅菜系"
verbose_name_plural = "餐厅菜系"
def __str__(self):
return "{}-{}".format(self.name,self.restaurant)
|
993,806 | c899eadf0960d91e9d4a3f661328e9f1122cba9b | from django.shortcuts import render, get_object_or_404, redirect, reverse, HttpResponseRedirect
from recipe.forms import AddRecipeForm, AddAuthorForm, LoginForm
from recipe.models import Author, Recipe
from django.contrib import messages
from django.contrib.auth import login, logout, authenticate
from django.contrib.auth.forms import UserCreationForm
from django.contrib.auth.decorators import login_required
# Create your views here.
def loginview(request):
if request.method == "POST":
form = LoginForm(request.POST)
if form.is_valid():
data = form.cleaned_data
user = authenticate(
request, username=data['username'], password=data['password']
)
if user:
login(request, user)
return HttpResponseRedirect(
request.GET.get('next', reverse('home'))
)
form = LoginForm()
return render(request, 'generic_form.html', {'form': form})
# https://pythonprogramming.net/user-login-logout-django-tutorial/
def logoutview(request):
logout(request)
messages.info(request, "Logged out successfully!")
return HttpResponseRedirect(
request.GET.get('next', reverse('home'))
)
def index(request):
return render(request, 'index.html',
{'recipes': Recipe.objects.all()})
def recipe(request, id):
recipe = get_object_or_404(Recipe, pk=id)
return render(request, 'recipe.html', {'recipe': recipe})
def author(request, id):
author = get_object_or_404(Author, pk=id)
# Got help from Peter on 5/7 for redering and sorting recipes
recipes = Recipe.objects.filter(author=author)
return render(request, 'author.html', {'author': author, 'recipes': recipes})
# https://www.youtube.com/watch?time_continue=399&v=Tja4I_rgspI&feature=emb_logo
# https://simpleisbetterthancomplex.com/tutorial/2017/02/18/how-to-create-user-sign-up-view.html
@login_required
def add_author(request):
if request.method == 'POST':
form = AddAuthorForm(request.POST)
if form.is_valid():
user = form.save()
user.refresh_from_db()
user.author.name = form.cleaned_data.get('name')
user.author.bio = form.cleaned_data.get('bio')
user.save()
raw_password = form.cleaned_data.get('password1')
user = authenticate(username=user.username, password=raw_password)
login(request, user)
return redirect('home')
else:
form = AddAuthorForm()
return render(request, 'addauthor.html', {'form': form})
@login_required
def add_recipe(request):
if request.method == 'POST':
form = AddRecipeForm(request.POST)
if form.is_valid():
form.save()
return redirect('home')
else:
form = AddRecipeForm()
return render(request, 'addrecipe.html', {'form': form})
|
993,807 | 97b796e041bd1f7c5c7a1579f13e7883b1a35eaa | import json, os, datetime, uuid, logging
import issues_conf as conf
logger = logging.getLogger(__name__)
DATETIME_FORMAT = "%Y-%m-%d %H:%M:%S"
def timestamp():
return datetime.datetime.utcnow().strftime(DATETIME_FORMAT)
def load_data(filename, **kwargs):
try:
return json.load(open(filename, 'r'), **kwargs)
except IOError:
return {}
def save_data(filename, data, **kwargs):
json.dump(data, open(filename, 'w'), indent=0)
class PyIssues(object):
'''
Simple issues collection, based on a directory.
'''
def __init__(self, directory):
self.directory = directory
self.settings = conf.__dict__.copy()
# load local settings
try:
execfile("{0}/conf.py".format(self.directory), self.settings)
logger.debug("Loaded custom settings")
except IOError:
logger.debug("No custom settings")
self.obj_dir = "{0}/objs".format(self.directory)
if not os.path.isdir(self.obj_dir):
os.makedirs(self.obj_dir)
self.issues_file = "{0}/issues.db".format(directory)
self.issues_data = load_data(self.issues_file)
self.dirty = False
def flush(self):
'''
Writes out database if necessary.
Returns whether the database was written or not
'''
if self.issues_data and self.dirty:
logger.debug("Saving database...")
save_data(self.issues_file, self.issues_data)
self.dirty = False
return True
return False
def close(self):
'''
Saves the database if necessary and cleans up.
'''
self.flush()
self.issues_data = None
def filter(self, filters=None, sort=None):
'''
Returns index items matching the criteria.
filters can be a func or a dict of exact matches.
sort is the name of a field, optionally prefixed with '-' for reverse.
'''
items = [ Issue.expand_index(x, self.issues_data[x], self.settings['_index']) for x in self.issues_data ]
if filters:
if hasattr(filters, '__iter__'):
d = filters
filters = lambda x: { i: x.get(i) for i in d } == d
items = [ x for x in items if filters(x) ]
if sort:
reverse = False
if sort[0] == '-':
sort = sort[1:]
reverse = True
items = sorted(items, key=lambda x: x[sort], reverse=reverse)
return items
def match(self, uuid):
'''
Match a single file from a uuid stub so we dont have to type full uuids.
Returns the full file path or throws an error if none or multiple matches.
'''
import glob
matches = glob.glob('{0}/{1}*'.format(self.obj_dir, uuid))
if not matches:
raise PyIssuesException("No match for uuid: {0}".format(uuid))
if len(matches) > 1:
raise PyIssuesException("Multiple matches for {0} - be more specific".format(uuid))
return matches[0]
def get(self, uuid):
'''
Get a single issue.
uuid can be a partial uuid.
'''
with open(self.match(uuid)) as bob:
data = json.load(bob)
return self.create(**data)
def create(self, **params):
return Issue(self.directory, self.settings['_fields'], self.settings['_required'], **params)
def delete(self, uuid):
f = self.match(uuid)
issue = self.get(uuid)
del self.issues_data[issue.uuid]
os.unlink(f)
self.dirty = True
def update(self, issue):
issue.updated = timestamp()
self.issues_data[issue.uuid] = issue.index(self.settings['_index'])
with open('{0}/{1}'.format(self.obj_dir, issue.uuid), 'w') as bob:
issue.write(bob)
self.dirty = True
def rebuild(self):
self.issues_data = {}
self.dirty = True
c = 0
for uuid in os.listdir(self.obj_dir):
issue = self.get(uuid)
if issue.status != 'archived':
self.issues_data[issue.uuid] = issue.index(self.settings['_index'])
c += 1
return c
class Issue(object):
def __init__(self, directory, fields, required, **kwargs):
for f in required:
if not f in kwargs:
raise PyIssuesException("Missing required field: {0}.".format(f))
data = dict(fields)
data.update(kwargs)
for key in data:
setattr(self, key, data[key])
if self.uuid is None:
self.uuid = str(uuid.uuid4())
if self.created is None:
self.created = timestamp()
self._data_dir = "{0}/files/{1}".format(directory, self.uuid)
def index(self, fields):
return [ getattr(self, x) for x in fields ] + [len(self.comments), len(self.attachments)]
@classmethod
def expand_index(cls, uuid, index, fields):
d = dict(zip(fields + ('comments', 'attachments'), index))
d['uuid'] = uuid
return d
def add_comment(self, comment, user):
'''
Add a comment.
Comment is stored in the form (comment, user, timestamp)
'''
self.comments.append((comment, user, timestamp()))
def remove_comment(self, index):
'''
Remove a comment (zero based index)
'''
del(self.comments[index])
def attach_file(self, filename, user):
'''
Attach a file to the issue.
Returns (original_name, stored_name, user, timestamp) as stored
'''
import shutil
if not os.path.exists(filename):
raise Exception("No such file: {0}".format(filename))
if not os.path.exists(self._data_dir):
os.makedirs(self._data_dir)
f = os.path.basename(filename)
i=0
while os.path.exists("{0}/{1}_{2}".format(self._data_dir, f, i)):
i += 1
target = "{0}/{1}_{2}".format(self._data_dir, f, i)
logger.debug("Storing {0} at {1}".format(f, target))
shutil.copy(filename, target)
data = (f, os.path.basename(target), user, timestamp())
self.attachments.append(data)
return data
def remove_file(self, index):
'''
Remove an attached file (zero based index)
'''
_, f = self.get_file(index)
os.unlink(f)
del(self.attachments[index])
def get_file(self, index):
'''
Get path information about an attached file (zero based index)
returns (original, path)
'''
attachment = self.attachments[index]
return (attachment[0], "{0}/{1}".format(os.path.realpath(self._data_dir), attachment[1]))
def write(self, stream):
stream.write(str(self))
def __str__(self):
return json.dumps({ x: self.__dict__[x] for x in self.__dict__ if not x[0] == '_' }, self.__dict__, indent=2)
def __repr__(self):
return "<Issue #{0}>".format(self.uuid[:8])
class PyIssuesException(Exception):
pass |
993,808 | f0b6010b9a454e0ba92f8292f81afabd2ec7da71 | from project.player.player import Player
class BattleField():
@staticmethod
def fight(attacker: Player, enemy: Player):
if attacker.is_dead or enemy.is_dead:
raise ValueError("Player is dead!")
if attacker.__class__.__name__=="Beginner":
attacker.health+=40
for card in attacker.card_repository.cards:
card.damage_points+=30
if enemy.__class__.__name__=="Beginner":
enemy.health+=40
for card in enemy.card_repository.cards:
card.damage_points+=30
get_bonus_health_points(attacker)
get_bonus_health_points(enemy)
for card in attacker.card_repository.cards:
if enemy.is_dead or attacker.is_dead:
return
enemy.take_damage(card.damage_points)
for card in enemy.card_repository.cards:
if attacker.is_dead or enemy.is_dead:
return
attacker.take_damage(card.damage_points)
def get_bonus_health_points(someone: Player):
bonus=0
for card in someone.card_repository.cards:
bonus+=card.health_points
someone.health+=bonus
|
993,809 | 0c442b5279a9ac5e810be6f5e774b866701e2569 | # Generated by Django 2.2.6 on 2019-11-02 03:19
from django.db import migrations
class Migration(migrations.Migration):
dependencies = [
('courses', '0012_delete_courseresource'),
]
operations = [
migrations.RenameModel(
old_name='SharedResource',
new_name='CourseResource',
),
]
|
993,810 | ed460893474f568e801e8d02e8351d56229ea6f8 | """
Useful subroutines for working with bokeh in general.
"""
from functools import wraps
def servable(title=None):
"""
Parametrizes a decorator which returns a document handle to be passed to bokeh.
Search "embed server" on bokeh.org to find more context.
Example:
```python
@servable()
def dummy(*args, **kwargs):
from hover.core.explorer import BokehCorpusAnnotator
annotator = BokehCorpusAnnotator(*args, **kwargs)
annotator.plot()
return annotator.view()
# in Jupyter
from bokeh.io import show, output_notebook
output_notebook()
show(dummy(*args, **kwargs))
# in <your-bokeh-app-dir>/main.py
from bokeh.io import curdoc
doc = curdoc()
dummy(*args, **kwargs)(doc)
# embedding a bokeh server
from bokeh.server.server import Server
app_dict = {
'my-app': dummy(*args, **kwargs),
'my-other-app': dummy(*args, **kwargs),
}
server = Server(app_dict)
server.start()
```
"""
def wrapper(func):
@wraps(func)
def wrapped(*args, **kwargs):
def handle(doc):
"""
Note that the handle must create a brand new bokeh model every time it is called.
Reference: https://github.com/bokeh/bokeh/issues/8579
"""
bokeh_model = func(*args, **kwargs)
doc.add_root(bokeh_model)
doc.title = title or func.__name__
return handle
return wrapped
return wrapper
|
993,811 | 56c6eca440d4b85f960378a4739f31b6fd4b30bf | # -*- coding: utf-8 -*-
"""
@author:XuMing(xuming624@qq.com)
@description:
"""
import numpy as np
from text2vec.utils.rank_bm25 import BM25Okapi
from text2vec.utils.tokenizer import Tokenizer
from text2vec.utils.log import logger
class BM25(object):
def __init__(self, corpus):
"""
Search sim doc with rank bm25
:param corpus: list of str.
A list of doc.(no need segment, do it in init)
"""
self.corpus = corpus
self.corpus_seg = None
self.bm25_instance = None
self.tokenizer = Tokenizer()
def init(self):
if not self.bm25_instance:
if not self.corpus:
logger.error('corpus is None, set corpus with documents.')
raise ValueError("must set corpus, which is documents, list of str")
if isinstance(self.corpus, str) or not hasattr(self.corpus, '__len__'):
self.corpus = [self.corpus]
self.corpus_seg = {k: self.tokenizer.tokenize(k) for k in self.corpus}
self.bm25_instance = BM25Okapi(corpus=list(self.corpus_seg.values()))
def get_similarities(self, query, n=5):
"""
Get similarity between `query` and this docs.
:param query: str
:param n: int, num_best
:return: result, dict, float scores, docs rank
"""
scores = self.get_scores(query)
rank_n = np.argsort(scores)[::-1]
if n > 0:
rank_n = rank_n[:n]
return [self.corpus[i] for i in rank_n]
def get_scores(self, query):
"""
Get scores between query and docs
:param query: input str
:return: numpy array, scores for query between docs
"""
self.init()
tokens = self.tokenizer.tokenize(query)
return self.bm25_instance.get_scores(query=tokens)
|
993,812 | 2b62b08594ed4fed564b5152d8cc6d17da233423 | import unittest
import numpy as np
from sklearn.utils._testing import assert_array_almost_equal
from maze import MazeEnvSample3x3, MazeEnvSpecial4x4, MazeEnvSpecial5x5
from ipendulum import InvertedPendulumEnv
from qlearning import QTableLearning
from value_iteration import ValueIteration
from policy_learning import CELearning
class TestMaze(unittest.TestCase):
def test_3x3_maze(self):
env = MazeEnvSample3x3()
current_state, _, _, _ = env.step(1) # right
self.assertEqual(current_state, [0, 1])
current_state, _, _, _ = env.step(3) # down
self.assertEqual(current_state, [1, 1])
current_state, _, _, _ = env.step(3) # down
self.assertEqual(current_state, [2, 1])
current_state, reward, done, _ = env.step(1) # right
self.assertEqual(current_state, [2, 2])
self.assertTrue(done)
self.assertTrue(reward > 0)
env.reset()
self.assertEqual(env.state, [0, 0])
def test_4x4_maze(self):
env = MazeEnvSpecial4x4()
env.step(1) # right
current_state, reward, done, _ = env.step(3)
self.assertTrue(reward < 0)
self.assertTrue(done)
def test_3x3_maze_value_iteration(self):
env = MazeEnvSample3x3()
alg = ValueIteration(env, max_iter=90)
alg.train()
expected_values = np.array([[2.048, 2.56, 3.2], [2.56, 3.2, 4], [3.2, 4, 5]])
# expected values are solved by Bell equation x = 1 + 0.8 * x for V[2, 2] = 5, etc..
assert_array_almost_equal(alg.values, expected_values)
done_cnt = 0
current_state = env.reset()
while True:
action = alg.predict(current_state)
current_state, reward, done, _ = env.step(action)
if done:
break
done_cnt += 1
self.assertEqual(done_cnt, 3)
self.assertEqual(reward, 1)
def test_4x4_maze_value_iteration(self):
env = MazeEnvSpecial4x4()
alg = ValueIteration(env)
alg.train()
done_cnt = 0
current_state = env.reset()
while True:
action = alg.predict(current_state)
current_state, reward, done, _ = env.step(action)
if done:
break
done_cnt += 1
self.assertEqual(done_cnt, 5)
self.assertEqual(reward, 4)
def test_5x5_maze_value_iteration(self):
env = MazeEnvSpecial5x5()
alg = ValueIteration(env)
alg.train()
done_cnt = 0
current_state = env.reset()
while True:
action = alg.predict(current_state)
current_state, reward, done, _ = env.step(action)
if done:
break
done_cnt += 1
self.assertEqual(done_cnt, 15)
self.assertEqual(reward, 1)
class TestPendulum(unittest.TestCase):
def test_model(self):
env = InvertedPendulumEnv()
env.step(1) # right force
state, _, _, _ = env.step(1)
self.assertTrue(state[0] > 0)
self.assertTrue(state[2] > 0)
self.assertTrue(state[3] > 0)
env.step(0)
env.step(0)
env.step(0)
state, _, done, _ = env.step(0)
self.assertTrue(state[2] < 0)
self.assertTrue(state[3] < 0)
self.assertFalse(done)
def test_cross_entropy_learning(self):
env = InvertedPendulumEnv()
alg = CELearning(env)
alg.train()
done_cnt = 0
current_state = env.reset()
while True:
action = alg.predict(current_state)
# action = env.action_space.sample()
current_state, reward, done, _ = env.step(action)
if done or done_cnt == 1000:
break
done_cnt += 1
self.assertTrue(done_cnt > 25)
if __name__ == '__main__':
unittest.main()
|
993,813 | a8d83f1e1c516ab94e2e4dd0062986439c1b2eed | import pandas as pd
import numpy as np
import matplotlib.pyplot as plt
import matplotlib as mpl
import seaborn as sns
from sklearn.preprocessing import MinMaxScaler
from sklearn.model_selection import train_test_split
from sklearn.metrics import mean_absolute_error, mean_squared_error, explained_variance_score
from tensorflow.keras.models import Sequential, load_model
from tensorflow.keras.layers import Dense
import tensorflow as tf
mpl.rcParams['patch.force_edgecolor'] = True
sns.set()
sns.set_style('whitegrid')
#real world data
df = pd.read_csv('DATA/kc_house_data.csv')
print(df.head())
#print(df.isnull().sum()) #no missing data!
#print(df.describe().transpose())
plt.figure(figsize = (10,6))
sns.distplot(df['price'])
plt.show()
sns.countplot(df['bedrooms'])
plt.show()
print(df.corr()['price'].sort_values()) # - checking for correlations
plt.figure(figsize = (10,6))
sns.scatterplot(x = 'price', y = 'sqft_living', data = df )
plt.show()
plt.figure(figsize = (10,6))
sns.boxplot(x='bedrooms', y = 'price', data=df)
plt.show()
plt.figure(figsize = (12,9))
sns.scatterplot(x='price', y='long', data =df)
plt.show()
plt.figure(figsize = (12,9))
sns.scatterplot(x='price', y='lat', data =df)
plt.show()
#at a certain value of latitiude and longitude, the price is very expensive
bottom_ninety_nine_perc = df.sort_values('price', ascending=False).iloc[216:]
plt.figure(figsize=(10,6))
sns.scatterplot(x='long', y='lat', data = bottom_ninety_nine_perc, hue = 'price', edgecolor=None, alpha=0.2, palette='RdYlGn')
plt.show()#we basically recreated the map of King County, Seattle
sns.boxplot(x='waterfront', y='price', data = df)
plt.show()
#now we drop unnecesary data
df = df.drop('id', axis = 1)
df['date'] = pd.to_datetime(df['date']) #now we can extract month or year automatically
df['year'] = df['date'].apply(lambda date: date.year)
df['month'] = df['date'].apply(lambda date: date.month)
plt.figure(figsize=(10,6))
sns.boxplot(x='month', y='price', data = df)
plt.show()
#or we could just do
df.groupby('month').mean()['price'].plot()
plt.show()
df.groupby('year').mean()['price'].plot()
plt.show()
df=df.drop('date', axis = 1)
df = df.drop('zipcode', axis=1)
#data preprocessing
X = df.drop('price', axis = 1).values
y = df['price'].values
X_train, X_test, y_train, y_test = train_test_split( X, y, test_size=0.3, random_state=101)
print(X_train.shape)
scaler = MinMaxScaler()
X_train = scaler.fit_transform(X_train)
X_test = scaler.transform(X_test)
model = Sequential()
model.add(Dense(19, activation='relu'))
model.add(Dense(19, activation='relu'))
model.add(Dense(19, activation='relu'))
model.add(Dense(19, activation='relu'))
model.add(Dense(1))
model.compile(optimizer='adam', loss='mse')
model.fit(x=X_train, y = y_train, validation_data =(X_test, y_test) ,batch_size=128, epochs = 400)
#model evaluations and predictions
loss_df = pd.DataFrame(model.history.history)
print(loss_df) #now we get the loss on the training set and val_loss - the lost on the test set in order to see if you're over fitting
loss_df.plot()
plt.show()
predictions = model.predict(X_test)
print(np.sqrt(mean_squared_error(y_test, predictions)))
print(mean_absolute_error(y_test,predictions))
print(explained_variance_score(y_test, predictions))
plt.scatter(y_test, predictions)
plt.plot(y_test, y_test, 'r')
plt.show()
#now we predict
single_house = df.drop('price', axis = 1).iloc[0]
single_house = scaler.transform(single_house.values.reshape(-1,19))
print(model.predict(single_house) #we are overshooting the value by a little which isn't too bad given the price range and number of houses
|
993,814 | af403de35b3e9f2132bb98b27d2c4a7a1f4c93f3 | #! /bin/python
import argparse
parser = argparse.ArgumentParser(description='Search for words inclupding partial words')
parser.add_argument('snippet', help='partial or complete string to search for in words file')
args = parser.parse_args()
snippet = args.snippet.lower()
words = open('/usr/share/dict/words').readlines()
# first value in the list is retunred if the other items are met.
# this single line replaces the 5 lines beneith it.
print([word.strip() for word in words if snippet in word.lower()])
#matches = []
#for word in words:
# if snippet in word.lower():
# matches.append(word)
#print(matches)
|
993,815 | 4be3b80b8aa9f447bfff1019e12b51334fd35d65 | import csv
print ("Stand by please, building list......")
#r = raw_input("Enter Yes or No: ") # -- ASk if you want Yes people or No people
ylst = list() # Create empty list for Yes People
nlst = list() # Create empty list for No people
d = dict() # Create a Dictionary if needed
with open('vcwater.csv', 'rb') as f:
reader = csv.reader(f)
# the code below will sort the rows into two list ylst and nlst based on what they start with
for row in f:
row = row.rstrip()
if row.startswith("Yes"):
ylst.append(row[4:])
if row.startswith("No"):
nlst.append(row[3:])
#this is a test line
print ylst
print nlst
|
993,816 | 40577eec44d287fa651baab0519608be2ca54318 | import sys
#sys.stdin = open('input.txt','r')
k,n = map(int, input().split())
a = [int(input()) for _ in range(k)]
def Count(num):
cnt = 0
for i in a:
cnt += (i // num)
return cnt
res = 0
left = 1
right = max(a)
while left <= right:
mid = (left+right)//2
if Count(mid)>= n:
result = mid
left = mid+1
else:
right = mid-1
print(result)
|
993,817 | 39e400ba3465dc617761b6e0fd06d8eb305a2441 | import unittest
import shutil
import os
import tempfile
from trovenames.index import TroveIndexBuilder, TroveIndex, TroveSwiftIndexBuilder, TroveSwiftIndex
class testIndex(unittest.TestCase):
def setUp(self):
pass
def test_create_index(self):
"""Create an index"""
indexfile = tempfile.mktemp()
self.addCleanup(os.unlink, indexfile)
index = TroveIndexBuilder("test/short.dat", out=indexfile)
# read the index file that was created
with open(indexfile, 'r+b') as fd:
indextext = fd.read()
indexlines = indextext.split('\n')
# 11 lines includes on blank line at the end
self.assertEquals(11, len(indexlines))
del indexlines[10]
# check the first character of each line
docs = [line[0] for line in indexlines]
self.assertEquals(['1', '2', '3', '4', '5', '6', '7', '8', '9', '1'], docs)
# check some lines from the index
ref = "1, 0, 31, test/short.dat"
self.assertEqual(ref, indexlines[0])
ref = "10, 279, 32, test/short.dat"
self.assertEqual(ref, indexlines[9])
def test_read_index(self):
"""Read an index from a file"""
indexfile = tempfile.mktemp()
self.addCleanup(os.unlink, indexfile)
TroveIndexBuilder("test/short.dat", out=indexfile)
index = TroveIndex()
index.reload(indexfile)
docs = sorted([doc for doc in index.documents])
self.assertEquals(10, len(docs))
self.assertEquals([1, 2, 3, 4, 5, 6, 7, 8, 9, 10], docs)
doc = index.get_document(1)
ref = {u"id":"1",u"titleName":u"Hello"}
self.assertNotEquals(None, doc, "Document not found for id 1")
self.assertDictEqual(ref, doc)
doc = index.get_document(10)
ref = {"id":"10","titleName":"Hello"}
self.assertNotEquals(None, doc)
self.assertDictEqual(ref, doc)
def test_create_index_swift(self):
"""Create an index of a swift document"""
indexfile = tempfile.mktemp()
self.addCleanup(os.unlink, indexfile)
index = TroveSwiftIndexBuilder("short.dat", out=indexfile)
# read the index file that was created
with open(indexfile, 'r+b') as fd:
indextext = fd.read()
indexlines = indextext.split('\n')
# 11 lines includes on blank line at the end
self.assertEquals(11, len(indexlines))
del indexlines[10]
# check the first character of each line
docs = [line[0] for line in indexlines]
self.assertEquals(['1', '2', '3', '4', '5', '6', '7', '8', '9', '1'], docs)
# check some lines from the index
ref = "1, 0, 31, short.dat"
self.assertEqual(ref, indexlines[0])
ref = "10, 279, 32, short.dat"
self.assertEqual(ref, indexlines[9])
def test_read_index_swift(self):
"""Read an index on a swift document from a file"""
indexfile = tempfile.mktemp()
self.addCleanup(os.unlink, indexfile)
TroveSwiftIndexBuilder("short.dat", out=indexfile)
index = TroveSwiftIndex()
index.reload(indexfile)
docs = sorted([doc for doc in index.documents])
self.assertEquals(10, len(docs))
self.assertEquals([1, 2, 3, 4, 5, 6, 7, 8, 9, 10], docs)
doc = index.get_document(1)
ref = {"id":"1","titleName":"Hello"}
self.assertDictEqual(ref, doc)
doc = index.get_document(10)
ref = {"id":"10","titleName":"Hello"}
self.assertNotEquals(None, doc)
self.assertDictEqual(ref, doc)
if __name__=='__main__':
unittest.main()
|
993,818 | 345d321e1eb7694f172905f0bb7fe810c2d80f3f | #!/usr/bin/env python
# coding: utf-8
print(list)
print(tuple)
print(set)
print(dict)
print({a for a in dir(__builtins__) if isinstance(getattr(__builtins__, a), type)
and a.islower()} - {'__loader__'})
|
993,819 | 8a8e8785b073a65aaaf381f38cb17eed5b5df112 | class Node:
def __init__(self, data):
self.data = data
self.next = None
def insert_at(prev_node, data):
new_node = Node(data)
new_node.next = prev_node.next
prev_node.next = new_node
class LinkedList:
def __init__(self):
self.head = None
def print(self):
cur_list = self.head
while cur_list:
print(cur_list.data)
cur_list = cur_list.next
def append(self, data):
new_node = Node(data)
if self.head is None:
self.head = new_node
return
next_node = self.head
while next_node.next:
next_node = next_node.next
next_node.next = new_node
def prepend(self, data):
new_node = Node(data)
new_node.next = self.head
self.head = new_node
def delete_node(self, key):
prev = None
cur = self.head
while cur and cur.data != key:
prev = cur
cur = cur.next
prev.next = cur.next
def delete_node_at_position(self, pos):
prev = None
cur = self.head
c = 0
while cur and c != pos:
prev = cur
cur = cur.next
c += 1
prev.next = cur.next
def reverse_list(self):
prev = None
cur = self.head
while cur:
nxt = cur.next
print(nxt.data)
cur.next = prev
print(cur.data)
prev = cur
print(prev.data)
cur = nxt
print(cur.data)
self.head = prev
def swap_nodes(self, node1, node2):
prev1 = None
current_node = self.head
while current_node and current_node.data != node1:
prev1 = current_node
current_node = current_node.next
prev2 = None
current_node1 = self.head
while current_node1 and current_node1.data != node2:
prev2 = current_node1
current_node1 = current_node1.next
if prev1:
prev1.next = current_node1
else:
self.head = current_node1
if prev2:
prev2.next = current_node
else:
self.head = current_node
current_node.next, current_node1.next = current_node1.next, current_node.next
ls = LinkedList()
ls.append("A")
ls.append("B")
ls.append("C")
ls.append("D")
ls.swap_nodes('A', 'C')
ls.print()
|
993,820 | eaadf8deeaf606108a39eab319f9169b0aa549a7 | # coding:utf-8
import socket
import threading
# 处理客户端请求的任务
def handle_client_req(new_socket, ip_port):
print('客户端的ip和端口号为：', ip_port)
while True:
# 5.循环接收容户端的数据
recv_data = new_socket.recv(1024)
if recv_data:
recv_content = recv_data.decode('gbk')
print('接收到的客户端数据为：', recv_content, ip_port)
print('接收数据长度为：', len(recv_data))
# 6.发送数据到客户端
send_content = '问题正在处理中。。。'
send_data = send_content.encode('gbk')
new_socket.send(send_data)
else:
# 客户端关闭连接
print('客户端下线了：', ip_port)
break
# 关闭服务端与客户端套接字，表示和客户端终止通信
new_socket.close()
if __name__ == '__main__':
# 1.创建tcp服务端套接字
# AF_INET:ipv4，AF_INET6:ipv6
tcp_server_socket = socket.socket(socket.AF_INET, socket.SOCK_STREAM)
# 设置端口号复用，表示的意思︰服务端程序退出端口号则立即释放
# 1.soL_SOCKET: 表示当前套接字
# 2.so_REUSEADDR: 表示复用端口号的选项
# 3.True: 确定复用
tcp_server_socket.setsockopt(socket.SOL_SOCKET, socket.SO_REUSEADDR, True)
# 2.绑定端口号
# 第一个参数表示ip地址，一般不用指定﹐表示本机的任何一个ip都可
# 第二个参数表示端口号
tcp_server_socket.bind(('', 9091))
# 3.设置监听
# 128:表示最大等待建立连接的个数
tcp_server_socket.listen(128)
while True:
# 4.循环等待接受容户端的连接请求
# 注意︰每次当客户端和服务端建立连接成功会返回一个新的套接字
# tcp_server_socket只负责等待接收客户端的连接请求﹐收发消息不使用该套接字
new_socket, ip_port = tcp_server_socket.accept()
# 代码执行到此，说明客户端和服务端建立连接成功
sub_thread = threading.Thread(target=handle_client_req, args=(new_socket, ip_port))
# 设置守护主线程，主线程退出子线程直接销毁
sub_thread.setDaemon(True)
sub_thread.start()
# 7.关闭套接字，表示服务端以后不再等待接受客户端的连接请求
# tcp_server_socket.close() # 因为服务端的程序要一直运行，可以不用关闭
|
993,821 | 75e1a75c3e49c132e726868e0f2d06a0bccbf211 | # https://www.hackerrank.com/challenges/maximize-it/problem
from typing import List
def just_maximize_it(lists: List[List[int]], m: int):
s_values = {0}
for list_ in lists:
new_s_values = set()
for prev_val in s_values:
for num in list_:
new_s_values.add((num**2 + prev_val) % m)
s_values = new_s_values
return max(s_values)
def test_just_maximize_it():
lists1 = [[1]]
m1 = 2
assert just_maximize_it(lists1, m1) == 1
m2 = 1_000
lists2 = [
[2, 5, 4],
[3, 7, 8, 9],
[5, 5, 7, 8, 9, 10],
]
s2 = 206
assert just_maximize_it(lists2, m2) == s2
m3 = 6
lists3 = [[2], [2]]
s3 = 2
assert just_maximize_it(lists3, m3) == s3
lists4 = [[1]]
m4 = 1
assert just_maximize_it(lists4, m4) == 0
def parse_maximize_it_input():
k_and_m_str = input()
k, m = map(int, k_and_m_str.split(' '))
lists = []
for i in range(k):
lists.append(list(map(int, input().split(' '))))
return lists, m
def maximize_it():
lists, m = parse_maximize_it_input()
return just_maximize_it(lists, m)
if __name__ == '__main__':
test_just_maximize_it()
# print(parse_maximize_it_input()) |
993,822 | e3bcac08a4f082453df8c5db62a0513f20d54a9d | secreat_number=8
guess_limit=3
number_of_tries=0
while number_of_tries < guess_limit:
number=input('Please guess the number ')
if secreat_number == int(number):
print('You guessed it right !')
break
else:
print('Try Again !!')
number_of_tries+=1
else:
print('Sorry you failed !!')
print('DONE')
|
993,823 | 6dcbede79e1d4884dc729d45a84676759eae1084 | from django.conf.urls import patterns, url
from .views import LoginView, RegisterView
urlpatterns = patterns('',
url(r'^login$', LoginView.as_view(), name='login'),
url(r'^salir$', 'users.views.salir', name='salir'),
url(r'^register$', RegisterView.as_view(), name='register'),
) |
993,824 | 06cdf88de856399ffcc01ed5a4717fac841d0f25 | import pickle
import numpy as np
from scipy.sparse import csr_matrix
import pandas as pd
def load_dict1():
with open('model1/id_to_itemid.p', 'rb') as fp:
id_to_itemid = pickle.load(fp)
with open('model1/id_to_userid.p', 'rb') as fp:
id_to_userid = pickle.load(fp)
with open('model1/itemid_to_id.p', 'rb') as fp:
itemid_to_id = pickle.load(fp)
with open('model1/userid_to_id.p', 'rb') as fp:
userid_to_id = pickle.load(fp)
return id_to_itemid, id_to_userid, itemid_to_id, userid_to_id
def load_model1():
return pickle.load(open('model1/model1.pickle', 'rb'))
if __name__ == '__main__':
id_to_itemid, id_to_userid, itemid_to_id, userid_to_id = load_dict1()
test_item_ids = np.array([
itemid_to_id['00701100130290059612'],
itemid_to_id['00707010220059611241'],
itemid_to_id['020041235Д5019522251'],
itemid_to_id['02204023026794451530'],
])
model1 = load_model1()
user_feat_lightfm = pd.read_csv('model1/user_feat_lightfm.csv', index_col=0)
predictions = model1.predict(user_ids=0, item_ids=test_item_ids,
user_features=csr_matrix(user_feat_lightfm.values).tocsr(),
num_threads=4)
print(predictions)
print(1) |
993,825 | e5aceeee28c5443637ed1e566c0d6ebe439d5b80 | # Generated by Django 2.2 on 2020-12-07 13:58
from django.db import migrations
class Migration(migrations.Migration):
dependencies = [
('store', '0012_auto_20201207_2116'),
]
operations = [
migrations.RenameField(
model_name='commoditytoshop',
old_name='shop_id',
new_name='shop_owner',
),
]
|
993,826 | 14c47b4e4e2d6709835b35017d65f6585ed628a7 | #!/usr/bin/python
import sys
import json
class LexicalEntry:
def __init__(self, l, f, t):
self.lexeme=l.strip()
self.formula=f.strip()
self.type=t.strip()
out = open(sys.argv[2],'w')
with open(sys.argv[1]) as f:
for line in f:
tokens = line.split("\t")
if len(tokens) > 2:
continue
if(tokens[0] == "loc_city"):
index = tokens[1].rfind('.')
citystate = tokens[1][index+1:]
city = citystate[0:citystate.rfind('_')]
city = city.replace('_',' ').strip()
entry = LexicalEntry(city, tokens[1], "fb:en.city")
out.write(json.dumps(entry.__dict__)+'\n')
elif (tokens[0] == "loc_state"):
index = tokens[1].rfind('.')
state = tokens[1][index+1:].strip()
state = state.replace('_',' ').strip()
entry = LexicalEntry(state, tokens[1], "fb:en.state")
out.write(json.dumps(entry.__dict__)+'\n')
elif tokens[0] == "loc_river":
index = tokens[1].rfind('.')
river = tokens[1][index+1:].strip()
river = river.replace('_',' ').strip()
entry = LexicalEntry(river+" river", tokens[1], "fb:en.river")
out.write(json.dumps(entry.__dict__)+'\n')
elif (tokens[0] == "loc_place"):
index = tokens[1].rfind('.')
place = tokens[1][index+1:].strip()
place = place.replace('_',' ').strip()
entry = LexicalEntry(place, tokens[1], "fb:en.place")
out.write(json.dumps(entry.__dict__)+'\n')
elif (tokens[0] == "loc_lake"):
index = tokens[1].rfind('.')
lake = tokens[1][index+1:].strip()
lake = lake.replace('_',' ').strip()
if not 'lake' in lake:
lake = lake + " lake"
entry = LexicalEntry(lake, tokens[1], "fb:en.lake")
out.write(json.dumps(entry.__dict__)+'\n')
elif (tokens[0] == "loc_mountain"):
index = tokens[1].rfind('.')
mountain = tokens[1][index+1:].strip()
mountain = mountain.replace('_',' ').strip()
entry = LexicalEntry("mount " + mountain, tokens[1], "fb:en.mountain")
out.write(json.dumps(entry.__dict__)+'\n')
elif (tokens[0] == "loc_country"):
index = tokens[1].rfind('.')
country = tokens[1][index+1:].strip()
country = country.replace('_',' ').strip()
entry = LexicalEntry(country, tokens[1], "fb:en.country")
out.write(json.dumps(entry.__dict__)+'\n')
out.close()
|
993,827 | 57d8d04f999bce6bb6107373756dabb157985e23 | #!/usr/bin/env python3
from app import app
from app import views
app.run()
|
993,828 | 15f93f7d7c65d78d8e959db6e53eb3c5d5b94a9c | {
'name':'Last Mile theme',
'description': 'A starting theme ex.',
'version':'0.1.0',
'author':'Samuel Smith',
'data': [ 'views/layout.xml' ],
'category': 'Theme/Creative',
'depends': ['website', 'website_theme_install', 'sale'],
} |
993,829 | f1c4e8eae52a67eda043ccf1d0632f4689725398 | import logging , sys
def get_logger(name, flog_name, stdout=True):
logger = logging.getLogger(name)
h_file = logging.FileHandler(flog_name)
h_file.setFormatter(logging.Formatter('%(message)s'))
logger.addHandler(h_file)
logger.setLevel(logging.INFO)
if stdout:
h_stdout = logging.StreamHandler(sys.stdout)
logger.addHandler(h_stdout)
return logger
if __name__ == '__main__':
logger1 = get_logger('log1', 'log1.log')
logger2 = get_logger('log2', 'log2.log', False)
logger1.info('test')
logger2.info('test') |
993,830 | cb8bedae198a5ea4b5a0f7b67ee4d9cf755cff35 | import sys
sys.path.append(r"C:\Users\evans\Google Drive\Code\Python\Project Euler\functions")
|
993,831 | 6372dba59444526fa39d0d1301da2cde28a87812 | from django.http import Http404
from rest_framework.response import Response
from rest_framework.mixins import ListModelMixin
from rest_framework.decorators import list_route
class AutoCompletionMixin(ListModelMixin):
"""
Add a route to the ViewSet to get a list of completion suggestion.
"""
@list_route()
def autocomplete(self, request, **kwargs):
try:
qp = request.query_params
except AttributeError:
# restframework 2
qp = request.QUERY_PARAMS
field_name = qp.get('f', None)
query = qp.get('q', '')
try:
data = self.model.es.complete(field_name, query)
except ValueError:
raise Http404("field {0} is either missing or "
"not in Elasticsearch.completion_fields.")
return Response(data)
|
993,832 | 0e90afd37a17d4e2815c8e821386c4320988cd0f | from pages.StartSurvey import StartSurvey
import unittest
import pytest
import utilities_config.custom_logger as cl
from utilities_config.Input import *
import logging
@pytest.mark.usefixtures("start_survey")
class TestStartSurvey(unittest.TestCase):
log = cl.customLogger(logging.DEBUG)
@pytest.fixture(autouse=True)
def objectSetup(self, start_survey):
self.ss = StartSurvey(start_survey)
# @pytest.mark.run(order=3)
# def test_start_survey(self):
# print("start survey creation")
# self.ss.start_survey_operation(new_survey_title, new_page_title)
# result = self.ss.surveyoperation_successful()
# assert result == True
@pytest.mark.run(order=3)
def test_survey_operation(self):
print("survey operation started")
self.ss.start_survey_operation(new_survey_title, new_page_title)
result = self.ss.valid_page_title()
assert result == str(new_page_title)
# @pytest.mark.run(order=3)
# def test_survey_title(self):
# print(" survey title")
# self.ss.survey_title_operation(new_survey_title)
# result = self.ss.valid_survey_title()
# assert result == True
#
# @pytest.mark.run(order=4)
# def test_start_survey(self):
# print(" page title")
# self.ss.page_title_operation(new_page_title)
# result = self.ss.valid_page_title()
# assert result == True
|
993,833 | 2468bbd5e9e0f73cd86c1c36307f2e7c8e3bf341 | def findrepn(li):
li.sort()
rep=[]
a=len(li)
for i in range(1,a):
if li[i]==li[i-1] :
if li[i] in rep :
continue
rep.append(li[i])
print(rep)
def main():
try:
li=[]
a=int(input())
for i in range(a):
li.append(int(input()))
findrepn(li)
except:
print('Unique')
main() |
993,834 | 757eb5e64733a88d21177f259025f1dcadb1d2c2 | """
Lagrello API
API specification for Lagrello, a simple to use authentication service # noqa: E501
The version of the OpenAPI document: 1.0.0
Contact: support@lagrello.com
Generated by: https://openapi-generator.tech
"""
import unittest
import lagrello_client
from lagrello_client.api.internal_api import InternalApi # noqa: E501
class TestInternalApi(unittest.TestCase):
"""InternalApi unit test stubs"""
def setUp(self):
self.api = InternalApi() # noqa: E501
def tearDown(self):
pass
def test_tenants_cardtoken(self):
"""Test case for tenants_cardtoken
Returns token that is used by stripe to save card number. # noqa: E501
"""
pass
if __name__ == '__main__':
unittest.main()
|
993,835 | 324f7b12060fb8eb24f9b5c072ce84c54c5c9c9d | from rest_framework import serializers
from .models import MuscleSubportion, Muscle, MuscleGrouping
class MuscleSubportionSerializer(serializers.HyperlinkedModelSerializer):
class Meta:
model = MuscleSubportion
fields = ['name']
class MuscleSerializer(serializers.HyperlinkedModelSerializer):
sub = MuscleSubportionSerializer(source='subportions', many=True)
class Meta:
model = Muscle
fields = ['name', 'sub']
class MuscleGroupingSerializer(serializers.HyperlinkedModelSerializer):
sub = MuscleSerializer(source='muscles', many=True)
class Meta:
model = MuscleGrouping
fields = ['url', 'name', 'snames', 'sub']
extra_kwargs = {
'url': {'view_name': 'muscle-detail', 'lookup_field': 'pk'}
}
|
993,836 | 388bb0519ba623b9d99f7600eea064cef188af6e | #!/usr/bin/env python3
"""
Usage: ./day5-q6.py <histone_file1> <histone_file2> ... <histone_file4> <samples.ctab>
"""
import sys
import os
import pandas as pd
import numpy as np
import statsmodels.formula.api as sm
import matplotlib.pyplot as plt
means = {}
h3k4me1 = sys.argv[1].split(os.sep)[-1]
m1 = pd.read_table(sys.argv[1], sep = "\t").iloc[:, 4]
h3k4me3 = sys.argv[2].split(os.sep)[-1]
m2 = pd.read_table(sys.argv[2], sep = "\t").iloc[:, 4]
h3k9ac = sys.argv[3].split(os.sep)[-1]
m3 = pd.read_table(sys.argv[3], sep = "\t").iloc[:, 4]
h3k27ac = sys.argv[4].split(os.sep)[-1]
m4 = pd.read_table(sys.argv[4], sep = "\t").iloc[:, 4]
h3k27me3 = sys.argv[5].split(os.sep)[-1]
m5 = pd.read_table(sys.argv[5], sep = "\t").iloc[:, 4]
fpkm_name = sys.argv[6].split(os.sep)[-1]
fpkm = pd.read_csv(sys.argv[6], sep = "\t").loc[:, "FPKM"]
fpkm_log = np.log(fpkm + 1)
means = {h3k4me1 : m1, h3k4me3 : m2, h3k9ac : m3, h3k27ac: m4, h3k27me3: m5, "fpkm": fpkm_log}
means_df = pd.DataFrame(means)
means_df.columns = ["h3k4me1", "h3k4me3", "h3k9ac", "h3k27ac", "h3k27me3", "fpkm"]
result = sm.ols(formula = 'fpkm ~ h3k4me1 + h3k4me3 + h3k9ac + h3k27ac + h3k27me3', data = means_df).fit()
fig, ax = plt.subplots()
ax.hist(result.resid, bins=1000)
ax.set_title("Residuals of Model")
ax.set_xlabel("Residuals")
ax.set_ylabel("Frequency")
ax.set_xlim(left=-2.5, right=5)
fig.savefig("893_resid_log.png")
plt.close(fig) |
993,837 | 41ef43560fa6060b9ab0e8500ff736db2e66cf88 | """
Auto catalogue.
Developed by Grigorev Albert.
"""
from CarClass import Car
def accept_command():
""" Asks for the command number. """
# TODO
def add(lst):
""" Asks user about the car. """
# TODO
def read():
""" Reading from the file. """
# TODO
def write(lst):
""" Writing to the file. """
# TODO
def show(lst):
""" Print objects. """
# TODO
def main():
""" Main function. """
# TODO
if __name__ == '__main__':
main()
|
993,838 | 1237b7c27d6b5796d71bc172ff52284415612ec7 | #coding=utf8
# auther:shixingjian time:2020/07/15
from poWeb.BasePage import BasePage
from selenium.webdriver.common.by import By
class LoginPage(BasePage):
def __init__(self):
super().__init__()#执行父类的构造方法
self.username_locator = (By.ID,'username')#元素定位方法
self.password_locator = (By.ID,'password')
self.loginbutton_locator = (By.TAG_NAME,'button')
#抽离具体的元素控件
def username_element(self):
return self.get_Element(self.username_locator)
#密码输入框元素
def password_element(self):
return self.get_Element(self.password_locator)
#登录按钮
def loginbutton_element(self):
return self.get_Element(self.loginbutton_locator)
class LoginPageAction(LoginPage):#执行登录动作
def login(self):
self.username_element().clear()
self.password_element().clear()
self.username_element().send_keys('auto')
self.password_element().send_keys('sdfsdfsdf')
self.loginbutton_element().click()
if __name__ == '__main__':
LoginPageAction().login() |
993,839 | ddacbd6e947c1c9d8316494229996613e98700a8 | dict1 = {'Name': 'Nataraj', 'colour': 'black'}
print(dict1.items())
k = dict1.keys()
courses = {'raja': ["math", "scenice", "social"],
'pradheep': ["tamil", "programming"]}
keys = courses.keys()
for name in keys:
print("The course taken by", name)
for listofcourse in courses[name]:
print(listofcourse)
|
993,840 | 1fc7205c3a27a01fab02aad2b08d2e12b98cb350 | from moddemodfunc import *
def generateIQ(Nt, N, mod, tx_mode):
# Nt = no. of antennas
# N = number of samples
# mod = modulation scheme, BPSK, QPSK, ..., 1024QAM
# BPSK=1, QPSK=2, ... 1024QAM=10
# TODO: improve the time-consuming loop
c = np.asmatrix([[np.complex(0,0)]*N]*Nt)
if tx_mode == 0: #sending same data in all antennas
for j in range(N):
# get random bits
b = np.random.randint(2, size=mod)
# encode bits into samples
temp_c = qammod(b, mod)
for i in range(Nt):
c[i,j] = temp_c
elif tx_mode == 1: #sending different data in each antenna
for i in range(Nt):
for j in range(N):
# get random bits
b = np.random.randint(2, size=mod)
# encode bits into samples
c[i,j] = qammod(b, mod)
# Normalize the signal to unit power
c = c/np.sqrt(normFactor(mod))
return c
|
993,841 | e023f5d95da5f8d0baa3b0623fd5e411f340b0ec | A=int(input())
B=int(input())
C=int(input())
D=int(input())
tickets = [A,B]
kippu = [C,D]
print(min(tickets)+min(kippu)) |
993,842 | c75ede00a6c1a1a3b8d03b7a5aa49cec5c1c8f8b | __version__ = '1.1.0'
from .urljects import url, URLView, url_view, view_include
from .routemap import RouteMap
from .patterns import *
|
993,843 | b4f7a9b4b43a9a34e7c64d52d294c805df87c1ed | class Solution:
def lengthOfLongestSubstring(self, s: str) -> int:
d = {}
start_point = -1
answer = 0
for idx, c in enumerate(s):
if c in d:
start_point = max(start_point, d[c]) #update for start_point(boundary)
d[c] = idx
#update for length
answer = max(answer, idx-start_point)
return answer
|
993,844 | 21ecc7253fd11ab193cee8ad85c881c9bc996008 | # 读取文件，传入文件名和标示符
f = open('D:/test.txt', 'r') # 标示符'r'表示读,文件路径用"/"或者"\\"
# 如果文件不存在，open()函数会抛出IOError错误，并给出错误码和详细信息
# 若文件打开成功，接下来调用read()方法一次读取文件的全部内容，用str对象表示
a = f.read()
print(a)
# 最后调用close()关闭文件。文件使用完后必须关闭，因为文件对象会占用操作系统的资源，且操作系统同一时间能打开的文件数量也是有限的
f.close()
# 由于文件读写时都有可能产生IOError，一旦出错，f.close()就不会调用。
# 为了保证无论是否出错都能正确地关闭文件，可try...finally实现
try:
f = open('d:/test.txt', 'r') # 返回的f存在为空和不为空2种情况
print(f.read())
finally:
if f: # 这句的意思是f不为空
f.close()
# 每次都用try...finally太繁琐，Python引入了with语句自动调用close()方法
with open('d:/test.txt', 'r') as f:
print(f.read())
# with跟try...finally是一样的，代码更简洁，且不必调用close()方法
"""
调用read()会一次性读取文件的全部内容，如果文件有10G，内存就爆了
所以保险起见，可以反复调用read(size)方法，每次最多读取size个字节的内容
另外，调用readline()可以每次读取一行内容，调用readlines()一次读取所有内容并按行返回list
"""
# 若文件很小，read()一次性读取最方便；若不确定文件大小，反复调用read(size)比较保险；若是配置文件，调用readlines()最方便
with open('d:/test.txt') as t:
for line in t.readlines():
print(line.strip()) # 把末尾的'\n'删掉
"""
file-like Object
像open()函数返回的这种有个read()方法的对象，在Python中统称为file-like Object
除了file外，还可以是内存的字节流，网络流，自定义流等
file-like Object不要求从特定类继承，只要写个read()方法就行
StringIO就是在内存中创建的file-like Object，常用作临时缓冲
"""
"""
二进制文件
上面所讲的默认都是读取文本文件，且是utf-8编码的文本文件，
要读取二进制文件，如图片、视频等，用'rb'模式打开文件即可
"""
i = open('d:/test.jpg', 'rb')
print(i.read())
"""
字符编码
读取非utf-8编码的文本文件，需要给open()函数传入encoding参数
"""
# 例如，读取gbk编码的文件
s = open('d:/gbk.txt', 'r', encoding='gbk')
print(s.read())
"""
遇到有些编码不规范的文件，可能会遇到UnicodeDecodeError，因为在文本文件中可能夹杂了一些非法编码的字符
这种情况，open()函数还接收一个error参数，表示如果遇到编码错误后如何处理，最简单的方式就是直接忽略
"""
h = open('d:/gbk.txt', 'r', encoding='gbk', errors='ignore')
"""
写文件
写文件和读文件是一样的，唯一的区别是调用open()函数是，传入标示符'w'或者'wb'，表示写文本文件或二进制文件
"""
x = open('d:/test.txt', 'w')
x.write('you are nice')
x.close()
"""
可以反复调用write()来写入文件，然后务必要调用close()来关闭文件
当写文件时，操作系统往往不会立刻把数据写入磁盘，而是放到内存缓存起来，空闲的时候再慢慢写入。
只有调用close()方法时，操作系统才保证把没有写入的数据全部写入磁盘。
忘记调用close()的后果是数据可能只写了一部分到磁盘，剩下的丢失了。
所以，还是用with语句来得保险
"""
with open('d:/test.txt', 'w') as f: # 如果文件不存在，会自动创建，'w'表示写数据，写之前会清空原文件中的原有数据
f.write('you are lucky')
with open('d:/test.txt', 'a') as f: # 'a'表示append，即在原来文件内容后继续写数据（不清除原有数据）
f.write('you are lucky')
# 要写入特定编码的文本文件，给open()函数传入encoding参数，将字符串自动转换成指定编码
# 定位
f.seek(2, 0) # 第一个数字表示偏移量，第二个数字表示文件从哪里开始的位置
# 当前位置
f.tell() # 返回当前文件位置
"""
标示符 打开方式 指针位置
b 二进制模式
t 文字模式（默认）
+ 打开磁盘文件进行更新（读写）
r 只读（默认模式） 指针在文件开头，文件不存在则报错
rb 二进制格式只读 指针在文件开头，文件不存在则报错
r+ 可读可写 指针在文件开头，文件不存在则报错
rb+ 二进制格式可读可写 指针在文件开头，文件不存在则报错
w 只写 文件存在，则覆盖；不存在，则创建
wb 二进制格式只写 文件存在，则覆盖；不存在，则创建
w+ 可读可写 文件存在，则覆盖；不存在，则创建
wb+ 二进制格式可读可写 文件存在，则覆盖；不存在，则创建
a 追加 文件存在，指针在文件尾追加；不存在，则创建
ab 二进制格式追加 文件存在，指针在文件尾追加；不存在，则创建
a+ 可读可写 文件存在，指针在文件尾追加；不存在，则创建
ab+ 二进制格式可读可写 文件存在，指针在文件尾追加；不存在，则创建
"""
|
993,845 | 04ec3614136f91cc593974ac5610f6116d9e4338 | import abc
import warnings
from typing import Optional, Set
from observer.lib import Observer
class Subject(metaclass=abc.ABCMeta):
""" Abstract event producer. """
_observers: Set[Observer] = set()
def attach(self, observer: Observer) -> None:
""" Attaches new observer to the subject. """
self._observers.add(observer)
def detach(self, observer: Observer) -> None:
""" Detaches observer from the subject. """
if observer not in self._observers:
warnings.warn('Observer {} was not attached to a Subject {}'.format(observer, self))
else:
self._observers.remove(observer)
def notify(self, value: Optional[any] = None) -> None:
""" Sends event changes to all attached observers. """
for observer in self._observers:
observer.update(value)
def __enter__(self) -> 'Subject':
return self
def __exit__(self, exc_type, exc_val, exc_tb) -> None:
""" Clear all observers to prevent dangling references. """
self._observers.clear()
|
993,846 | 28559ba9ea074d81add7eea1103b61b6c6706cce |
import numpy
from matplotlib import pyplot as plot
from matplotlib.backends.backend_pdf import PdfPages
# Generate the data
data = numpy.random.randn(24, 1024)
# The PDF document
pdf_pages = PdfPages('histograms.pdf')
# Generate the pages
nb_plots = data.shape[0]
nb_plots_per_page = 6
nb_plots_per_page_two=nb_plots_per_page*2
nb_pages = int(numpy.ceil(nb_plots / float(nb_plots_per_page)))
grid_size = (nb_plots_per_page, 2)
row_pos=0
for i, samples in enumerate(data):
# Create a figure instance (ie. a new page) if needed
print "i=",i
if i % nb_plots_per_page == 0:
fig = plot.figure(figsize=(8.27, 11.69), dpi=100)
print 'new page'
if i==0:
col_pos=0
elif i%2==0:
col_pos=0
else:
col_pos=1
# Plot stuffs !
print "row,col:",row_pos,col_pos
plot.subplot2grid(grid_size, (row_pos, col_pos))
plot.hist(samples, 32, normed=1, facecolor='#808080', alpha=0.75)
if i%2!=0:
row_pos+=1
# Close the page if needed
if (i + 1) % nb_plots_per_page == 0 or (i + 1) == nb_plots:
print 'saved page'
plot.tight_layout()
pdf_pages.savefig(fig)
# Write the PDF document to the disk
pdf_pages.close() |
993,847 | 5096c1b5838f9152e6555aa57cf2760dc8c3e38f | import requests
import json
import dateutil.parser
from dateutil.tz import tzutc
import datetime
import os
import humanize
valid_execution_states = {'starting', 'idle', 'busy'}
from typing import List, TypedDict
class Session(TypedDict):
path: str
last_activity: datetime.datetime
execution_state: str
def get_sessions(url: str, token: str) -> List[Session]:
"""
return sessions sorted by last activity
"""
sessions_url = f'{url}api/sessions'
response = requests.get(sessions_url, params={'token': token})
assert(response.status_code == 200)
sessions_raw = json.loads(response.text)
sessions = []
for session_raw in sessions_raw:
session = Session(
path = session_raw['path'],
last_activity = dateutil.parser.isoparse(session_raw['kernel']['last_activity']),
execution_state = session_raw['kernel']['execution_state']
)
assert(session['execution_state'] in valid_execution_states)
sessions.append(session)
sessions.sort(key=lambda session: session['last_activity'], reverse=True)
return sessions
def test_running(sessions: List[Session], url: str) -> bool:
running = [session for session in sessions if session['execution_state'] != 'idle']
if len(running) == 0:
message = f"No notebooks currently running on {url}"
else:
plural = '' if len(running) == 1 else 's'
message_lines = [f"{len(running)} notebook{plural} running on {url}"]
for session in running:
message_lines.append(f" - {session['path']}: {session['execution_state']}")
message = os.linesep.join(message_lines)
print(message)
return len(running) > 0
def test_active(sessions: List[Session], url: str, timeout: datetime.timedelta) -> bool:
now = datetime.datetime.now(tzutc())
active = [session for session in sessions if session['last_activity'] >= now - timeout]
if len(active) == 0:
message = f"No notebooks active in {humanize.naturaldelta(timeout)} on {url}"
else:
plural = '' if len(active) == 1 else 's'
message_lines = [f"{len(active)} notebook{plural} active in {humanize.naturaldelta(timeout)} on {url}"]
for session in active:
message_lines.append(f" - {session['path']}: {humanize.naturaldelta(now-session['last_activity'])}")
message = os.linesep.join(message_lines)
print(message)
return len(active) > 0
|
993,848 | a388acd673a3d80133ae963c86633c4c214283a5 | import pandas as pd
import numpy as np
from generator import *
# data and metric imports
import sklearn.model_selection
import sklearn.datasets
import sklearn.metrics
from sklearn.feature_selection import SelectFromModel
from sklearn.ensemble import RandomForestClassifier
from skbio.stats.distance import permanova
from scipy.spatial import distance
from skbio import DistanceMatrix
import matplotlib.pyplot as plt
#Build discriminator
#X1xlsx is the excel wkbk with the real samples
#X2xlsx is the excek ekbk with the fake samples
#Uses Jensen Shannon Distance Metric with a PERMANOVA test
def discriminator(X1xlsx, X2xlsx):
X1_df = pd.read_excel(X1xlsx, index_col=0)
X1 = X1_df.to_numpy()
y1 = np.ones(X1.shape[0])
X2_df = pd.read_excel(X2xlsx, index_col=0)
X2 = X2_df.to_numpy()
y2 = np.zeros(X2.shape[0])
X = np.concatenate((X1,X2), axis=0)
y = np.concatenate((y1,y2))
dm = np.zeros((X.shape[0],X.shape[0]))
for i in range(X.shape[0]):
for j in range(i+1):
if i==j:
dm[i][j] = 0
else:
dm[i][j] = dm[j][i] = distance.jensenshannon(X[i], X[j])
dm_from_np = DistanceMatrix(dm)
results = permanova(dm_from_np, y, permutations=10000)
return results
|
993,849 | 47a1fb3a1303cc1b981dfd7a783bc2ee7de88146 | #!/usr/bin/env python
import numpy as np
import h5py
import matplotlib.pyplot as plt
from matplotlib.colors import LogNorm
def get_wave(data, key, ch, tick_min, tick_max):
frame = np.array(data[key])
wave = frame[tick_min:tick_max, ch]
return wave
def show_frame(data, event, apa, tag) :
key = '/%d/frame_%s%d'%(event,tag,apa)
f = data.get(key)
if f is None:
print('f is None')
return
frame = np.array(f)
print(frame.shape)
frame_ma = np.ma.array(frame)
# plt.gca().set_title(tag)
# plt.hist(frame.reshape(-1), bins=100, range=(-2000,2000))
plt.imshow(frame_ma, cmap="jet", interpolation="none"
# , extent = [0 , 2560, 0 , 6000]
, origin='lower'
# , aspect='auto'
, aspect=0.8/4.7
# , aspect=0.1
)
# plt.colorbar()
# plt.xlim([0, 1600])
# plt.xlim([0, 800]) # U
# plt.xlim([800, 1600]) # V
plt.xlim([2080, 2560]) # W
# plt.clim([2300,2400]) # orig U, V
# plt.clim([885,915]) # orig W
plt.clim([0,10000])
# plt.grid()
# plt.show()
plt.savefig('{}-{}.png'.format(tag,event), dpi=300)
return
if __name__ == '__main__':
apa = 0
tags = [
# 'ductor'
# 'orig',
'gauss'
]
data = h5py.File('g4-rec-%d.h5'%apa, 'r')
# data = h5py.File('g4-tru-%d.h5'%apa, 'r')
plt.figure()
start = 100
for event in range(start, start+len(list(data.keys()))) :
print("hioplot event: ", event)
for tag in tags:
show_frame(data, event, apa, tag) |
993,850 | 8072836644f02d179530bbdfed7c328df7e8a202 | #!/usr/bin/env python3
import __main__
import math
import os.path
from northstar.srv import get_orientation
from turtlesim.msg import Pose
import rospy
class orientation_service:
"""
A turtlesim-specific service to return the orientation of the turtle with respect to
True North. An orientation returned of zero indicates turtle is pointing directly to North.
"""
def __init__(self):
# Keep the received theta (rotation) of the turtle in counter-clockwise radians where
# zero rotation indicates pointing due East.
self.current_theta = None
# listen to changes in pose from the turtle simulation
sub = rospy.Subscriber('turtle1/pose', Pose, lambda x: self.handle_incoming(x))
# provide a service to return the current clockwise rotation from the desired target
svc = rospy.Service("get_orientation", get_orientation, lambda x: self.get_orientation(x))
def handle_incoming(self, msg):
if self.current_theta is None or msg.theta != self.current_theta:
self.current_theta = msg.theta
rospy.logdebug("< %s", msg)
def _calc_orientation(self):
# We want to return degrees clockwise rotation from North calculated from
# counter-clockwise radians rotation from East
# once converted add on 90deg due to turtle pose's theta of zero referring
# to due East
if self.current_theta is not None:
return math.degrees(-self.current_theta) + 90
return math.nan
def get_orientation(self, request_msg):
response_msg = self._calc_orientation()
rospy.logdebug("> %s", response_msg)
return response_msg
if __name__ == '__main__':
node_name = os.path.basename(__main__.__file__[:-3])
rospy.init_node(node_name)
rospy.loginfo("Started [%s] node", node_name)
svc = orientation_service()
rospy.spin()
|
993,851 | cc5bea557193ca8c59b31a62d71cd06e898ea72e | from conans import ConanFile, tools, CMake
from conans.tools import check_min_cppstd
import os
class RepackedCorrade(ConanFile):
name = "repacked_corrade"
homepage = "https://github.com/werto87/repacked_corrade"
description = "repacked corrade https://github.com/mosra/corrade.git"
topics = ("utility", "magnum-dependency")
license = "BSL-1.0"
url = "https://gitlab.com/werto87/conan-the-example"
settings = "compiler"
@property
def _source_subfolder(self):
return "source_subfolder"
def configure(self):
if self.settings.compiler.cppstd:
check_min_cppstd(self, "20")
def source(self):
tools.get(**self.conan_data["sources"][self.version])
extracted_dir = self.name + "-" + self.version
os.rename(extracted_dir, self._source_subfolder)
def build(self):
cmake = CMake(self)
cmake.build()
def package(self):
self.copy("*.h*", dst=f"include/{self.name}",
src=f"{self._source_subfolder()}/{self.name}")
def package_info(self):
self.cpp_info.libs = ["corrade"]
|
993,852 | 977c3b9b17c414be9a54390f33f7a012123e6393 | datefile=open('date.info', 'r')
date = datefile.read()[:-1]
rasterfilepath=open('FLIP_'+ date + '.asc', 'r')
rasterfile = rasterfilepath.readlines()
with open('Corr_' + date + '.asc', 'w') as txt:
for i in xrange(6):
txt.write(rasterfile[i])
for j in xrange(len(rasterfile)-6):
txt.write(rasterfile[len(rasterfile)-j-1])
|
993,853 | 7289434f0410f73305efa0c752878508d08af3d6 | # ---
# jupyter:
# jupytext:
# formats: ipynb,py:percent
# notebook_metadata_filter: all
# text_representation:
# extension: .py
# format_name: percent
# format_version: '1.3'
# jupytext_version: 1.14.5
# kernelspec:
# display_name: Python 3 (ipykernel)
# language: python
# name: python3
# language_info:
# codemirror_mode:
# name: ipython
# version: 3
# file_extension: .py
# mimetype: text/x-python
# name: python
# nbconvert_exporter: python
# pygments_lexer: ipython3
# version: 3.10.8
# latex_envs:
# LaTeX_envs_menu_present: true
# autoclose: false
# autocomplete: true
# bibliofile: biblio.bib
# cite_by: apalike
# current_citInitial: 1
# eqLabelWithNumbers: true
# eqNumInitial: 1
# hotkeys:
# equation: Ctrl-E
# itemize: Ctrl-I
# labels_anchors: false
# latex_user_defs: false
# report_style_numbering: false
# user_envs_cfg: false
# widgets:
# application/vnd.jupyter.widget-state+json:
# state:
# 073ec5924a5140b0ba11f5621a36f605:
# model_module: '@jupyter-widgets/base'
# model_module_version: 2.0.0
# model_name: LayoutModel
# state:
# _model_module: '@jupyter-widgets/base'
# _model_module_version: 2.0.0
# _model_name: LayoutModel
# _view_count: null
# _view_module: '@jupyter-widgets/base'
# _view_module_version: 2.0.0
# _view_name: LayoutView
# align_content: null
# align_items: null
# align_self: null
# border_bottom: null
# border_left: null
# border_right: null
# border_top: null
# bottom: null
# display: null
# flex: null
# flex_flow: null
# grid_area: null
# grid_auto_columns: null
# grid_auto_flow: null
# grid_auto_rows: null
# grid_column: null
# grid_gap: null
# grid_row: null
# grid_template_areas: null
# grid_template_columns: null
# grid_template_rows: null
# height: null
# justify_content: null
# justify_items: null
# left: null
# margin: null
# max_height: null
# max_width: null
# min_height: null
# min_width: null
# object_fit: null
# object_position: null
# order: null
# overflow: null
# padding: null
# right: null
# top: null
# visibility: null
# width: null
# 0a5481c8f31445a8ae1bfd572c4fbf0e:
# model_module: '@jupyter-widgets/controls'
# model_module_version: 2.0.0
# model_name: FloatSliderModel
# state:
# _dom_classes: []
# _model_module: '@jupyter-widgets/controls'
# _model_module_version: 2.0.0
# _model_name: FloatSliderModel
# _view_count: null
# _view_module: '@jupyter-widgets/controls'
# _view_module_version: 2.0.0
# _view_name: FloatSliderView
# behavior: drag-tap
# continuous_update: false
# description: Capital Share $\epsilon$
# description_allow_html: false
# disabled: false
# layout: IPY_MODEL_5b3b8de406c54117a97d00ff1a145152
# max: 0.4
# min: 0.2
# orientation: horizontal
# readout: true
# readout_format: .3f
# step: 0.01
# style: IPY_MODEL_1bf74a4563a5400cab0a2cbc66fd656b
# tabbable: null
# tooltip: null
# value: 0.33
# 1301681f1a614df6bdbcd5d59092d742:
# model_module: '@jupyter-widgets/controls'
# model_module_version: 2.0.0
# model_name: SliderStyleModel
# state:
# _model_module: '@jupyter-widgets/controls'
# _model_module_version: 2.0.0
# _model_name: SliderStyleModel
# _view_count: null
# _view_module: '@jupyter-widgets/base'
# _view_module_version: 2.0.0
# _view_name: StyleView
# description_width: initial
# handle_color: null
# 15d20500b128469faa56c52d2aa22a62:
# model_module: '@jupyter-widgets/base'
# model_module_version: 2.0.0
# model_name: LayoutModel
# state:
# _model_module: '@jupyter-widgets/base'
# _model_module_version: 2.0.0
# _model_name: LayoutModel
# _view_count: null
# _view_module: '@jupyter-widgets/base'
# _view_module_version: 2.0.0
# _view_name: LayoutView
# align_content: null
# align_items: null
# align_self: null
# border_bottom: null
# border_left: null
# border_right: null
# border_top: null
# bottom: null
# display: null
# flex: null
# flex_flow: null
# grid_area: null
# grid_auto_columns: null
# grid_auto_flow: null
# grid_auto_rows: null
# grid_column: null
# grid_gap: null
# grid_row: null
# grid_template_areas: null
# grid_template_columns: null
# grid_template_rows: null
# height: null
# justify_content: null
# justify_items: null
# left: null
# margin: null
# max_height: null
# max_width: null
# min_height: null
# min_width: null
# object_fit: null
# object_position: null
# order: null
# overflow: null
# padding: null
# right: null
# top: null
# visibility: null
# width: null
# 16b16b73803d4363a99b20d397a1d70d:
# model_module: '@jupyter-widgets/controls'
# model_module_version: 2.0.0
# model_name: VBoxModel
# state:
# _dom_classes:
# - widget-interact
# _model_module: '@jupyter-widgets/controls'
# _model_module_version: 2.0.0
# _model_name: VBoxModel
# _view_count: null
# _view_module: '@jupyter-widgets/controls'
# _view_module_version: 2.0.0
# _view_name: VBoxView
# box_style: ''
# children:
# - IPY_MODEL_0a5481c8f31445a8ae1bfd572c4fbf0e
# - IPY_MODEL_402d256f65a44dcbb97f5fc7c8043158
# - IPY_MODEL_e1209d2465264af88edbecc606f231de
# layout: IPY_MODEL_4f1ba9d72962449a9f8888787d017d5e
# tabbable: null
# tooltip: null
# 1bf74a4563a5400cab0a2cbc66fd656b:
# model_module: '@jupyter-widgets/controls'
# model_module_version: 2.0.0
# model_name: SliderStyleModel
# state:
# _model_module: '@jupyter-widgets/controls'
# _model_module_version: 2.0.0
# _model_name: SliderStyleModel
# _view_count: null
# _view_module: '@jupyter-widgets/base'
# _view_module_version: 2.0.0
# _view_name: StyleView
# description_width: initial
# handle_color: null
# 2433dd36ca1f4e85aa4e68f185427393:
# model_module: '@jupyter-widgets/controls'
# model_module_version: 2.0.0
# model_name: FloatSliderModel
# state:
# _dom_classes: []
# _model_module: '@jupyter-widgets/controls'
# _model_module_version: 2.0.0
# _model_name: FloatSliderModel
# _view_count: null
# _view_module: '@jupyter-widgets/controls'
# _view_module_version: 2.0.0
# _view_name: FloatSliderView
# behavior: drag-tap
# continuous_update: false
# description: Capital Share $\epsilon$
# description_allow_html: false
# disabled: false
# layout: IPY_MODEL_49d4e63b2183471aa786d7ff10367d47
# max: 0.4
# min: 0.2
# orientation: horizontal
# readout: true
# readout_format: .3f
# step: 0.01
# style: IPY_MODEL_969905a4ce5b494f8c39b5fd8fca7159
# tabbable: null
# tooltip: null
# value: 0.33
# 2ae0480191fb463e94d8ddb0e07e40d9:
# model_module: '@jupyter-widgets/base'
# model_module_version: 2.0.0
# model_name: LayoutModel
# state:
# _model_module: '@jupyter-widgets/base'
# _model_module_version: 2.0.0
# _model_name: LayoutModel
# _view_count: null
# _view_module: '@jupyter-widgets/base'
# _view_module_version: 2.0.0
# _view_name: LayoutView
# align_content: null
# align_items: null
# align_self: null
# border_bottom: null
# border_left: null
# border_right: null
# border_top: null
# bottom: null
# display: null
# flex: null
# flex_flow: null
# grid_area: null
# grid_auto_columns: null
# grid_auto_flow: null
# grid_auto_rows: null
# grid_column: null
# grid_gap: null
# grid_row: null
# grid_template_areas: null
# grid_template_columns: null
# grid_template_rows: null
# height: null
# justify_content: null
# justify_items: null
# left: null
# margin: null
# max_height: null
# max_width: null
# min_height: null
# min_width: null
# object_fit: null
# object_position: null
# order: null
# overflow: null
# padding: null
# right: null
# top: null
# visibility: null
# width: null
# 2f4a706726b040cfb27e624519511fcd:
# model_module: '@jupyter-widgets/controls'
# model_module_version: 2.0.0
# model_name: FloatSliderModel
# state:
# _dom_classes: []
# _model_module: '@jupyter-widgets/controls'
# _model_module_version: 2.0.0
# _model_name: FloatSliderModel
# _view_count: null
# _view_module: '@jupyter-widgets/controls'
# _view_module_version: 2.0.0
# _view_name: FloatSliderView
# behavior: drag-tap
# continuous_update: false
# description: Discount Factor $\beta$
# description_allow_html: false
# disabled: false
# layout: IPY_MODEL_94e0ba73fe56465da000062125bde52b
# max: 0.99
# min: 0.94
# orientation: horizontal
# readout: true
# readout_format: .3f
# step: 0.001
# style: IPY_MODEL_e1d670f3f9fb467bb214057129777dc7
# tabbable: null
# tooltip: null
# value: 0.96
# 36256c84647b49a68a7f9ec4f7c2cc7a:
# model_module: '@jupyter-widgets/base'
# model_module_version: 2.0.0
# model_name: LayoutModel
# state:
# _model_module: '@jupyter-widgets/base'
# _model_module_version: 2.0.0
# _model_name: LayoutModel
# _view_count: null
# _view_module: '@jupyter-widgets/base'
# _view_module_version: 2.0.0
# _view_name: LayoutView
# align_content: null
# align_items: null
# align_self: null
# border_bottom: null
# border_left: null
# border_right: null
# border_top: null
# bottom: null
# display: null
# flex: null
# flex_flow: null
# grid_area: null
# grid_auto_columns: null
# grid_auto_flow: null
# grid_auto_rows: null
# grid_column: null
# grid_gap: null
# grid_row: null
# grid_template_areas: null
# grid_template_columns: null
# grid_template_rows: null
# height: null
# justify_content: null
# justify_items: null
# left: null
# margin: null
# max_height: null
# max_width: null
# min_height: null
# min_width: null
# object_fit: null
# object_position: null
# order: null
# overflow: null
# padding: null
# right: null
# top: null
# visibility: null
# width: null
# 36e9bed81a104775952b02d494e6ddfa:
# model_module: '@jupyter-widgets/controls'
# model_module_version: 2.0.0
# model_name: SliderStyleModel
# state:
# _model_module: '@jupyter-widgets/controls'
# _model_module_version: 2.0.0
# _model_name: SliderStyleModel
# _view_count: null
# _view_module: '@jupyter-widgets/base'
# _view_module_version: 2.0.0
# _view_name: StyleView
# description_width: initial
# handle_color: null
# 3b2e3635af704f079a205df3a0ed553c:
# model_module: '@jupyter-widgets/base'
# model_module_version: 2.0.0
# model_name: LayoutModel
# state:
# _model_module: '@jupyter-widgets/base'
# _model_module_version: 2.0.0
# _model_name: LayoutModel
# _view_count: null
# _view_module: '@jupyter-widgets/base'
# _view_module_version: 2.0.0
# _view_name: LayoutView
# align_content: null
# align_items: null
# align_self: null
# border_bottom: null
# border_left: null
# border_right: null
# border_top: null
# bottom: null
# display: null
# flex: null
# flex_flow: null
# grid_area: null
# grid_auto_columns: null
# grid_auto_flow: null
# grid_auto_rows: null
# grid_column: null
# grid_gap: null
# grid_row: null
# grid_template_areas: null
# grid_template_columns: null
# grid_template_rows: null
# height: null
# justify_content: null
# justify_items: null
# left: null
# margin: null
# max_height: null
# max_width: null
# min_height: null
# min_width: null
# object_fit: null
# object_position: null
# order: null
# overflow: null
# padding: null
# right: null
# top: null
# visibility: null
# width: null
# 3eb6165e3e724dc0b3439e948bfbbd7f:
# model_module: '@jupyter-widgets/base'
# model_module_version: 2.0.0
# model_name: LayoutModel
# state:
# _model_module: '@jupyter-widgets/base'
# _model_module_version: 2.0.0
# _model_name: LayoutModel
# _view_count: null
# _view_module: '@jupyter-widgets/base'
# _view_module_version: 2.0.0
# _view_name: LayoutView
# align_content: null
# align_items: null
# align_self: null
# border_bottom: null
# border_left: null
# border_right: null
# border_top: null
# bottom: null
# display: null
# flex: null
# flex_flow: null
# grid_area: null
# grid_auto_columns: null
# grid_auto_flow: null
# grid_auto_rows: null
# grid_column: null
# grid_gap: null
# grid_row: null
# grid_template_areas: null
# grid_template_columns: null
# grid_template_rows: null
# height: null
# justify_content: null
# justify_items: null
# left: null
# margin: null
# max_height: null
# max_width: null
# min_height: null
# min_width: null
# object_fit: null
# object_position: null
# order: null
# overflow: null
# padding: null
# right: null
# top: null
# visibility: null
# width: null
# 3f87327fbf1f464185f30d9143ec841e:
# model_module: '@jupyter-widgets/base'
# model_module_version: 2.0.0
# model_name: LayoutModel
# state:
# _model_module: '@jupyter-widgets/base'
# _model_module_version: 2.0.0
# _model_name: LayoutModel
# _view_count: null
# _view_module: '@jupyter-widgets/base'
# _view_module_version: 2.0.0
# _view_name: LayoutView
# align_content: null
# align_items: null
# align_self: null
# border_bottom: null
# border_left: null
# border_right: null
# border_top: null
# bottom: null
# display: null
# flex: null
# flex_flow: null
# grid_area: null
# grid_auto_columns: null
# grid_auto_flow: null
# grid_auto_rows: null
# grid_column: null
# grid_gap: null
# grid_row: null
# grid_template_areas: null
# grid_template_columns: null
# grid_template_rows: null
# height: null
# justify_content: null
# justify_items: null
# left: null
# margin: null
# max_height: null
# max_width: null
# min_height: null
# min_width: null
# object_fit: null
# object_position: null
# order: null
# overflow: null
# padding: null
# right: null
# top: null
# visibility: null
# width: null
# 402d256f65a44dcbb97f5fc7c8043158:
# model_module: '@jupyter-widgets/controls'
# model_module_version: 2.0.0
# model_name: FloatSliderModel
# state:
# _dom_classes: []
# _model_module: '@jupyter-widgets/controls'
# _model_module_version: 2.0.0
# _model_name: FloatSliderModel
# _view_count: null
# _view_module: '@jupyter-widgets/controls'
# _view_module_version: 2.0.0
# _view_name: FloatSliderView
# behavior: drag-tap
# continuous_update: false
# description: Pop. Growth $\Xi$
# description_allow_html: false
# disabled: false
# layout: IPY_MODEL_073ec5924a5140b0ba11f5621a36f605
# max: 1.05
# min: 0.98
# orientation: horizontal
# readout: true
# readout_format: .3f
# step: 0.001
# style: IPY_MODEL_9f5a3c55e8874c4f934b352e91b94601
# tabbable: null
# tooltip: null
# value: 1.01
# 49d4e63b2183471aa786d7ff10367d47:
# model_module: '@jupyter-widgets/base'
# model_module_version: 2.0.0
# model_name: LayoutModel
# state:
# _model_module: '@jupyter-widgets/base'
# _model_module_version: 2.0.0
# _model_name: LayoutModel
# _view_count: null
# _view_module: '@jupyter-widgets/base'
# _view_module_version: 2.0.0
# _view_name: LayoutView
# align_content: null
# align_items: null
# align_self: null
# border_bottom: null
# border_left: null
# border_right: null
# border_top: null
# bottom: null
# display: null
# flex: null
# flex_flow: null
# grid_area: null
# grid_auto_columns: null
# grid_auto_flow: null
# grid_auto_rows: null
# grid_column: null
# grid_gap: null
# grid_row: null
# grid_template_areas: null
# grid_template_columns: null
# grid_template_rows: null
# height: null
# justify_content: null
# justify_items: null
# left: null
# margin: null
# max_height: null
# max_width: null
# min_height: null
# min_width: null
# object_fit: null
# object_position: null
# order: null
# overflow: null
# padding: null
# right: null
# top: null
# visibility: null
# width: null
# 4f1ba9d72962449a9f8888787d017d5e:
# model_module: '@jupyter-widgets/base'
# model_module_version: 2.0.0
# model_name: LayoutModel
# state:
# _model_module: '@jupyter-widgets/base'
# _model_module_version: 2.0.0
# _model_name: LayoutModel
# _view_count: null
# _view_module: '@jupyter-widgets/base'
# _view_module_version: 2.0.0
# _view_name: LayoutView
# align_content: null
# align_items: null
# align_self: null
# border_bottom: null
# border_left: null
# border_right: null
# border_top: null
# bottom: null
# display: null
# flex: null
# flex_flow: null
# grid_area: null
# grid_auto_columns: null
# grid_auto_flow: null
# grid_auto_rows: null
# grid_column: null
# grid_gap: null
# grid_row: null
# grid_template_areas: null
# grid_template_columns: null
# grid_template_rows: null
# height: null
# justify_content: null
# justify_items: null
# left: null
# margin: null
# max_height: null
# max_width: null
# min_height: null
# min_width: null
# object_fit: null
# object_position: null
# order: null
# overflow: null
# padding: null
# right: null
# top: null
# visibility: null
# width: null
# 5b3b8de406c54117a97d00ff1a145152:
# model_module: '@jupyter-widgets/base'
# model_module_version: 2.0.0
# model_name: LayoutModel
# state:
# _model_module: '@jupyter-widgets/base'
# _model_module_version: 2.0.0
# _model_name: LayoutModel
# _view_count: null
# _view_module: '@jupyter-widgets/base'
# _view_module_version: 2.0.0
# _view_name: LayoutView
# align_content: null
# align_items: null
# align_self: null
# border_bottom: null
# border_left: null
# border_right: null
# border_top: null
# bottom: null
# display: null
# flex: null
# flex_flow: null
# grid_area: null
# grid_auto_columns: null
# grid_auto_flow: null
# grid_auto_rows: null
# grid_column: null
# grid_gap: null
# grid_row: null
# grid_template_areas: null
# grid_template_columns: null
# grid_template_rows: null
# height: null
# justify_content: null
# justify_items: null
# left: null
# margin: null
# max_height: null
# max_width: null
# min_height: null
# min_width: null
# object_fit: null
# object_position: null
# order: null
# overflow: null
# padding: null
# right: null
# top: null
# visibility: null
# width: null
# 5c01c9d84b1e4208b5c6e305f5280a32:
# model_module: '@jupyter-widgets/controls'
# model_module_version: 2.0.0
# model_name: SliderStyleModel
# state:
# _model_module: '@jupyter-widgets/controls'
# _model_module_version: 2.0.0
# _model_name: SliderStyleModel
# _view_count: null
# _view_module: '@jupyter-widgets/base'
# _view_module_version: 2.0.0
# _view_name: StyleView
# description_width: initial
# handle_color: null
# 6491284613264ac389258acc14449532:
# model_module: '@jupyter-widgets/controls'
# model_module_version: 2.0.0
# model_name: FloatSliderModel
# state:
# _dom_classes: []
# _model_module: '@jupyter-widgets/controls'
# _model_module_version: 2.0.0
# _model_name: FloatSliderModel
# _view_count: null
# _view_module: '@jupyter-widgets/controls'
# _view_module_version: 2.0.0
# _view_name: FloatSliderView
# behavior: drag-tap
# continuous_update: false
# description: Capital Share $\epsilon$
# description_allow_html: false
# disabled: false
# layout: IPY_MODEL_3b2e3635af704f079a205df3a0ed553c
# max: 0.4
# min: 0.2
# orientation: horizontal
# readout: true
# readout_format: .3f
# step: 0.01
# style: IPY_MODEL_ca6fa929f35a4438b0249ad44c03c26f
# tabbable: null
# tooltip: null
# value: 0.33
# 66635fdac1714c2ebf56dceb0a22f5d5:
# model_module: '@jupyter-widgets/controls'
# model_module_version: 2.0.0
# model_name: FloatSliderModel
# state:
# _dom_classes: []
# _model_module: '@jupyter-widgets/controls'
# _model_module_version: 2.0.0
# _model_name: FloatSliderModel
# _view_count: null
# _view_module: '@jupyter-widgets/controls'
# _view_module_version: 2.0.0
# _view_name: FloatSliderView
# behavior: drag-tap
# continuous_update: true
# description: Init. capital ratio
# description_allow_html: false
# disabled: false
# layout: IPY_MODEL_f38de7846e9d42daa339e1189dd6fcd7
# max: 0.1
# min: 0.01
# orientation: horizontal
# readout: true
# readout_format: .3f
# step: 0.01
# style: IPY_MODEL_36e9bed81a104775952b02d494e6ddfa
# tabbable: null
# tooltip: null
# value: 0.08
# 737f6d7583d94118a09d1eab85968920:
# model_module: '@jupyter-widgets/controls'
# model_module_version: 2.0.0
# model_name: FloatSliderModel
# state:
# _dom_classes: []
# _model_module: '@jupyter-widgets/controls'
# _model_module_version: 2.0.0
# _model_name: FloatSliderModel
# _view_count: null
# _view_module: '@jupyter-widgets/controls'
# _view_module_version: 2.0.0
# _view_name: FloatSliderView
# behavior: drag-tap
# continuous_update: false
# description: Pop. Growth $\Xi$
# description_allow_html: false
# disabled: false
# layout: IPY_MODEL_3f87327fbf1f464185f30d9143ec841e
# max: 1.05
# min: 0.98
# orientation: horizontal
# readout: true
# readout_format: .3f
# step: 0.001
# style: IPY_MODEL_1301681f1a614df6bdbcd5d59092d742
# tabbable: null
# tooltip: null
# value: 1.01
# 7682a977124d4614a03e7dbd0e442e9a:
# model_module: '@jupyter-widgets/base'
# model_module_version: 2.0.0
# model_name: LayoutModel
# state:
# _model_module: '@jupyter-widgets/base'
# _model_module_version: 2.0.0
# _model_name: LayoutModel
# _view_count: null
# _view_module: '@jupyter-widgets/base'
# _view_module_version: 2.0.0
# _view_name: LayoutView
# align_content: null
# align_items: null
# align_self: null
# border_bottom: null
# border_left: null
# border_right: null
# border_top: null
# bottom: null
# display: null
# flex: null
# flex_flow: null
# grid_area: null
# grid_auto_columns: null
# grid_auto_flow: null
# grid_auto_rows: null
# grid_column: null
# grid_gap: null
# grid_row: null
# grid_template_areas: null
# grid_template_columns: null
# grid_template_rows: null
# height: null
# justify_content: null
# justify_items: null
# left: null
# margin: null
# max_height: null
# max_width: null
# min_height: null
# min_width: null
# object_fit: null
# object_position: null
# order: null
# overflow: null
# padding: null
# right: null
# top: null
# visibility: null
# width: null
# 93df4a2b3c0142919734e7b4e743aa64:
# model_module: '@jupyter-widgets/base'
# model_module_version: 2.0.0
# model_name: LayoutModel
# state:
# _model_module: '@jupyter-widgets/base'
# _model_module_version: 2.0.0
# _model_name: LayoutModel
# _view_count: null
# _view_module: '@jupyter-widgets/base'
# _view_module_version: 2.0.0
# _view_name: LayoutView
# align_content: null
# align_items: null
# align_self: null
# border_bottom: null
# border_left: null
# border_right: null
# border_top: null
# bottom: null
# display: null
# flex: null
# flex_flow: null
# grid_area: null
# grid_auto_columns: null
# grid_auto_flow: null
# grid_auto_rows: null
# grid_column: null
# grid_gap: null
# grid_row: null
# grid_template_areas: null
# grid_template_columns: null
# grid_template_rows: null
# height: null
# justify_content: null
# justify_items: null
# left: null
# margin: null
# max_height: null
# max_width: null
# min_height: null
# min_width: null
# object_fit: null
# object_position: null
# order: null
# overflow: null
# padding: null
# right: null
# top: null
# visibility: null
# width: null
# 94e0ba73fe56465da000062125bde52b:
# model_module: '@jupyter-widgets/base'
# model_module_version: 2.0.0
# model_name: LayoutModel
# state:
# _model_module: '@jupyter-widgets/base'
# _model_module_version: 2.0.0
# _model_name: LayoutModel
# _view_count: null
# _view_module: '@jupyter-widgets/base'
# _view_module_version: 2.0.0
# _view_name: LayoutView
# align_content: null
# align_items: null
# align_self: null
# border_bottom: null
# border_left: null
# border_right: null
# border_top: null
# bottom: null
# display: null
# flex: null
# flex_flow: null
# grid_area: null
# grid_auto_columns: null
# grid_auto_flow: null
# grid_auto_rows: null
# grid_column: null
# grid_gap: null
# grid_row: null
# grid_template_areas: null
# grid_template_columns: null
# grid_template_rows: null
# height: null
# justify_content: null
# justify_items: null
# left: null
# margin: null
# max_height: null
# max_width: null
# min_height: null
# min_width: null
# object_fit: null
# object_position: null
# order: null
# overflow: null
# padding: null
# right: null
# top: null
# visibility: null
# width: null
# 95aa9c64cd25419abc1e260f23103509:
# model_module: '@jupyter-widgets/controls'
# model_module_version: 2.0.0
# model_name: SliderStyleModel
# state:
# _model_module: '@jupyter-widgets/controls'
# _model_module_version: 2.0.0
# _model_name: SliderStyleModel
# _view_count: null
# _view_module: '@jupyter-widgets/base'
# _view_module_version: 2.0.0
# _view_name: StyleView
# description_width: ''
# handle_color: null
# 969905a4ce5b494f8c39b5fd8fca7159:
# model_module: '@jupyter-widgets/controls'
# model_module_version: 2.0.0
# model_name: SliderStyleModel
# state:
# _model_module: '@jupyter-widgets/controls'
# _model_module_version: 2.0.0
# _model_name: SliderStyleModel
# _view_count: null
# _view_module: '@jupyter-widgets/base'
# _view_module_version: 2.0.0
# _view_name: StyleView
# description_width: initial
# handle_color: null
# 9d581788e31747aca484a871858b3ae7:
# model_module: '@jupyter-widgets/controls'
# model_module_version: 2.0.0
# model_name: SliderStyleModel
# state:
# _model_module: '@jupyter-widgets/controls'
# _model_module_version: 2.0.0
# _model_name: SliderStyleModel
# _view_count: null
# _view_module: '@jupyter-widgets/base'
# _view_module_version: 2.0.0
# _view_name: StyleView
# description_width: ''
# handle_color: null
# 9f5a3c55e8874c4f934b352e91b94601:
# model_module: '@jupyter-widgets/controls'
# model_module_version: 2.0.0
# model_name: SliderStyleModel
# state:
# _model_module: '@jupyter-widgets/controls'
# _model_module_version: 2.0.0
# _model_name: SliderStyleModel
# _view_count: null
# _view_module: '@jupyter-widgets/base'
# _view_module_version: 2.0.0
# _view_name: StyleView
# description_width: initial
# handle_color: null
# a036ec58068a48b3b2a709cee5690db8:
# model_module: '@jupyter-widgets/controls'
# model_module_version: 2.0.0
# model_name: VBoxModel
# state:
# _dom_classes:
# - widget-interact
# _model_module: '@jupyter-widgets/controls'
# _model_module_version: 2.0.0
# _model_name: VBoxModel
# _view_count: null
# _view_module: '@jupyter-widgets/controls'
# _view_module_version: 2.0.0
# _view_name: VBoxView
# box_style: ''
# children:
# - IPY_MODEL_6491284613264ac389258acc14449532
# - IPY_MODEL_2f4a706726b040cfb27e624519511fcd
# - IPY_MODEL_737f6d7583d94118a09d1eab85968920
# - IPY_MODEL_66635fdac1714c2ebf56dceb0a22f5d5
# - IPY_MODEL_c54e0574e60740818fad31b0b1f687f5
# layout: IPY_MODEL_f6a4919cc0504674a7ca7f6de8473b6f
# tabbable: null
# tooltip: null
# a51b014f6f254aaea8830531a0d85123:
# model_module: '@jupyter-widgets/base'
# model_module_version: 2.0.0
# model_name: LayoutModel
# state:
# _model_module: '@jupyter-widgets/base'
# _model_module_version: 2.0.0
# _model_name: LayoutModel
# _view_count: null
# _view_module: '@jupyter-widgets/base'
# _view_module_version: 2.0.0
# _view_name: LayoutView
# align_content: null
# align_items: null
# align_self: null
# border_bottom: null
# border_left: null
# border_right: null
# border_top: null
# bottom: null
# display: null
# flex: null
# flex_flow: null
# grid_area: null
# grid_auto_columns: null
# grid_auto_flow: null
# grid_auto_rows: null
# grid_column: null
# grid_gap: null
# grid_row: null
# grid_template_areas: null
# grid_template_columns: null
# grid_template_rows: null
# height: null
# justify_content: null
# justify_items: null
# left: null
# margin: null
# max_height: null
# max_width: null
# min_height: null
# min_width: null
# object_fit: null
# object_position: null
# order: null
# overflow: null
# padding: null
# right: null
# top: null
# visibility: null
# width: null
# b893afd518f44f43926c3dd3359fbaf5:
# model_module: '@jupyter-widgets/controls'
# model_module_version: 2.0.0
# model_name: FloatSliderModel
# state:
# _dom_classes: []
# _model_module: '@jupyter-widgets/controls'
# _model_module_version: 2.0.0
# _model_name: FloatSliderModel
# _view_count: null
# _view_module: '@jupyter-widgets/controls'
# _view_module_version: 2.0.0
# _view_name: FloatSliderView
# behavior: drag-tap
# continuous_update: false
# description: Discount Factor $\beta$
# description_allow_html: false
# disabled: false
# layout: IPY_MODEL_7682a977124d4614a03e7dbd0e442e9a
# max: 0.99
# min: 0.95
# orientation: horizontal
# readout: true
# readout_format: .3f
# step: 0.001
# style: IPY_MODEL_5c01c9d84b1e4208b5c6e305f5280a32
# tabbable: null
# tooltip: null
# value: 0.96
# c45ba4769ad84c8c8739df50daa5ae62:
# model_module: '@jupyter-widgets/controls'
# model_module_version: 2.0.0
# model_name: VBoxModel
# state:
# _dom_classes:
# - widget-interact
# _model_module: '@jupyter-widgets/controls'
# _model_module_version: 2.0.0
# _model_name: VBoxModel
# _view_count: null
# _view_module: '@jupyter-widgets/controls'
# _view_module_version: 2.0.0
# _view_name: VBoxView
# box_style: ''
# children:
# - IPY_MODEL_2433dd36ca1f4e85aa4e68f185427393
# - IPY_MODEL_b893afd518f44f43926c3dd3359fbaf5
# - IPY_MODEL_c925655fa1a64bd2b62178b0397158f5
# - IPY_MODEL_dbfe69ec0bf2411c969309e52682f30a
# - IPY_MODEL_ff0f6e7fb87e4797835013ad57a1d95f
# layout: IPY_MODEL_2ae0480191fb463e94d8ddb0e07e40d9
# tabbable: null
# tooltip: null
# c54e0574e60740818fad31b0b1f687f5:
# model_module: '@jupyter-widgets/output'
# model_module_version: 1.0.0
# model_name: OutputModel
# state:
# _dom_classes: []
# _model_module: '@jupyter-widgets/output'
# _model_module_version: 1.0.0
# _model_name: OutputModel
# _view_count: null
# _view_module: '@jupyter-widgets/output'
# _view_module_version: 1.0.0
# _view_name: OutputView
# layout: IPY_MODEL_15d20500b128469faa56c52d2aa22a62
# msg_id: ''
# outputs:
# - data:
# image/png: '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
#
# '
# text/plain: <Figure size 900x600 with 1 Axes>
# metadata: {}
# output_type: display_data
# tabbable: null
# tooltip: null
# c925655fa1a64bd2b62178b0397158f5:
# model_module: '@jupyter-widgets/controls'
# model_module_version: 2.0.0
# model_name: FloatSliderModel
# state:
# _dom_classes: []
# _model_module: '@jupyter-widgets/controls'
# _model_module_version: 2.0.0
# _model_name: FloatSliderModel
# _view_count: null
# _view_module: '@jupyter-widgets/controls'
# _view_module_version: 2.0.0
# _view_name: FloatSliderView
# behavior: drag-tap
# continuous_update: true
# description: $z$
# description_allow_html: false
# disabled: false
# layout: IPY_MODEL_36256c84647b49a68a7f9ec4f7c2cc7a
# max: 0.02
# min: 0
# orientation: horizontal
# readout: true
# readout_format: .3f
# step: 0.005
# style: IPY_MODEL_95aa9c64cd25419abc1e260f23103509
# tabbable: null
# tooltip: null
# value: 0.015
# ca6fa929f35a4438b0249ad44c03c26f:
# model_module: '@jupyter-widgets/controls'
# model_module_version: 2.0.0
# model_name: SliderStyleModel
# state:
# _model_module: '@jupyter-widgets/controls'
# _model_module_version: 2.0.0
# _model_name: SliderStyleModel
# _view_count: null
# _view_module: '@jupyter-widgets/base'
# _view_module_version: 2.0.0
# _view_name: StyleView
# description_width: initial
# handle_color: null
# dbfe69ec0bf2411c969309e52682f30a:
# model_module: '@jupyter-widgets/controls'
# model_module_version: 2.0.0
# model_name: FloatSliderModel
# state:
# _dom_classes: []
# _model_module: '@jupyter-widgets/controls'
# _model_module_version: 2.0.0
# _model_name: FloatSliderModel
# _view_count: null
# _view_module: '@jupyter-widgets/controls'
# _view_module_version: 2.0.0
# _view_name: FloatSliderView
# behavior: drag-tap
# continuous_update: true
# description: $Rfree$
# description_allow_html: false
# disabled: false
# layout: IPY_MODEL_93df4a2b3c0142919734e7b4e743aa64
# max: 1.1
# min: 1
# orientation: horizontal
# readout: true
# readout_format: .3f
# step: 0.0001
# style: IPY_MODEL_9d581788e31747aca484a871858b3ae7
# tabbable: null
# tooltip: null
# value: 1.03
# e1209d2465264af88edbecc606f231de:
# model_module: '@jupyter-widgets/output'
# model_module_version: 1.0.0
# model_name: OutputModel
# state:
# _dom_classes: []
# _model_module: '@jupyter-widgets/output'
# _model_module_version: 1.0.0
# _model_name: OutputModel
# _view_count: null
# _view_module: '@jupyter-widgets/output'
# _view_module_version: 1.0.0
# _view_name: OutputView
# layout: IPY_MODEL_3eb6165e3e724dc0b3439e948bfbbd7f
# msg_id: ''
# outputs:
# - data:
# image/png: 'iVBORw0KGgoAAAANSUhEUgAABZ8AAAJHCAYAAAD/vijJAAAAOXRFWHRTb2Z0d2FyZQBNYXRwbG90bGliIHZlcnNpb24zLjYuMywgaHR0cHM6Ly9tYXRwbG90bGliLm9yZy/P9b71AAAACXBIWXMAAA9hAAAPYQGoP6dpAADRO0lEQVR4nOzdd3QU5dvG8WtJIwk1SJHQW2jSpYN0NHSQJlVpUqVHEPEHKJrQe0eKVF8RRIpK70WQXgKhVwmEFkravn+sWRKSQMomm/L9nLOHMDsze092s3n2yjP3GIxGo1EAAAAAAAAAAFhQKmsXAAAAAAAAAABIfgifAQAAAAAAAAAWR/gMAAAAAAAAALA4wmcAAAAAAAAAgMURPgMAAAAAAAAALI7wGQAAAAAAAABgcYTPAAAAAAAAAACLI3wGAAAAAAAAAFgc4TMAAAAAAAAAwOIInwEAAAAAAAAAFkf4DAAAAAAAAACwOFtrFwAAAAAAAGLv6tWrqlevXrhllStX1o8//miligAAMGHmMwAASNKCg4OtXQISCM91eIn5+5GYa0PiwGvEss6ePRthWeHCha1QCQAA4THzGQDwRgEBAfrjjz+0Y8cOnTlzRnfu3FFAQIDSp0+vbNmyqUKFCvrwww9VsmRJa5eKaKpVq5Zu3rypPn36qG/fvjHezs7OTr/++qsKFiz41m06dOigQ4cOqUyZMlqxYkVcyo7Upk2b9Ndff2nixIkx2i70WKJiZ2cnJycnvfvuuypbtqxatGihYsWKxbXcBHH37l2tXbtWe/fu1cWLF/X48WPZ2dkpa9asKl26tBo3bqxKlSpZu8wIbty4odq1a0uSvv32W7Vs2TLc/bF9ri3Nz89P69at0549e+Tj46P79+/LYDAoc+bMKl68uOrVq6d69erJ1jZ+h9mJ5fsRmcRcW2IT+h4ZE7Vr19bMmTPjqaKEEdVr5G3vA8nV8ePHNWPGDJ04cUJPnz6Vi4uL2rVrpx49ekR7H+fOnYuwrEiRIpYsEwCAWCF8BgBEae3atRo3bpx8fX0j3Hf//n3dv39fp0+f1sKFC1W9enWNHTtWmTNntkKlSEiBgYH68ssvtXr1atnY2FitjsmTJ2vWrFkqX768xfcdGBioR48e6dGjRzp37pyWL1+uHj16aMCAARZ/LEt58eKFpkyZoqVLlyowMDDcfYGBgbp8+bIuX76sNWvWqGrVqvLy8lKmTJmsVG3MxOdzHV1Go1ELFizQrFmz9PTp0wj3X79+XdevX9emTZuUM2dOffPNN6pWrVq81JIYvh9RScy1IXHgNRLexYsX1b59ewUEBJiX3b17V2nSpInRfgifAQCJFeEzACBSY8aM0U8//SRJsre3V7NmzVS7dm3ly5dPadOm1ePHj3XixAmtWbNGe/fu1a5du9S2bVutWLGCADoFOHXqlObPnx+jWVmWdufOnTjvo2zZspo3b16E5SEhIXry5In+/vtvTZw4Ubdv39bs2bOVK1cutWjRIs6Pa2mPHz9W165ddfz4cUlS8eLF1bZtW5UuXVqZMmWSn5+fLl68qEWLFunvv//Wnj171KZNG61atUouLi5Wrt7Ezs5OuXLlkqQIoYslnuu4CAwM1KBBg/THH39Ikt555x21a9dO1apVk6urq/z9/XXt2jWtW7dOGzdu1PXr19WtWzcNHTpUn332mcXrsfb3400Sc22JWfbs2fX7779Ha934nlUf3970GnnT+0BytX79egUEBMjGxkbjx49XxYoVFRQUFOPjf73thoODg/Lly2fJUgEAiJWkPXIBAMSLefPmmYPnYsWKadq0aXJ1dQ23ToYMGZQrVy41bNhQa9eu1fDhw3X9+nV99913mjx5shWqRkKbPn26ateurQIFCli7lFizsbGRs7NzpPelTZtWjRs31nvvvacmTZro5cuXmjJlipo1a6ZUqRLXZTMGDBhgDp67du2qwYMHy2AwmO/PkCGD8ubNq7p162ru3LmaMGGCrl27pmHDhmnOnDnWKjucrFmz6q+//rJ2GZH6/vvvzcFzgwYN9O2338rJycl8v4uLi3LmzKkqVaqoa9eu+vzzz3Xz5k15enoqc+bMatSokbVKRxJhMBiifC9KSRLz+0B8CT27rHDhwnJ3d4/VPh4+fBgh1C9YsKBVz04CACBU4vrkBACwuuvXr2vatGmSpNy5c2vx4sURgufXNW3aVJ07d5Zk6uN49erV+C4TVpQ/f36lSpVKAQEBGjZsWLK/aFTevHnVoEEDSaZToU+dOmXlisJbu3at9uzZI0lq3bq1hgwZEi54fl337t1Vq1YtSdKOHTv0999/J0idSdX+/fu1bNkySVKdOnU0YcKEcMHz6woVKqQVK1Yoffr0kkxnkTx48CBBagWQ9IT+Do3LHx8iu9hgZC03li1bpiJFisjNzc18K1mypDZt2hTrxwYA4G2Y+QwACGfhwoV6+fKlJGnkyJFKmzZttLbr2rWrlixZIkk6evSocufOLUk6ePCgOnbsKEk6efKk5syZo1WrVunx48d699131bt3bzVu3Ni8nx07dujnn3/W8ePH9fDhQzk7O6tQoUJyd3fXxx9/LDs7uyhr2LRpk9auXauTJ0/q0aNHSpMmjfLly6datWqpbdu2UZ7CGtvt3iQgIEDr1q3T1q1bdebMGT18+FCpUqWSi4uLSpUqpZYtW0Z60bewF1v6888/5ejoqLlz52rnzp26c+eOnJ2dVbJkSbVr107Vq1eP9LFDQkK0efNmrVixQj4+Pnr+/LkKFCigTz75RM2aNYvxsbyuaNGiql69un788UedOHFCCxYsUPfu3WO9v3/++UfLli3T33//rfv378vR0VGFChVSgwYNIn3Op02bpunTp5v/f+jQIbm5uUmStm7dqhw5csS6lqgULVpUa9askSTdvHlTJUqUiPNxxPRnIypz586VJDk5Oal///7ROp6+fftqx44dKlCgQKSnwFvi9btx40bZ29tr2rRp2rdvn54+fSpXV1dVr15dn332WYT2PJFdaCwmz3Vsa36bWbNmSTK1Hxo9evQbg/1QWbNm1aBBgzRy5Eg9evRIy5cvV58+fcz3h33uf/zxR1WuXDnS/YQe6+eff64BAwZE6/sRl+cgtnVJlvu5jOvzmJjezxNCXJ4zS/y+kUwXy1u5cqWOHj2q27dvy87OToUKFVKjRo3UqlUrc5uQmL5+o7rgYGzGCZY61reJSW2vX2wy7PejWbNm+uGHH6L9uG/r92w0GjVx4kTz74tQWbJk0cyZM/Xee+9F+7EAAIgpwmcAgFlwcLA2bNggScqZM6eqVq0a7W1dXFy0aNEiFS5cOMoP999//72WL19u/v+VK1eUM2dOSdLz5881aNAgbd26Ndw2Dx8+1KFDh3To0CEtX75cs2fPjnQm9tChQ7Vu3boI2x49elRHjx7VsmXLtGTJEnMvybhu9ybXr19X165ddeXKlQj33bx5Uzdv3tSGDRvUp08f9e3bN8r9nD59WqNGjdLDhw/NywICArRjxw7t2LFD/fr1U+/evcNt8/z5cw0YMEDbt28Pt/zEiRM6ceKE9uzZo5CQkGgfS1T69++v7du368qVK5o2bZpq166t/Pnzx2gfISEh8vLy0o8//hhueUBAgA4fPqzDhw9r9erVmj17trJmzRrnmuMibOD4+mnMljiON/1svMn58+fl4+MjyTQrN7r9m4sWLarDhw9H+rNqqdfv2bNnNWrUKD1+/Ni87OLFi7p48aLWrFmjuXPnqmTJktGq920sVfPr7t27p4MHD0qS6tWrF6MLNDZr1kyenp7y9/fXunXrwoXPCSUhnwNLiOvzmJjfzxO72Py+CQkJ0aRJkyIEmi9fvjR/z9evX6958+ZZpH9zXMcJcTnWhKottiILnwsXLizJ1LN++PDh+u2338LdX6xYMc2aNcvqv18BAMkfbTcAAGanTp3So0ePJEk1atSI8fblypV74wfM5cuXq169etq8ebO2bdum//3vfypdurQkaeDAgeYPbR9++KFWrVqlgwcPauPGjerevbtsbW3l7e2tzz77TE+fPg23399//90cOHTq1Em//fabDhw4oD///FODBg2Sra2tbt++rVGjRllkuzcJDg5Wnz59dOXKFTk5OWnYsGHatGmTDhw4oN9//13Dhw9XhgwZJEkzZ858Y4uS4cOHKyQkRF9//bW2bdumPXv2aNy4cebtZ8yYEWH7UaNGmYPn1q1ba/369Tpw4IAWLVqkUqVK6ffff9ft27ejfTxRSZ06tcaOHRun9htTp041B7b16tXT8uXLdfDgQW3ZskXDhg1T2rRpdebMGfXo0cM8G1+SevTooaNHj5r76JYtW9YcdMTHh3pJ4VptvB6yx/Y4wnrTz8abHD161Px1+fLlY3RMkf2sWvL1+/XXX+vZs2fq1auXtmzZot27d+t///uf0qZNq4cPH6pr165vbUcRnefakjW/7vDhw+av33///WhvJ5lmSof+Ae/atWsW+bmL6WvfEs9BfNX2urg+j4n9/Tyxi83vm3nz5pmD5woVKmjx4sXav3+/Nm7cqA4dOkgyvUeFfs/j+hqJyzghrscaH7XNmzcvyu/H6NGjY/T4r7fdMBgMKly4sJ4+faru3btHCJ7r1aunZcuWETwDABIEM58BAGbe3t7mr0NP/bSkHDlyaNKkSeZTcNu2bStJ2r59u7Zt2ybJFBoMHz7cvE2GDBk0aNAgFStWTF988YWuXLmimTNnaujQoeZ1/vzzT0lSpUqVwm2bMWNGde/eXYGBgZo6dar27t0rPz8/ZcyYMU7bvcmePXvMM5BGjRoVrm1CxowZVbBgQbm6uqp3794KCQnRnj17zC1KXhcYGKjly5eraNGi5mWNGzdW2rRp9fnnnys4OFhbt27VZ599JskUkK5du1aS1KVLl3Dfo0qVKqlMmTLq2LGjjh079tbjiI6yZcuqQ4cOWrx4sY4fP66FCxeqW7du0dr26tWr5gvddejQQSNGjDDflyFDBnXu3FnlypVTq1atdPbsWa1YscLcV9ze3l729vbm19GbLhpoCd7e3tq4caMk0wWcwobPcTmOsKL62Xiba9eumb/Oly9fjI/tdZZ8/T579kyenp5q2rSpeVnbtm1VuHBhtW/fXo8fP9b06dM1cuTIKOuJznNtyZpfFzqrXFKsLqwZ9jm5dOmS3n333RjvI6yYvvYt8RzEV22vi+vzmNjfz9/GaDTK39//res5OjrGywVPY/r75u7du5oxY4YkqWbNmpoxY4b5rBAXFxeNGDFCBoNBS5Ys0e+//65+/fopZ86csX6NxHWcEJdjja/aUqdOLUlx/l0WEBCgS5cuhVuWK1cu+fv7q1u3bhFmRffo0UMDBgyIVgshAAAsgZnPAAAzPz8/89ev9wINKzAwUP7+/lHenj9/Hul2tWvXNn/ICmv16tWSpEyZMmnw4MGRbvvhhx+aL5K2evXqcLNsAwICJEmPHj1SUFBQhG3btm2ruXPnasOGDeF6WMd2uzdxdnZWx44d1aBBgyivWl+hQgXz12G/56+rWrVquA/HoapUqWL+Pt64ccO8fMOGDTIajXJycor09G8HBwcNGzYsWscRXQMHDjSHLdOmTQsX1r3JqlWrFBISIkdHR3P/0dcVL17cfKG/VatWWabg1wQHB0f6Gvbz89PZs2c1d+5ctWvXTi9fvpTBYNCQIUPi5Tii+tl4mydPnpi/jk6Y9jaWfP2WL18+XOgZqnTp0uaZfhs3boxzGxhL1vy60DNBJCldunQxru2dd94xfx32FP+EklDPgSXE9XlM7O/nb3Pr1i2VKVPmrbfz58/H+jHeJKa/b7Zs2WJ+Xxw+fHiEdkSS6VoQefLkUfXq1eM8wz6u44SwYnqsCVlbbPj4+CgwMDDcMicnJ7Vp0yZc8Gxvby8vLy8NHDiQ4BkAkKCY+QwAMItuADFv3jxNmTIlyvtdXV3Ns4DCiuzK69KrU9tr1qwpe3v7KPf74Ycfatu2bXry5InOnTunYsWKSTK1+9i+fbvOnDmjVq1a6eOPP1b16tXNF7hycXHRBx98EGF/sd3uTcqVK6dy5cpFef+jR4/0999/m/8fWUgSKrKL2kmmD5Bp06aVn59fuKA/tDdt+fLl5ejoGOm2pUqVUubMmXXv3r03Hkd0hbbfaN++vV6+fKnhw4drxYoVb52ZF3qRpdCZoVHN+CtRooR+++03Xbp0KdqzFWPiyJEjKlOmzFvXs7Oz01dffRXh9WCp44jqZ+Ntwn6fXw8fYsOSr9/QwD0yNWvW1K+//moO+UN/lmPDkjW/7sWLF+avHRwcYlxb2EDOaDTGePu4SqjnwBLi+jwm9vfzxC6mv2/2798vyfTeF1UP7axZs+qPP/6wSH1xHSeEFdNjTcjaYuP1lhuRLcuUKZOmT58erd93AABYGuEzAMAsffr05q8tFU6GFdpPMaynT5+aZ2++7YJ1Ye+/ffu2+YNbu3bttHnzZp08eVKnT5/W6dOnJUl58+ZV1apVVbt2bVWoUCFCIBrb7aLr+PHjOnbsmK5cuaJr167p8uXLunXrVrgQ6k2B1JuC1tAPuGG3D+0p+7YL1eXNm9eiz2+5cuXUvn17LV26VMeOHdOPP/6oLl26vHGb69evSzJd+Cm6H4bv3Llj8fA5Kg4ODkqbNq3y5s2rsmXLqmXLluYQKyxLHUdkPxvREXY/cZl1GZm4vn4LFSoU5X1hWxPcuXPHYiFMXGt+XdgZsmFnQUdX2G3Cvr8mFGs8B5YQm+cxsb+fv01Uf7RNKDH9fXP37l1JinWbkZiwxDghrJgea0LWFhuRXWzwdd9//z3BMwDAagifAQBmYT8YvemU0169eqlXr14Rlg8ePFjr16+PcrvIZg6GnSnq5OT0xvrCzuYNu52jo6OWL1+uJUuW6JdffjH3Prx8+bIuX76spUuXytXVVd988024mW+x3e5t/v77b/3www86efJkhPty5MihqlWrauXKlW/dj52dXbQfU3rVgiGqWc+honvKeUwMGjRIO3fu1LVr1zRlyhTVrFnzjT2I33QxKEtu8zbly5fX0qVLY729pY4jNrNqpfA/s9evX1fFihWjvW1wcHCkp8pb6vX7ptdZ2Ndo2NYhsWWpml8Xts/zpUuXopwxGZWwffTfFkzFh4R8DiwhLs9jYn8/T+xi+vsm9A8rb/t9YwmWGCeEFdNjfRNL1xYbkc18fp2Xl9cbz4oCACA+ET4DAMxKlSolJycnPXv2TFu2bImyh60lhf2w9uzZszeu+6YPefb29uratau6du2qK1euaO/evdq3b5/2798vf39/3bx5U71799by5cvDBUix3S4qJ0+eVOfOnRUYGCgnJyfVqVNHJUuWVIECBVSoUCG5uLgoKCgoXsKK9OnTy9fX962nC4f2RrUkR0dHjR07Vh06dDC331i+fHmU66dOnVpPnz5VgwYNNHHiRIvXk1CsfRxhWwLs27dPLVu2jPa2rVu3lsFg0AcffKA+ffpIsuzr9+XLl1HeF/ZnPa6z2ePzZ65SpUrmr/fu3Rtp/+SoBAUFmduy5MyZU9mzZ4/RY4dt+RFb8fEcWKKuyFjieUxO7+eWFB/PWejF8mLSniK2LDVOiA+JobbX+4BnyJBBJUuW1M6dO83LLl68qNGjR+v777+32OMCABBdhM8AADN7e3t9+OGHWrNmjS5evKi///77jf0uLSFNmjRKly6dHj9+/NaL1YW9/01BTp48eZQnTx61a9dOAQEBWrFihb7//nsFBgZqxYoVUYYOsd0urMmTJyswMFBp06bVL7/8EukpyZZujxDq3Xffla+vb4Sr3r8utFWEpb3//vtq166dfvrpJ/3zzz9atGhRlOtmz55d3t7eb72ok9FoTNQXRrL2cWTPnl0lSpTQiRMntGvXLj148EAuLi5v3c7Hx0enTp2S0WjUu+++a15uydfvtWvXVLx48Ujvu3z5svnryNqZxER8/sxlz55dFSpU0MGDB7V582YNHjxYWbNmjda2mzZtMre3ady4cbj7ws44j6pXtyUuUBjT5yCh6oqMpZ/HpP5+Hl3Wes6yZ8+u8+fPv/X3yZw5c2Rra6uyZcuqVKlSsXqs+BgnWIq1a7t161aElkBFihTR999/r0aNGun+/fvm5WvWrFGFChVi9Ec0AAAsIXZNzwAAyVaPHj3Mp6R6eHjEqK1AbC62ZDAYVLZsWUnS9u3b3zgrN/TCRU5OTuZeps+fP1eXLl30wQcfaNmyZRG2sbe3V6dOnczrh/apjO12b3P06FFJUuXKlaPshRl6oSbJshchq1atmiTTaeJRhQ3Xrl3T1atXLfaYrxs0aJC55/SUKVN08+bNSNcLfc5Pnz6tO3fuRLm/kSNHqkKFCmrRokWE12JiCKUtcRxx1blzZ0mmdh6TJk2K1jbjx483v/batm1rXm7J1+/u3bujvG/r1q2STKFndNpRvOm5ju+fudBZ4QEBARo2bFi0Lszq6+urH374QZKULl06tW/fPtz9YU99jyq8DD2uyET3tR/T5yCudcWktqj2G5vnMTm+n0eXJZ6z2AjtH3zx4kXdunUr0nWePn2qqVOnysvLS/v27TMvj+lrJK7jhPhk7doia7lRuHBhZcqUKdJZzqNGjXprSA4AgKURPgMAwsmTJ48GDRokydT3uW3btuH6lkbm0aNH+v7777V58+ZYPWarVq0kSffv39f48eMjXWfLli3avn27JKlZs2bmgNzR0VF3797VnTt3tGrVqkhPM3/06JH5w3GuXLnitN3bhM5Cu3z5cqQh1e3bt8MdY1Qz1WKjSZMmsrOz08uXLzV27NgI94eEhOj777+P14DEyclJ3333nQwGg168eBFl+Bz6nAcFBWnUqFEKDg6OsM7x48f166+/6uHDh8qQIYPSpEkT7v7Q77Ulv4cxZYnjiCt3d3e9//77kqTVq1dr3LhxUT7HISEh8vT0NF/YrFatWuFaS1jy9bt+/fpI++QeOnRIGzdulGT6WY6ONz3X8f0zV758ebVr106SqfVGv3793tiv9erVq+rQoYN8fX0lSV999VWE2eg5c+Y0X/guNJAK68WLF5o7d26UjxHd135Mn4O41hWT2qLaLjbPY3J8P48uSzxnsdGkSRPZ2trKaDTK09Mz0vecWbNmKSgoSAaDQe7u7ublsXmNxGWcEN+sWVtU4bMkffDBBxH+8PXs2TP1798/3trnAAAQGcJnAEAEn376qfmCgt7e3mrRooW++OIL/fHHH7p48aIePnyoa9euafv27Ro5cqRq1qypRYsWyWg0ytnZWV26dInR49WqVUu1atWSJC1evFj9+/fX8ePH9ejRI/n4+GjSpEnq37+/JNMH7YEDB4bbPvTxzp8/r08//VS7d+/W3bt3dffuXe3cuVOfffaZnjx5Ihsbm3CzPGO73ZtUqVLF/H0bPHiwzp49Kz8/P/n4+Gj+/Plq2rRpuFl3lrzoUJ48edS1a1dJ0rp169S7d2+dPHlSDx8+1LFjx9S9e3dt27Yt0ovMWVKFChXe+v0qWrSoeZ1t27apY8eO2rNnjx48eKBr167pp59+Urdu3RQYGCgHBwcNGTIkwj4yZMggyfT8HTt2TH5+fgke/ljiOOLKYDBo4sSJ5kBt/vz5aty4sVauXKlz587p4cOHun79utatW6eWLVtq4cKFkqSCBQtGmBlnyddvYGCgPvvsM61cuVL//vuv7ty5o8WLF6tHjx4KCQlR7ty51a1bt2gd45ue64T4mRs+fLgaNGggSfrrr79Uv359zZkzR2fOnNGDBw909+5dHTx4UN9++60aNmxobn0zYMCASE9xT5cunfkPBtu2bdOoUaN0+fJl+fr6atu2bWrTpo3Onj2rdOnSxfj7EVZMn4O41hWT2l4X1+cxub2fR5clnrPYyJo1q3r06CFJ2rx5s3r27Kl//vlHfn5+OnfunMaMGWN+r2ndurXy5Mlj3jY2r5G4jhPikzVre73fs/QqfJakoUOHRphl7e3tre+++85iNQAA8Db0fAYAROqLL75Q6dKl9f333+vSpUvavHnzG2c2Ozo6qnHjxurdu3e0e6KGNX78eA0ePFjbtm3Tpk2btGnTpgjrFCtWTFOmTIkwc7RZs2Y6fvy4VqxYoSNHjpgD2LDs7Ow0ZswYubm5xXm7NxkyZIiOHDmie/fuacOGDdqwYUOEdWrUqGEOhC3dAqNfv3569OiRli9fri1btmjLli0RHjswMFB79+616OO+bvDgwdq5c2eUM58l04zQwMBA/d///Z/+/vvvSP9o4ezsrIkTJ4b7MB2qQoUKmjdvnp49e6bWrVtLkpYsWaIKFSpY7kCiIa7HYQlZsmTRihUrNHz4cO3cuVPe3t765ptvoly/bt26+u6775Q+ffpwyy35+q1fv762b9+ub775JkItBQsW1Jw5c+Tg4BCt43vTc50QP3O2traaMGGCSpYsqalTp+revXuaOHFilBeZdHV11YgRI8yBVGSGDx+uTp066eHDh1q+fHm4C3QaDAYNHTpUu3fvDtfWIVR0X/uxeQ7iUldMantdXJ/H5Ph+Hl1xfc5iq0+fPnr48KGWLVum7du3m2f2hlW3bl199dVX4ZbF9jUSl3FCfLNWba/PfLazswvXysjBwUHjx4/Xxx9/HK4lyOrVq1WhQgU1bNjQYrUAABAVZj4DAKJUvXp1bdy4UfPnz1ebNm1UtGhRpU2bVjY2NkqfPr3y5s2rhg0b6ttvv9XOnTs1evToWAXPkimcmzVrlmbOnKnatWsrc+bMsrOzU5YsWVS5cmV5enpq5cqV5n7Cr/vf//6nuXPnql69esqWLZvs7Ozk5OSkvHnzql27dvr9998jPcU/tttFJUeOHPr111/VoUMH5cqVS3Z2drK3t9e7776r2rVra8aMGZozZ45q1qwpydSLM+wFgeIqVapU+uabb7Rw4ULVqFFDmTNnloODgwoVKiQPDw/NnDkzQXolOzs7m9tvRMXOzk7fffedfvrpJzVs2FCurq6yt7dX6tSpVbBgQX322WfauHGjatSoEen21apV08iRI5UnTx7Z2dnJxcXF3OogIcX1OCzlnXfe0dy5c7V8+XK1a9dORYsWVbp06WRjYyMnJyflz59frVq10ooVKzR9+vQIwbNk2ddvtWrV9H//93+qV6+eMmTIIGdnZxUvXlwjRozQzz//LFdX12gf25ue64T6mTMYDOrUqZO2bdumESNGqGrVqub3KQcHB+XKlUvu7u6aNGmS/vzzzzcGz5JpduJvv/1mrtve3l4uLi6qXbu2fvrppzeeQRLd135snoO41BWT2l5niecxub2fR1dcn7PYSpUqlUaOHKmlS5fqo48+UpYsWWRnZ6d06dKpUqVKmjhxoqZPny57e/tw28X2NRLXcUJ8skZtT58+jXCx2wIFCkRo6eHm5qbBgwdH2H7kyJG6cuWKxeoBACAqBqM1rowBAACAZOfGjRuqXbu2JOnbb79Vy5YtrVxRysNzAAAAgMSEmc8AAAAAAAAAAIsjfAYAAAAAAAAAWBzhMwAAAAAAAADA4gifAQAAAAAAAAAWR/gMAAAAAAAAALA4g9FoNFq7CAAAAAAAAABA8sLMZwAAAAAAAACAxRE+AwAAAAAAAAAsjvAZAAAAAAAAAGBxhM8AAAAAAAAAAIsjfAYAAAAAAAAAWBzhMwAAAAAAAADA4gifAQAAAAAAAAAWR/gMAAAAAAAAALA4wmcAAAAAAAAAgMURPgMAAAAAAAAALI7wGQAAAAAAAABgcYTPAAAAAAAAAACLI3wGAAAAAAAAAFgc4TMAAAAAAAAAwOIInwEAAAAAAAAAFkf4DAAAAAAAAACwOMJnAAAAAAAAAIDFET4DAAAAAAAAACyO8BkAAAAAAAAAYHGEzwAAAAAAAAAAiyN8BgAAAAAAAABYHOEzAAAAAAAAAMDiCJ8BAAAAAAAAABZH+AwAAAAAAAAAsLgYh89bt25Vq1atVKZMGVWvXl1jx47V8+fPI113165daty4sUqXLq3mzZvr4MGDcS4YAAAAgOUtWrRIbm5uWrBgQZTrML4HAABATMQofL569aq++OIL3b17V61atVKOHDm0ePFiTZ48OcK6Pj4+6tWrl/z8/NSyZUvdvn1b3bp1040bNyxVOwAAAIA4CgoK0vz58+Xl5fXG9RjfAwAAIKZiFD5ny5ZN69at0+rVq9W6dWsVLVpUkpQ6deoI6/72228KDAxU3759NXz4cHXr1k0vX77Uhg0bLFM5AAAAgDhr0KCBJk6cqJw5c75xPcb3AAAAiCnbmKzs4OCg/PnzS5IqVKighw8fqnz58urdu3eEdb29vSVJefPmlSTzdhcuXIhTwQAAAAAsp0KFCpowYYK2b9+u6dOnR7ke43sAAADEVKwuOBgSEqLBgwerdevWOnTokL766qsI6zx79kySKbCWJHt7+3DLAQAAAFjf6NGjVbx48beux/geAAAAMRWr8DlVqlRq2bKlRo0apSxZsui3337TkydPwq3j6OgoSQoICJAkvXz5UpLk5OQU6T4NBoP8/f1jUw4AAACAeBbT8b3EGB8AACCli1H4vH37djVs2FCLFi2SJAUGBurly5eys7Mzz4AIFXoa3uXLlyVJV65ckSQVLFgwjiUDAAAASGiM7wEAABBTMer5XLRoUd29e1dTpkzR7du3debMGT169EiffPKJ7t27p3Xr1ilPnjxyd3dXo0aN9OOPP2rq1Km6ePGi1q9fL3t7e7m7u8fXsQAAAACwkJs3bzK+BwAAQJzEaOZz1qxZtXDhQhUvXlyrV6/WzZs31atXLw0bNkw3btzQlClTtG7dOklS4cKFNWfOHGXKlEmrV69W1qxZNXfu3LdeRRsAAACA9TG+BwAAQFwZjEaj0dpFSKZ+cE+fPpWzs7O1SwEAAABgAYzxAQAAUrZYXXAQAAAAAAAAAIA3IXwGAAAAAAAAAFgc4TMAAAAAAAAAwOIInwEAAAAAAAAAFkf4DAAAAAAAAACwOMJnAAAAAAAAAIDFET4DAAAAAAAAACyO8BkAAAAAAAAAYHGEzwAAAAAAAAAAiyN8BgAAAAAAAABYHOEzAABAIrZs2TLVr19fxYsXl5ubm5YvXy4/Pz+VKVNGZcqU0ZMnT8zrfvnll3Jzc1OjRo0i3dfQoUPl5uam1atXJ1T5AAAAQLxKzuPlQYMGyc3NTU2bNo31Pt52zPGN8BkAACCR2rp1q0aPHq0rV64oR44cKlmypEqUKKEFCxbI399fLVq0UNq0ac3rnz17VpJUuHDhSPf32WefSZJmzpypwMDA+D8AAAAAIB4ltfHy1q1b5ebmJjc3N/3222/m5Xfu3FH16tXl5uamjz/+WM+ePZMkGY1GlSxZUvXr1490f25ubrpx48YbH/NtxxzfbK3yqAAAAHir0AFpwYIFtX79ehkMBgUGBqpr166SpMaNG5vXDQwMlI+PjySpSJEike6vcOHCKliwoC5cuKAdO3aobt268XwEAAAAQPxJauPl2rVrq0iRIjp79qxmzZqlhg0byt/fX926ddPdu3eVK1cuzZkzR05OTpKkiRMnRnnMYY9t165dunLlijp27Bhu3egcc3xj5jMAAEAi8+jRI7m5uWnz5s2SpAsXLqhw4cIaNmyY9u3bJz8/P2XOnFnvvfeeeRsfHx/z7IzQgeXMmTPl5uam4sWLa+3atZKkmjVrSpI2bNiQgEcEAAAAWE5SHi/36tVLknTp0iWtXbtWffr0kbe3t1xcXDR//nxlypRJkmnGcugs6UOHDpm3t7W1lZeXl9q1aydJ8vDw0IABAyJ9rOgec3xi5jMAAEhRjEbpv7PYEoyTk2QwRH/927dvm2dESFKePHmUPn16VaxYUX///bckhRtIS69Op5OkQoUKaeTIkVq1apVcXFw0bdo0lStXTpJUokQJSTLvBwAAAAjLaDSa2z4kFCcnJxliMGBOyuPlunXrqlChQvL29taIESMUHBwsR0dHzZkzR7lz5460Xjc3N/PX7u7uKl++vFq3bi3JFLyvXr1a+fPnj/BY0T3m+ET4DAAAUgyjUapaVdq3L2Eft0oVaffu6AfQhQsX1ldffaX27dtLkiZNmqSiRYtKkvr16ydJcnV1DbfNuXPnJEkZM2bUiBEjtG3bNhUoUECzZ89Wzpw5zetlz55dknTv3j09e/bMfEofAAAAYDQaVbVqVe1L4AFzlSpVtHv37mgH0El5vGwwGNSjRw8NGjRIwcHBsrGx0eTJk82h9+v1urq6Kn369Oblmzdv1rfffqucOXPqxo0byp8/v1q3bq0BAwaYZ0PH9JjjE+EzAABIUWIyA9maQgeKtra2KlCggHl56NW6nZ2dw60fOqvBz89P27ZtU7FixbRkyRKlSZMm3Hph///kyRPCZwAAAIQTkxnI1mSt8fK0adM0ffr0N9b2f//3fxFmXocKDg7W77//bv5/SEhIhKA8bL1hZz1L0osXLzRo0CA1bdpUhQsX1rhx43Tx4kVdvnw5yn287ZjjE+EzAABIMQwG0wzkxN52Q3o1mM6XL5/s7e3Ny0MHiq+fCnn+/HlJ0jvvvCNfX19dunRJd+7cCTcQl6SnT59G2BcAAAAgmYLn3bt3J/q2G5L1xstFixbVxx9//MbaXFxcorxv1KhR2r59u+zt7WVnZyd/f3/NnDlTkyZNCreet7e3pIgXCmzatGmEfdaoUUM1atSIsDy6xxyfCJ8BAECKYjBIr02CSJRCB9Ovz3QI7QN369Yt87Lbt2/r4cOHkqSvvvpK8+fP1+nTp9W/f3/9/PPPcnR0NK8but0777wTYTYIAAAAYDAYksQ40Vrj5dq1a6t27dqxqnnmzJlatWqVDAaDvLy85O3trZkzZ2rz5s3q06ePuW/zrVu3zPUWLlw4yv2FhsuRickxx6dUCfIoAAAAiLbg4GBdvHhRUsTBZtmyZSVJp0+fNi8LeyGR9957TxMmTJCTk5MuXLigMWPGhNv+xIkT4fYDAAAAJDVJcbz866+/asqUKZKkL7/8Uh999JE6deokZ2dnhYSEaPbs2eZ1Q4N1KeLM5+iKyTHHJ8JnAACARObKlSt68eKFpIgzOapUqaIMGTLo9u3b5gFl6ODU2dlZOXLkUN68eTV8+HBJ0i+//KJ169aZt9+xY4ckqUGDBvF9GAAAAEC8SGrj5d27d2vEiBGSpE8//VSdO3eWJGXIkMF8kcANGzbo6tWrkdYbGzE55vhkMBqNxgR5pLcwGAx6+vRpkpjWDwAAYE3jxo3T/Pnz9emnn+rLL7+M9nbnzp1TkyZN9O677+rPP/8M1xsPiA+M8QEAgDUwXk48mPkMAACQxHTt2lXOzs76+eefzVfzjo6FCxdKknr27MlAGgAAAMkW4+XEg5nPAAAAAOIFY3wAAICUjZnPAAAAAAAAAACLI3wGAAAAAAAAAFgc4TMAAAAAAAAAwOIInwEAAAAAAAAAFkf4DAAAAAAAAACwOMJnAAAAAAAAAIDFET4DAAAAAAAAACyO8BkAAAAAAAAAYHGEzwAAAMnAoEGD5ObmpqZNm8ZpP19++aXc3NzUqFEjyxQGAAAAIMUifAYAAEiEtm7dKjc3N7m5uem3334zL79z546qV68uNzc3ffzxx3r27JkkyWg0qmTJkqpfv36U+3Rzc9ONGzfe+Lhnz56VJBUuXNgCRwEAAAAgJbO1dgEAAACIqHbt2ipSpIjOnj2rWbNmqWHDhvL391e3bt109+5d5cqVS3PmzJGTk5MkaeLEiZHuJzS4bty4sXnZrl27dOXKFXXs2DHcuoGBgfLx8ZEkFSlSJD4OCwAAAEAKwsxnAACARKpXr16SpEuXLmnt2rXq06ePvL295eLiovnz5ytTpkySTLOVQ2dJHzp0KNw+bG1t5eXlpXbt2kmSPDw8NGDAgEgfz8fHR4GBgZJehc8zZ86Um5ubihcvrrVr18bHYQIAAABIppj5DAAAUhSj0WhuVZFQnJycZDAYYrxd3bp1VahQIXl7e2vEiBEKDg6Wo6Oj5syZo9y5c5vXC22VIZlaa4Tl7u6u8uXLq3Xr1pKkCxcuaPXq1cqfP3+Exwu7n0KFCmnkyJFatWqVXFxcNG3aNJUrVy7GxwAAAAAg5WLmMwAASDGMRqOqVq2qNGnSJOitWrVqMhqNMa7XYDCoR48ekqTg4GDZ2Nho8uTJKlGiRLj1zp07J0lydXVV+vTpw923efNmNW3aVFmyZJEk5c+fX61bt9ayZcsiPF7ofjJmzKgRI0Zo1apVKlCggFavXk3wDAAAACDGmPkMAABSlNjMQLaW4OBg/f777+b/h4SEyNXVNcJ6oTOWX5/1LEkvXrzQoEGD1LRpUxUuXFjjxo3TxYsXdfny5Sj34+fnp23btqlYsWJasmSJ0qRJY6lDAgAAAJCCED4DAIAUw2AwaPfu3Umm7caoUaO0fft22dvby87OTv7+/po5c6YmTZoUbj1vb29JkV8ksGnTphGW1ahRQzVq1Iiw/Pz585Kkd955R76+vrp06ZLu3LmjAgUKxLh2AAAAAKDtBgAASFEMBoOcnZ0T9Bab4HnmzJlatWqVDAaDvLy81KlTJ0mmNho+Pj7m9W7duqWHDx9KkgoXLvzGfZ4/f145cuSI9L7bt2+b9/PVV1+pWLFiev78ufr376/nz5/HuH4AAAAAIHwGAABIZH799VdNmTJFkvTll1/qo48+UqdOneTs7KyQkBDNnj3bvG5on2Yp8pnP0RX2YoPvvfeeJkyYICcnJ124cEFjxoyJ9X4BAAAApFyEzwAAAInI7t27NWLECEnSp59+qs6dO0uSMmTIoHbt2kmSNmzYoKtXr0p6FT47OztHOas5Ol7fT968eTV8+HBJ0i+//KJ169bFet8AAAAAUiaDMTaXXo8HBoNBT58+lbOzs7VLAQAAAGABjPEBAABSNmY+AwAAAAAAAAAsjvAZAAAAAAAAAGBxhM8AAAAAAAAAAIsjfAYAAAAAAAAAWBzhMwAAAAAAAADA4gifAQAAAAAAAAAWR/gMAAAAAAAAALA4wmcAAAAAAAAAgMURPgMAAAAAAAAALI7wGQAAAAAAAABgcYTPAAAAAAAAAACLI3wGAAAAAAAAAFgc4TMAAAAAAAAAwOIInwEAAAAAAAAAFkf4DAAAAAAAAACwOMJnAAAAAAAAAIDFET4DAAAAAAAAACyO8BkAAAAAAAAAYHGEzwAAAAAAAAAAiyN8BgAAAAAAAABYHOEzAAAAAAAAAMDiCJ8BAAAAAAAAABZH+AwAAAAAAAAAsDjCZwAAAAAAAACAxcUofD5x4oQ6duyoMmXKqFKlSho4cKDu3r0b6brDhg2Tm5tbuNukSZMsUjQAAAAAAAAAIHGzje6Kfn5+6tq1qwIDA9W6dWvduHFDGzZs0K1bt7Ry5coI6588eVLp0qXTp59+al5Wrlw5y1QNAAAAAAAAAEjUoh0++/j4KEeOHKpRo4b69esno9GoihUr6tixYwoICJC9vb153WfPnsnHx0fly5dXmzZtlDp1ajk5OcXLAQAAAAAAAAAAEp9oh8/lypXTmjVrzP/39vbW48ePlStXrnDBsySdPn1aISEhOnbsmCpVqiSDwaB69epp7NixSpMmjeWqBwAAAAAAAAAkStEOn8M6d+6cunXrppCQEPXv3z/C/b6+vnJ1dVWpUqVUtWpV/fnnn/rjjz+UIUMGjR49Oq41AwAAAAAAAAASOYPRaDTGZINdu3apf//+ev78ub766iu1b9/+rds8evRI5cuX17vvvqsdO3ZEXojBoKdPn8rZ2Tkm5QAAAABIpBjjAwAApGypYrLy5s2b1bNnT4WEhGj69OlRBs+bNm3SN998o7Nnz0qSHj9+LEn0fQYAAAAAAACAFCLabTeuX7+uoUOHKigoSBUrVtT58+d1/vx5SVKTJk20bt065cmTR+7u7jIajVq5cqV27dqlhg0bmmc7t2vXLl4OAgAAAAAAAACQuEQ7fF69erVevnwpSdqzZ4/27Nljvi9fvnyaMmWKatSoIXd3d7m7u8vf318LFy7UokWLlCVLFg0bNozwGQAAAAAAAABSiBj3fI4v9IMDAAAAkhfG+AAAAClbjHo+AwAAAAAAAAAQHYTPAAAAAAAAAACLI3wGAAAAAAAAAFgc4TMAAAAAAAAAwOIInwEAAAAAAAAAFkf4DAAAAKRw9+7dU79+/VS+fHnVrFlTU6dOVXBwcIT1QkJCNHnyZNWoUUMlSpRQ/fr1tWzZMitUDAAAgKSA8BkAAABI4b744gv98ccfqlOnjrJnz64ZM2ZowYIFEdZbsWKFZs2apdy5c6t9+/ZKlSqVRo8erb///tsKVQMAACCxI3wGAAAAUrArV67oyJEjKl26tMaOHatp06ZJkn755ZcI616+fFmSVLNmTTVs2FDvvfeeJMnOzi7hCgYAAECSYWvtAgAAAABYz4ULFyRJefPmlSS5uLjIxcVFV69eVWBgYLhguU2bNvrjjz/0/fffm5cNHz5cJUuWTNiiAQAAkCQw8xkAAABIwfz9/SVJDg4O5mUODg4yGo16/vx5uHXTpEmj7Nmzq1KlSho7dqzy5cunCRMm6MSJEwlaMwAAAJIGZj4DAAAAKZiTk5MkKSAgwLzsxYsXMhgMcnR0DLfu6NGjdezYMW3evFl58+ZV2rRp1bdvXy1btkwlSpRI0LoBAACQ+DHzGQAAAEjB8ufPL8nU+1mSHj58KD8/P+XOnTtCL+fQns9PnjyRJD179kySZG9vn0DVAgAAIClh5jMAAACQguXPn1/FixfXkSNHNHz4cF27dk2S1Lx5c928eVPr1q1Tnjx55O7ururVq+vSpUsaMmSI6tSpo7Vr18pgMMjd3d3KRwEAAIDEyGA0Go3WLkKSDAaDnj59KmdnZ2uXAgAAAKQovr6+Gj16tA4cOCBHR0c1bdpUffv21ZEjR9SxY0fVqFFDc+bMUWBgoObMmaNff/1Vvr6+ypUrl3r27Bll+MwYHwAAIGUjfAYAAAAQLxjjAwAApGz0fAYAAAAAAAAAWBzhMwAAAAAAAADA4gifAQAAAAAAAAAWR/gMAAAAAAAAALA4wmcAAAAAAAAAgMURPgMAAAAAAAAALI7wGQAAAAAAAABgcYTPAAAAAAAAAACLI3wGAAAAAAAAAFgc4TMAAAAAAAAAwOIInwEAAAAAAAAAFkf4DAAAAAAAAACwOMJnAAAAAAAAAIDFET4DAAAAAAAAACyO8BkAAAAAAAAAYHGEzwAAAAAAAAAAiyN8BgAAAAAAAABYHOEzAAAAAAAAAMDiCJ8BAAAAAAAAABZH+AwAAAAAAAAAsDjCZwAAAAAAAACAxRE+AwAAAAAAAAAsjvAZAAAAAAAAAGBxhM8AAAAAAAAAAIsjfAYAAAAAAAAAWBzhMwAAAAAAAADA4gifAQAAAAAAAAAWR/gMAAAAAAAAALA4wmcAAAAAAAAAgMURPgMAAAAAAAAALI7wGQAAAAAAAABgcYTPAAAAAAAAAACLI3wGAAAAAAAAAFgc4TMAAAAAAAAAwOIInwEAAAAAAAAAFkf4DAAAAAAAAACwOMJnAAAAAAAAAIDFET4DAAAAAAAAACyO8BkAAAAAAAAAYHGEzwAAAAAAAAAAiyN8BgAAAAAAAABYHOEzAAAAAAAAAMDiCJ8BAAAAAAAAABZH+AwAAAAAAAAAsDjCZwAAAAAAAACAxRE+AwAAAAAAAAAsjvAZAAAAAAAAAGBxhM8AAAAAAAAAAIsjfAYAAAAAAAAAWBzhMwAAAAAAAADA4gifAQAAAAAAAAAWF6Pw+cSJE+rYsaPKlCmjSpUqaeDAgbp7926k6+7atUuNGzdW6dKl1bx5cx08eNAiBQMAAAAAAAAAEj+D0Wg0RmdFPz8/1a9fX4GBgWrdurVu3Lihv/76S6VLl9bKlSvDrevj46MmTZooY8aM+uijj7R+/Xr5+/tr48aNypEjR+SFGAx6+vSpnJ2d435UAAAAAKyOMT4AAEDKZhvdFX18fJQjRw7VqFFD/fr1k9FoVMWKFXXs2DEFBATI3t7evO5vv/2mwMBA9e3bV61atVK2bNnk6empDRs2qEePHvFyIAAAAAAAAACAxCPa4XO5cuW0Zs0a8/+9vb31+PFj5cqVK1zwHHqfJOXNm1eSlD9/fknShQsX4lwwAAAAAAAAACDxi9UFB8+dO6euXbsqJCRE/fv3j3D/s2fPJEkODg6SZA6nQ5cDAAAAAAAAAJK3GIfPu3bt0ieffCJfX199/fXXcnd3j7COo6OjJCkgIECS9PLlS0mSk5NTXGoFAAAAAAAAACQRMQqfN2/erJ49eyokJETTp09X+/btI10vtM3G5cuXJUlXrlyRJBUsWDAOpQIAAAAAAAAAkgqD0Wg0RmfF69evq0GDBnr58qWqVq2qsmXLmu9r0qSJ1q1bpzx58sjd3V3nzp1T8+bNlSlTJrm7u2v9+vV68uSJNm7cqJw5c0ZeCFfCBgAAAJIVxvgAAAApW7QvOLh69Wpz+4w9e/Zoz5495vvy5cunKVOmqEaNGnJ3d1fhwoU1Z84cTZgwQatXr1aePHk0YcKEKINnAAAAAAAAAEDyEu2Zz/GNWREAAABA8sIYHwAAIGWL9sxnAAAAAEDi9+TJE129elXXrl2Tr6+vHjx4oAcPHsjPz08PHjzQ8+fPFRgYqICAAPO/qVKlkr29vflmZ2cnZ2dnZcyYURkyZFCGDBmUMWNGvfPOO8qePbtcXV2VKVMmpUoV42vYAwCAFITwGQAAAACSmMDAQF24cEGnTp3SqVOndPr0aV2+fFlXr17VgwcPEqQGOzs7vfvuu8qZM6cKFCigggULqkCBAuZb+vTpE6QOAACQeNF2AwAAAEC8YIxvObdu3dLevXt14MABHThwQEeOHDFfkycyGTNmVO7cuZUlSxa5uLjIxcVFGTNmVMaMGeXs7Cw7O7tws5yNRqMCAgLMt5cvX8rf319+fn56+PCh/Pz85Ofnp3v37unWrVv6999/31pz5syZ9d5776lkyZLmW5EiReTg4GDJbw0AAEjECJ8BAAAAxAvG+LHn7++v7du3a8uWLfrrr7905syZCOukSZNGxYsXV/HixVWsWDEVKFBAuXPnVu7cuZUuXbp4rS8gIEB3797VrVu3dOXKFV28eFEXLlww/xtVOG1ra6vChQurXLlyqlixoipVqqRixYrJxsYmXusFAADWQfgMAAAAIF4wxo+Zhw8fav369VqzZo02b96sFy9emO8zGAwqVaqUKlWqpIoVK6pixYoqUKCADAaDFSuO2uPHj+Xt7a0TJ07o+PHj5tvDhw8jrJsmTRqVL19eFStWVOXKlVWtWrV4D88BAEDCIHwGAAAAEC8Y479dYGCgNm3apIULF2rjxo0KDAw035cnTx7Vr19fdevWVc2aNeXi4mLFSuPOaDTq+vXrOnbsmA4dOqQDBw7o4MGDevr0abj1bGxsVLZsWdWqVUs1a9ZUlSpVeA0BAJBEET4DAAAAiBeM8aN26dIlzZo1S0uWLAnXoqJYsWJq3ry5WrRooRIlSiTamc2WEhwcrDNnzmj//v3av3+/du/eLR8fn3Dr2NnZqUKFCqpfv77c3d1VqlQppUqVykoVAwCAmCB8BgAAQLz4999/NX36dI0ePdrapcBKGOOHZzQatW/fPk2cOFFr165VSEiIJClLlizq0KGDOnfurOLFi1u5Suu7du2atm/fru3bt2vbtm26fv16uPuzZcumjz76SO7u7qpbt67Sp09vpUoBAMDbED4DAADAoi5duqQJEyZo4cKFevHihRLJcBNWwBjfxGg0avPmzRo1apQOHjxoXl6/fn317NlT7u7usrOzs2KFiZfRaNSlS5e0ZcsWbdq0SVu2bJG/v7/5fhsbG1WvXl3NmzdX06ZNlSNHDitWCwAAXkf4DAAAAIv4559/5OXlpdWrV5tndL7//vs6dOiQlSuDtTDGl7Zv364RI0Zo3759kiQHBwd16NBB/fv3V7FixaxcXdLz8uVL7dmzRxs3btTGjRt17ty5cPdXqFBBzZs3V/PmzVWgQAErVQkAAEIRPgMAACDWjEajtm/fLk9PT/3555/m5R9++KE8PDz0wQcfJPuetYhaSh7jnzp1Sv3799fWrVslSalTp1bv3r01ZMgQZc2a1crVJR+XLl3S2rVrtWbNGu3bty/cmRYlSpRQ8+bN1bp1axUuXNiKVQIAkHIRPgMAACDGgoODtWbNGnl5eenvv/+WZDr9vXXr1ho6dKhKlixp5QqRGKTEMf7Dhw/1zTffaMaMGQoODpadnZ169Oih4cOH691337V2ecna7du3zUH09u3bFRwcbL6vTJkyateundq0aaPs2bNbsUoAAFIWwmcAAABE24sXL7R48WKNHz9eFy9elCQ5OjqqS5cuGjRokPLkyWPdApGopKQxvtFo1JIlSzRkyBDdu3dPktS8eXNNmDCBnwsruH//vn7//XetXr1af/zxhzmINhgMqlWrlj755BO1aNGCixUCABDPCJ8BAADwVg8fPtSsWbM0ZcoU3b17V5Lk4uKiPn36qE+fPsqcObOVK0RilFLG+Ldv31b37t31+++/S5IKFy6sqVOnqm7dulauDJJ07949/fzzz1q2bJm597Zk6r/dqFEjffbZZ6pXr55sbGysWCUAAMkT4TMAAACidPPmTU2ePFlz5szRkydPJEm5cuXSwIED1bVrV8ZueKPkPsY3Go1asWKF+vTpIz8/P9nb22vUqFEaNGiQ7OzsrF0eInH58mWtWLFCy5Yt05kzZ8zLXV1d1blzZ3322WfKly+fFSsEACB5IXwGAABABOfOndO4ceO0dOlSBQYGSpKKFy8uDw8PtW7dmmAN0ZKcx/hPnz5V9+7dtWLFCkmmnsKLFy9W8eLFrVwZosNoNOr48eNatGiRli5dqgcPHpjvq1Gjhrp06aIWLVrI0dHRilUCAJD0ET4DAADA7MCBA/L09NS6desUOkysXr26PDw89NFHH8lgMFi5QiQlyXWMf/bsWbVo0UJnz56VjY2NRo4cqWHDhvFHmSTq5cuX+u2337RgwQL9+eef5ve+9OnTq3379urZs6eKFStm5SoBAEiaCJ8BAABSOKPRqE2bNsnT01O7du0yL2/atKk8PDxUsWJFK1aHpCw5jvFXrlyprl27yt/fX9mzZ9fq1atVpUoVa5cFC7l27ZoWLVqkH3/8UVeuXDEvr1Gjhnr16qWmTZvyRwYAAGKA8BkAACCFCgwM1KpVq+Tl5aWTJ09Kkuzs7NShQwcNHjxYRYoUsXKFSOqS0xg/JCREw4cPl6enpySpVq1aWr58ubJmzWrlyhAfQkJCtG3bNs2aNUvr1q1TcHCwJClbtmzq3r27unXrphw5cli5SgAAEj/CZwAAgBTG399f8+fP18SJE3Xt2jVJUpo0afT555+rf//+cnV1tXKFSC6Syxj/5cuX6ty5s1auXClJGjZsmMaMGSMbGxsrV4aEcOPGDc2dO1fz5s3TnTt3JEk2NjZq0qSJvvjiC1WrVo2WRAAARIHwGQAAIIXw9fXVtGnTNH36dPPFtbJmzaovvvhCPXv2VIYMGaxbIJKd5DDG9/PzU9OmTbVr1y7Z2tpqwYIF6tixo7XLghUEBARo7dq1mjlzpnbu3GleXqZMGQ0YMECtWrWSvb29FSsEACDxIXwGAABI5q5cuaIJEyZowYIFev78uSQpf/78GjJkiDp16qTUqVNbuUJY27179zRmzBgdOHBAzs7OatasmXr37h3pzN5z587J09NTR48elZOTk5o2barBgwdHum5SH+PfuHFD9erV09mzZ5UuXTqtWbNGtWvXtnZZSAROnz6tadOmafHixXrx4oUkKXv27OrTp4969OghFxcXK1cIAEDiQPgMAACQTB0/flxeXl5atWqVuV9p2bJl5eHhoebNm9MyAGaffPKJjhw5ohYtWujq1av6+++/NWjQIHXv3j3cev/++68aNmwog8Gg5s2b69ChQzp16pS+/PJLffrppxH2m5TH+FeuXFGtWrV0+fJlubq6atOmTXrvvfesXRYSGV9fX82ZM0fTp083t+RwcnJSp06d1L9/fxUqVMjKFQIAYF2EzwAAAMmI0WjUzp075enpqc2bN5uX161bVx4eHqpVqxa9SRHOlStXVL9+fZUuXVorV67UgwcPVKlSJeXJk0d//PFHuHVnz56tSZMm6dtvv1WDBg306NEj+fr6ytXVNdKZnkl1jO/j46NatWrp2rVryp8/v7Zt26ZcuXJZuywkYi9fvtSqVas0adIkHTt2zLy8YcOG8vDwUNWqVa1XHAAAVpTK2gUAAAAg7oKDg7VmzRpVrFhRNWvW1ObNm5UqVSq1bt1aR44c0Z9//qnatWsTPCOCCxcuSJLy5s0rSXJxcZGLi4uuXr2qwMDAcOuePn1akrR161aVK1dONWvW1KJFi5JcuPwm165dU82aNXXt2jUVKlRIO3fuJHjGWzk4OKhjx446evSotm3bpkaNGkmSfv/9d1WrVk3VqlXTxo0blUjmfgEAkGAInwEAAJKwly9fav78+SpatKhatGihQ4cOKXXq1OrZs6e8vb21cuVKlSlTxtplIhHz9/eXZArPQjk4OMhoNJp7hId68uSJJNPM4NGjR6tmzZr6/fffNXPmzIQrOB7dvXtXderU0fXr1+Xm5qadO3fK1dXV2mUhCTEYDKpZs6Z+++03nT9/Xt26dZO9vb327NmjBg0amM8wCG2FBABAckf4DAAAkAQ9evRInp6eypMnj7p16yZvb29lzJhRI0aM0NWrVzVz5kzlz5/f2mUiCXBycpIkBQQEmJe9ePFCBoNBjo6O4dYN/X+vXr308ccfa+zYsZKkXbt2JVC18cfPz0/16tXThQsXlDt3bm3ZskXZsmWzdllIwgoVKqS5c+fq0qVLGjRokJydnXX8+HG1bdtWbm5umjt3rl6+fGntMgEAiFeEzwAAAEnI7du39eWXXypXrlz68ssvdefOHeXIkUMTJ07U1atXNWbMGGXJksXaZSIJCf0jxZUrVyRJDx8+lJ+fn3Lnzi07O7tw67q5uUkyXWRNknn25ushdVITEBCgFi1a6MSJE8qWLZu2bNmiHDlyWLssJBOurq4aP368rl27plGjRilTpkzy8fFRjx49lDdvXo0fP958BgIAAMkNFxwEAABIAry9vTV+/HgtXrzYPEO1aNGiGjp0qNq2bSt7e3srV4ikrEWLFjp16pRatGiha9eu6fDhwxo4cKAaNmyodevWKU+ePHJ3d9f169fVqFEjGQwGtWzZUidPntTRo0c1cuRItWvXLsJ+k8IY32g0qnPnzlqyZInSpk2r3bt3q2TJktYuC8mYv7+/5s2bpwkTJujGjRuSpMyZM2vIkCHq1atXov55AQAgpgifAQAAErHDhw/L09NTa9asMV+oqkqVKvLw8FCDBg2UKhUnsiHufH19NXr0aB04cECOjo5q2rSp+vbtqyNHjqhjx46qUaOG5syZI0k6cuSIPD09dfbsWWXNmlVt2rRRly5dIr2YZVIY448ZM0YjR46UjY2NNmzYoPr161u7JKQQAQEB+umnnzR27Fj5+PhIkt555x1zCJ0mTRorVwgAQNwRPgMAACQyRqNRf/zxhzw9PbVjxw7z8kaNGsnDw0NVqlSxXnFADCT2Mf66devUtGlTSdKcOXPUvXt36xaEFCkoKEjLli3TmDFjCKEBAMkO4TMAAEAiERQUpNWrV8vLy0vHjx+XJNna2qpdu3YaMmSIihUrZuUKo8/PT7p40XRr29ba1cBaEvMY//z583r//ff15MkT9evXT1OmTLF2SUjhgoKCtHz5co0ZM0YXL16URAgNAEj6CJ8BAACs7NmzZ1q4cKEmTJhgvuibs7OzunfvrgEDBihnzpzWLTAKDx5IFy68CpnDfn3//qv1EsdoE9aQWMf4T548UYUKFXT27FlVr15dW7ZsiXBxRcBaogqhhw8frp49eyp16tRWrhAAgOgjfAYAALCS+/fva8aMGZo2bZp8fX0lmS461a9fP/Xq1UsuLi5Wrc9oDB8wvx40+/m9efvs2aUCBaSdOxOmXiQ+iXWM3759ey1btkzZs2fX0aNHlTVrVmuXBEQQFBSkFStWaMyYMbpw4YIkKUeOHPrmm2/UuXNn2draWrlCAADejvAZAAAggV27dk0TJ07UvHnz9OzZM0lSvnz5NHjwYHXu3FmOjo4JWs/Dh6Yw2ds7/L8XL5ruexNXV1PAXLCg6d/Qr/PnlxjWITGO8ZcuXaqOHTvKxsZGu3btUuXKla1dEvBGQUFBWrx4sf73v//pxo0bkqSCBQtqzJgxatmyJReeBQAkaoTPAAAACeTUqVPy8vLSihUrFBQUJEkqXbq0PDw81KJFi3idxfbsmSlMDhswh359796bt82RI/KAOV8+Ama8WWIb41+8eFGlS5fW06dPNWbMGI0YMcLaJQHR9uLFC82ePVvfffed+WyZUqVK6bvvvtNHH30kg8Fg5QoBAIiI8BkAACAeGY1G7dmzR56entqwYYN5ee3ateXh4aE6depYLDAICJAuXYp8FvN/k+WilC2bVKiQKVQO/Tc0YHZyskh5SIES0xg/MDBQVapU0eHDh/XBBx9o69atsrGxsXZZQIw9efJEkydP1vjx4/X48WNJUtWqVTV27FhVq1bNytUBABAe4TMAAEA8CAkJ0fr16+Xp6an9+/dLklKlSqUWLVpo6NChKleuXKz2GxwsXbsWMVz29pauXJFCQqLe1sUlfLgcNmROmzZW5QBvlJjG+GPHjtVXX32ljBkz6vjx44n2Qp5AdN2/f18//PCDpk+frhcvXkiSPvzwQ3l6eqpEiRJWrg4AABPCZwAAAAsKCAjQTz/9pHHjxuncuXOSJAcHB3Xu3FmDBw9WgQIF3roPo1G6fTvygNnHxzTDOSrOzhHD5dB/M2Wy1FEC0ZNYxvinT59WmTJlFBAQoKVLl6p9+/ZWrQewpJs3b2rMmDFasGCBgoKCZDAY1LlzZ40ZM0aurq7WLg8AkMIRPgMAAFjA48ePNXfuXE2aNEm3bt2SJKVPn169evVSv379lC1btgjb+PlJ58+bbmFD5gsXJH//qB/LwcF0Qb/IQuZs2STafiKxSAxj/KCgIFWuXFmHDx9Ww4YN9dtvv9EbF8nSxYsXNXz4cP3888+SJEdHRw0aNEhDhw5VWk5vAQBYCeEzAABAHNy9e1dTpkzRzJkz9ejRI0lS9uzZNWDAAHXv3l2Ojul06dKrkDns7U0X+rOxkfLmjXwGc86cpvuBxC4xjPHHjx+vIUOGKH369Dp9+jQzQZHs7d+/X4MHD9a+ffskSVmyZNH//vc/devWLV4vbAsAQGQInwEAAGLh4sWLGj9+vBYtWqSXL19KknLlclOtWkOVMWM7+fg46Nw50wUAg4Ki3o+rq+TmZgqWw4bMefJI9vYJcyxAfLH2GP/GjRsqXLiw/P39NX/+fHXp0sUqdQAJzWg06tdff5WHh4cuXrwoSSpcuLA8PT3VqFEjZv8DABIM4TMAAEA0vXwprVt3RJMne+rAgV9kNJqu7mdjU1HBwR6SGktKFWE7JydTwPz6rVAhKU2ahD0GICFZe4zfunVrrV69WpUrV9bu3buVKlXEn08gOQsICNCcOXM0atQo3b9/X5JUvXp1jR8/Xu+//76VqwMApASEzwAAAGEYjdKdO+HbY5w9a9SxY1t0546npK1h1naX5CGpmgwGg3LlijxkdnWVyLyQEllzjL9lyxbVrVtXqVKl0pEjR1SqVKkErwFILB49eqQffvhBkydP1osXLyRJHTp00Pfff08rGgBAvCJ8BgAAKdLz56YL/EXWi/nJk9C1giT9IslT0j//LbNRpkxtVbHiUFWs+J45YC5YUHJ0tMaRAImXtcb4AQEBKlGihM6fP6++fftq6tSpCfr4QGJ17do1jRgxQkuXLpUkOTk5afjw4Ro4cKAc+SUGAIgHhM8AACBZe/BAOnvWdDt37tXXV66YZjlHxmB4rkyZftSzZxP07NklSVLq1E5q376bvvpqgPLkyZ1wBwAkYdYa40+fPl19+/ZVlixZdP78eWXIkCFBHx9I7A4fPqwvvvhC+/fvlyTlzp1b48ePV4sWLegHDQCwKMJnAACQ5BmN0vXr4cPl0LD533+j3i5jxletMQoXllxd/XT48AytXDlV9+7dkyRlypRJ/fr1U+/evZUpU6YEOiIgebDGGP/x48fKnz+/fH19NWvWLH3++ecJ9thAUmI0GrVixQp5eHjoxo0bkkz9oKdMmUKbGgCAxRA+AwCAJCMwULp4MeJM5nPnJH//qLfLmVMqUsR0K1z41deZM0sGg3T9+nVNmjRJc+fOlf9/O8qTJ48GDRqkzz77TE5OTgl0hEDyYo0x/ogRI/Tdd9/Jzc1NJ0+elJ2dXYI9NpAU+fv7y8vLS15eXnrx4oUMBoO6deumMWPGKEuWLNYuDwCQxBE+AwCAROfJE1Og/PpMZh8fKSgo8m1sbU19l18PmN3cpDRpIt/mzJkz8vLy0rJlyxT0345LlCghDw8PtWrVSra2tvF0hEDKkNBj/Fu3bqlAgQJ6/vy51qxZo2bNmiXI4wLJwbVr1+Th4aGVK1dKktKlS6eRI0eqb9++sre3t3J1AICkivAZAABYhdFoaonxepuMs2el/87+jVSaNOHD5dCwOX9+KboTHPfu3StPT0+tX7/evKxGjRry8PBQ/fr16XcJWEhCj/F79OihuXPnqnLlytqzZw8/y0As7N69W/3799fRo0clSYULF9b06dNVu3ZtK1cGAEiKCJ8BAEC8MhqlW7ekM2ek06df/Xv2rOTnF/V2WbNG3irD1dXUKiOmQkJCtGHDBnl6emrv3r2STOOPZs2aycPDQ+XLl4/lEQKISkKO8a9evaqCBQsqMDBQu3fvVtWqVeP9MYHkKjg4WIsXL9awYcP0738XT2jVqpUmTJigHDlyWLk6AEBSQvgMAAAswmiUbt6MGDKfOSM9ehT5NgaDlDdvxFnMRYqYLgZoCQEBAVqxYoW8vLx05swZSZK9vb06duyowYMHy83NzTIPBCCChBzj9+zZU7Nnz1bt2rW1ZcuWeH88ICV4+PChRo4cqRkzZigkJETOzs765ptv9MUXX9CKAwAQLYTPAAAgRoxGU1uMyELmx48j38bGRipQQCpWTCpa9NW/hQpJqVPHT51Pnz7VvHnzNHHiRN34r49HunTp9Pnnn6t///5699134+eBAZgl1Bj/+vXryp8/vwIDA7Vz505Vr149Xh8PSGmOHTum3r17a9++fZKkIkWKaPr06apVq5aVKwMAJHaEzwAAIFJGo3T9euQh85MnkW9jY2O66F9kIbODQ8LU/e+//2ratGmaMWOG/P7r65EtWzYNGDBAPXr0UPr06ROmEAAJNsbv06ePZsyYoRo1amj79u3x+lhAShUSEqKlS5dqyJAhunfvniSpdevWmjBhglxdXa1cHQAgsSJ8BgAghQsNmV8PmN8UMtvaRh0yW+ss3EuXLmnChAlauHChXrx4IUkqVKiQhgwZog4dOsghodJvAGYJMca/ffu28uTJo4CAAG3btk01a9aMt8cCYGrF8fXXX2vmzJm04gAAvBXhMwAAKYTRKN2+LZ08KZ06ZbqFhsxPn0a+ja2tKVAOGzAXK2YKnhPL58t//vlHXl5eWr16tUJCQiRJ5cuXl4eHh5o0aSIbGxsrVwikXAkxxv/qq680duxYVa5cWXv27JEhNlckBRBjx44dU69evbR//35JplYcM2bM4A9AAIBwCJ8BAEiG/PxMM5hDg+bQf//rQhGBra3k5hYxZC5QIPGEzGEZjUZt375dnp6e+vPPP83LP/zwQ3l4eOiDDz4ggAISgfge4/v7+ytXrlx68OCB1qxZo2bNmsXL4wCIXEhIiJYsWaKhQ4eaW3F06NBBEyZMUObMma1cHQAgMSB8BgAgCXv+XDp7Nvxs5pMnpZs3I18/VSrTTObixU230KC5YEHJzi5ha4+N4OBgrVmzRl5eXvr7778lSTY2NmrdurWGDh2qkiVLWrlCAGHF9xh/1qxZ6tWrl/Llyydvb2/OdACsxM/PTyNGjNCsWbNkNBrl4uKicePG6dNPP+WPwQCQwhE+AwCQBAQFSRcvhp/FfPKk5OMj/ddpIoJcuV6FzO+9Z/q3cGEpdeqErd0SXrx4ocWLF2v8+PG6ePGiJMnR0VFdunTRoEGDlCdPHusWCCBS8TnGDwkJUeHChXXhwgVNnTpVffv2tfhjAIiZQ4cOqXv37jp+/LgkqXr16po9e7aKFCli5coAANZC+AwAQCISevG/sCHzqVOm2c0vX0a+TaZMr8Ll0H+LFZPSp0/Y2uPDw4cPNWvWLE2ZMkV3796VJLm4uKhPnz7q06cPp/QCiVx8jvHXr1+vxo0bK3369Lpx44bSpElj8ccAEHNBQUGaMmWKRo4cqWfPnsnOzk5ffvmlhg8frtRJ8S/gAIA4IXwGAMBKHj40BczHj4cPmh8/jnx9Z2dTqBw2ZC5eXMqaVUpuZ7TevHlTkydP1pw5c/TkyRNJUq5cuTRw4EB17dqV8QKQRMTnGL9OnTraunWrhgwZIi8vL4vvH0DcXL16Vb1799aGDRskSQUKFNCsWbNUp04dK1cGAEhIhM8AAMSz4GDpwgXpxAnT7fhx07/XrkW+vq2tqT3G6y0z8uQx9WxOzs6dO6dx48Zp6dKlCgwMlCQVL15cQ4cOVZs2bWSXFBpTAzCLrzH+xYsXVbBgQRkMBl2+fFm5c+e26P4BWIbRaNSaNWvUr18/3bp1S5LUvn17TZgwQVmyZLFydQCAhED4DACABT14EDFkPnVKevEi8vVz5ZJKljQFzKEhc6FCkr19wtZtbQcOHJCnp6fWrVun0KFJtWrV5OHhIXd3dy5WBCRR8TXG//LLL+Xp6akPP/xQmzZtsui+AVje48ePNWLECE2fPl1Go1EZM2aUl5eXPvvsM6VK7n9ZB4AUjvAZAIBYCAqSvL3Dh8wnTkg3bkS+vpOTKVwuUcJ0Cw2cM2RI0LITFaPRqE2bNsnT01O7du0yL2/SpIk8PDxUqVIlK1YHwBLiY4wfEBCgnDlz6t9//9WaNWvUrFkzi+0bQPw6fPiwevTooX/++UeS9MEHH2jevHkqWLCglSsDAMQXwmcAAN7C1zd8wHzihHT6dNQXAMyb91XAHBo258sn2dgkbN2JVWBgoFatWiUvLy+dPHlSkmRnZ6cOHTpo8ODBKlKkiJUrBGAp8THG/+WXX/Txxx8rW7ZsunbtGu14gCQmKChIU6dO1ddff61nz54pderUGjVqlAYOHChbW1trlwcAsDDCZwAA/hMcLJ0/Lx07ZgqbQwPn27cjXz9NGtPs5bAh83vvSenSJWjZSYa/v7/mz5+viRMn6tp/Da/TpEmjzz//XP3795erq6uVKwRgafExxq9fv77+/PNPDR8+XN99953F9gsgYV2+fFk9evTQX3/9JUkqU6aMFixYoFKlSlm3MACARRE+AwBSJH9/6eRJU9D8zz+mf0+elJ4/j3z9AgXCt8woUSJlXADQEnx9fTVt2jRNnz5dDx48kCRlzZpVX3zxhXr27KkMKbn3CJDMWXqMf+XKFeXLl09Go1E+Pj7Kly+fRfYLwDqMRqOWLFmiAQMGyM/PTzY2Nho6dKhGjhyp1KlTW7s8AIAFED4DAJK9f/8NHzIfO2bq1xwSEnFdZ2dTuFyq1KuQuXhx0yxnxMyVK1c0YcIELViwQM//S/Xz58+vIUOGqFOnTnyoBFIAS4/xv/vuO40YMUK1a9fWli1bLLJPANZ3584d9evXTz///LMkqVChQpo/f76qVatm5coAAHEVq/B50aJF+v777zV06FB16dIl0nWGDRumNWvWhFv2+eefa8CAAZEXQvgMAIijkBDp0qWIQfOtW5Gvny2bVLq0KWguVcr0df78zGaOq+PHj8vLy0urVq1ScHCwJKls2bLy8PBQ8+bNZUPzayDFsOQY32g0qlixYjp79qwWLVqkTp06WaBCAInJ2rVr1atXL93+r+dZz5499cMPPygdPc0AIMmKUTf/oKAgLVq0SBMnTnzruidPnlS6dOn06aefmpeVK1cu5hUCABCJly9NF/0LGzIfPy49eRJxXYNBKlQofMhcsqQpfIZlGI1G7dy5U56entq8ebN5ed26deXh4aFatWrJYDBYsUIASd3x48d19uxZOTg4qFmzZtYuB0A8aNq0qWrUqKEhQ4Zo/vz5mjVrltavX6/Zs2erQYMG1i4PABALMQqfGzRooOvXrytnzpy6cuVKlOs9e/ZMPj4+Kl++vNq0aaPUqVPLyckprrUCAFKox49NIfPRo6+C5jNnpKCgiOumTm266F9oyFyqlOn/tM2IH8HBwVq3bp08PT116NAhSVKqVKnUsmVLDR06VGXKlLFyhQCSi+XLl0uSGjVqxCxIIBnLkCGD5s2bp7Zt26pbt266dOmSGjZsqE8++URTp05VpkyZrF0iACAGYhQ+V6hQQRMmTND27ds1ffr0KNc7ffq0QkJCdOzYMVWqVEkGg0H16tXT2LFjlYZP/wCAN/DzMwXNR46YwuYjR6QLFyJf18XlVcAc+q+bm2Qbo99uiI2XL19q6dKlGjdunLy9vSVJqVOn1qeffqpBgwYpf/78Vq4QQHISEhKiFStWSJI++eQTK1cDICHUqlVLJ0+e1DfffKOJEydq+fLl2rp1q2bPnq2mTZtauzwAQDTF6OP56NGjJUnbt29/43q+vr5ydXVVqVKlVLVqVf3555/6448/lCFDBvM+AAC4f/9VwBz676VLka+bK5dUpowpZA4NmnPkMLXUQMJ59OiRZs+ercmTJ+vOnTuSpIwZM6p3797q27evsmTJYuUKASRHe/bs0Y0bN5Q+fXp99NFH1i4HQAJxcnLSuHHj1LJlS3Xu3Flnz55Vs2bNmAUNAElIvMwN++ijj8INCmvXrq3y5ctr165d8fFwAIAk4N9/w4fMR49KV69Gvm6+fKaguWxZ079lykjvvJOw9SK827dva/LkyZo9e7YeP34sScqRI4cGDhyorl27Km3atFauEEBytmzZMklSixYtlDp1aitXAyChlS9fXkePHtX//vc/jRs3zjwLes6cOWrSpIm1ywMAvEG8hM+bNm3SgQMH1KZNGxUpUsT8IZW+zwCQMty+HT5oPnJEunkz8nULFnwVNJcta5rVnDFjwtaLqHl7e2vcuHFasmSJAgICJElFixbV0KFD1bZtW9nb21u5QgDJXXBwsNasWSNJatu2rZWrAWAtqVOn1g8//KBmzZqpc+fOOnfunJo2bap27dpp6tSpcnFxsXaJAIBIWCR8vnnzptatW6c8efLI3d1dRqNRK1eu1K5du9SwYUPt2LFDktSuXTtLPBwAIBG5dUv6+2/TLTRs/q8bQzgGg6kfc+hs5rJlTa0z0qdP8JIRDYcOHZKnp6d+/fVXGY1GSVKVKlXk4eGhBg0aKFWqVFauEEBKsW/fPvn6+ipjxoyqUaOGtcsBYGUVKlTQP//8o2+++Ubjx4/XsmXLzLOgGzdubO3yAACvsUj4fOPGDU2ZMkU1atSQu7u73N3d5e/vr4ULF2rRokXKkiWLhg0bRvgMAEncgwemkPnw4Ve3W7cirpcqlVSkyKvZzGXKmIJmrjmbuBmNRv3xxx/y9PQ0/+FYkho1aiQPDw9VqVLFesUBSLHWrVsnSWrYsKFsuaIsAJlmQXt6eqp58+bmWdBNmjRR+/btNWXKFGZBA0AiYjCGTmeyMoPBoKdPn8rZ2dnapQAAJD19aprJHDZojuxigKlSScWKSeXKvQqaS5aU6LSUdAQFBWn16tXy8vLS8ePHJUm2trZq166dhgwZomLFilm5QgBJVVzH+EajUQULFpSPj4/+7//+Ty1atLBwhQCSuhcvXphnQYeEhChbtmyaO3euGjVqZO3SAAAifAYASHr5Ujp+PHzQfPasFNlviIIFTUHz+++bbqVLS7x1J03Pnj3TwoULNWHCBF25ckWS5OzsrO7du2vAgAHKmTOndQsEkOTFdYx/+vRpFS9eXA4ODvL19VUaTqEBEIUDBw7o008/1blz5yRJHTp00JQpU5SRi4kAgFVx3hoApDBBQdKZM+GD5pMnpcDAiOvmyPEqZH7/fdPMZsbvSd/9+/c1Y8YMTZs2Tb6+vpKkzJkzq1+/furVqxenqgJINEJbbtSuXZvgGcAbVaxYUUePHtU333yjCRMmaOnSpdq6dat+/PFH1atXz9rlAUCKxcxnAEjGjEZTq4yDB6VDh15dFPD584jrvvPOq5A5dGZztmwJXzPiz7Vr1zRx4kTNmzdPz549kyTly5dPgwcPVufOneXo6GjlCgEkN3Ed41eoUEGHDh3SnDlz1L17dwtXByC52r9/vzp37ixvb29JUq9eveTl5UXeAABWQPgMAMnIw4emkPngwVe3/ya2hpM2rWkWc9hZzblzSwZDgpeMBHDq1Cl5eXlpxYoVCgoKkiSVLl1aHh4eatGiBRfwAhBv4jLGv3XrllxdXWUwGHTr1i1l4y+iAGLg2bNn8vDw0PTp0yVJBQsW1JIlS1SxYkUrVwYAKQufNgEgiQoKMrXLOHhQOnDA9O9/Le7Csbc39WUuX/5V0FyokOlCgUi+jEaj9uzZI09PT23YsMG8vHbt2vLw8FCdOnVk4K8NABKxjRs3SpLKly9P8AwgxpycnDRt2jQ1btxYn376qS5cuKAqVapo2LBhGjlypOzt7a1dIgCkCITPAJBE3LjxKmQ+eNDUQiOy9hn58kkVKkgVK5r+LVVKcnBI8HJhJSEhIVq/fr08PT21f/9+SaaZhx9//LGGDh2qcuXKWblCAIiev/76S5L00UcfWbkSAElZ3bp1dfLkSfXt21fLli3Td999p40bN2rp0qUqVqyYtcsDgGSPthsAkAj5+5vC5bCzmm/dirheunSmgDnsLXPmhK8X1hcQEKCffvpJ48aNM1/l3cHBQZ07d9bgwYNVoEABK1cIICWK7Rg/JCREWbJk0f3797V3715Vrlw5nioEkJL8/PPP+vzzz/XgwQM5ODjou+++U//+/WVjY2Pt0gAg2SJ8BgArMxolb29p375XQfPJk1JISPj1bGyk994LP6vZzY32GSnd48ePNXfuXE2aNEm3/vsLRfr06dWrVy/169ePU9UBWFVsx/hHjx5V2bJllTZtWt2/f192dnbxVCGAlOb27dvq2rWrubVP9erVtXjxYuXJk8e6hQFAMkXbDQBIYM+eSYcPm8Lmffuk/ful+/cjrufq+ipkrlDBdIFA/j6HUHfv3tWUKVM0c+ZMPXr0SJKUPXt2DRgwQN27d1e6dOmsXCEAxN6WLVskSTVr1iR4BmBR7777rn7//XfNnz9fAwYM0K5du1SiRAlNnjxZn376KdfEAAALI3wGgHh2/fqroHnfPunYMdPFAsNKnVoqV06qVOlV4OzqapVykchdvHhR48eP16JFi/Ty5UtJkpubm4YOHap27drJgQbfAJKB0H7PderUsXIlAJIjg8Ggbt26qVatWurUqZP27t2rLl26aN26dZo3b56yZMli7RIBINmg7QYAWFBAgClcDhs237wZcb3s2aUqVaTKlU23UqUkLriNNzly5Ig8PT31yy+/KOS/niwVK1aUh4eHGjdurFT0XwGQCMVmjP/8+XNlzJhRL1++1JkzZ1SkSJF4rBBAShccHKzx48fr66+/VmBgoLJkyaJFixZxsVMAsBDCZwCIg3v3TG0zQoPmw4elFy/Cr2NjYwqXQ4PmypWlnDklzujD2xiNRm3ZskWenp7aunWrebm7u7s8PDxUrVo1Tg0FkKjFZoy/detW1alTR66urrp+/TrvcwASxIkTJ/TJJ5/o9OnTkqQ+ffrIy8tLjo6OVq4MAJI22m4AQDQZjdL589Lu3dLevaaw+cKFiOu5uJjaZ4QGze+/T69mxExQUJB++eUXeXp66p9//pEk2djYqG3btho6dKjee+89K1cIAPEnbMsNgmcACaVEiRI6fPiwvvzyS02dOlXTp0/Xtm3btHz5cpUsWdLa5QFAkkX4DABRCAqS/vnHFDbv3i3t2SP5+kZcr2jRV0FzpUpSoUISHRAQG8+fP9ePP/6oCRMm6NKlS5IkJycndevWTQMGDFDu3LmtXCEAxL/Qiw3WrVvXypUASGkcHR01ZcoUffTRR+rcubPOnDmj8uXL6/vvv1f//v1pcwYAsUDbDQD4j7+/dPDgq7D5wAHTsrAcHEwXA6xWzdSzuWJFKWNG69SL5MPPz08zZszQ1KlTde/ePUlSpkyZ1K9fP/Xu3VuZMmWycoUAEDsxHeM/fvxYGTNmVEhIiG7evKns2bPHc4UAELl79+6pS5cuWr9+vSTT2RiLFy/mfQkAYojwGUCK5etrms28Z48pbD561DTbOawMGaSqVU23atWksmVNATRgCdevX9ekSZM0d+5c+f/3l448efJo0KBB+uyzz+Tk5GTlCgEgbmI6xt+yZYvq1q2rvHnzms8AAQBrMRqNmjt3rgYMGKDnz5/LxcVF8+fPV7NmzaxdGgAkGbTdAJAiGI3S1auv2mfs3i2dPRtxvRw5TCFztWqmwLlYMVpowPLOnDkjLy8vLVu2TEH//cWjRIkS8vDwUKtWrWRry69nACnTvn37JEmVK1e2ciUAYPoDWo8ePfTBBx+oXbt2Onr0qJo3b66uXbtq0qRJSpMmjbVLBIBEj0+3AJIlo9F0McAdO6SdO6Vdu6QbNyKuV7Toq6C5WjWJlrqIT3v37pWnp6f59E1JqlGjhjw8PFS/fn0urAUgxSN8BpAYFS5cWPv379fIkSPl5eWl+fPna8eOHVq+fLnef/99a5cHAIkabTcAJAtGo3T+vClo3rHDdLtzJ/w6tramthmhYXOVKtI771ijWqQkISEh2rBhgzw9PbV3715Jpt95zZo1k4eHh8qXL2/lCgEg/sRkjB8SEqKMGTPq8ePH+ueff1SqVKn4LxAAYmjHjh3q0KGDbty4IVtbW/3vf//Tl19+KRsbG2uXBgCJEuEzgCTJaJTOnXsVNO/cKd29G34dBwfTBQFr1JCqVzddKJC3GCSUgIAArVixQl5eXjpz5owkyd7eXh07dtTgwYPl5uZm5QoBIP7FZIx/6tQpvffee0qTJo38/PxoQQQg0fLz81OPHj30888/S5KqVaumn376Sbly5bJyZQCQ+DCiA5AkGI3SmTOvguadO6V//w2/joODVLmy9MEHpsC5QgUpdWprVIuU7OnTp5o3b54mTpyoG//1ekmXLp0+//xz9e/fX++++66VKwSAxCm05UaFChUIngEkahkzZtSqVavUoEED9enTR7t371bJkiW1YMECNW/e3NrlAUCiwqgOQKIUEvIqbN6xw9Sz+d698OukTm0Km2vUMAXO5csTNsN6/v33X02dOlUzZ86Un5+fJClbtmwaMGCAevToofTp01u5QgBI3ELD50qVKlm5EgB4O4PBoE6dOqlq1ar65JNPdOjQIbVo0UKff/65Jk6cKEdHR2uXCACJAm03ACQKRqPk4yNt3Wq6bd8u+fqGX8fR0dSnOXRm8/vvm2Y7A9Z06dIljR8/Xj/++KNevHghSSpUqJCGDBmiDh06yIEXKYAULCZj/EKFCunChQvauHGjPvroowSoDgAsIyAgQF9//bW8vLwkScWLF9fKlStVrFgxK1cGANZH+AzAam7dkrZtM4XN27ZJ166Fv9/JyRQ2h85sfv99yd7eKqUCEfzzzz/y9PTUzz//rJCQEElS+fLl5eHhoSZNmnDRGQBQ9Mf49+7dU5YsWSRJDx48UMaMGROiPACwqD///FMdO3bU3bt35ejoqMmTJ6tbt24yGAzWLg0ArIbwGUCCefDA1EIjNHA+dy78/XZ2UqVKUu3aUq1apjYahM1ITIxGo7Zt2yZPT0/99ddf5uUffvihPDw89MEHH/DhAgDCiO4Y/7ffflOTJk1UtGhRnT59OoGqAwDLu3v3rjp16qQ//vhDkvTxxx9r7ty5/FENQIpFz2cA8cbfX9qz59XM5qNHTe01QhkMUpkyprC5dm3TLGf+/oTEKDg4WGvWrJGnp6eOHDkiSbKxsVHr1q01dOhQlSxZ0soVAkDSduDAAUn0ewaQ9GXNmlUbN27UxIkTNWzYMP3f//2fDh06pBUrVqhy5crWLg8AEhzhMwCLCQqSDh2S/vrLFDgfOCAFBoZfp3DhV2FzjRoSEwCQmL148UKLFy/W+PHjdfHiRUmSo6OjunTpokGDBilPnjzWLRAAkomjR49KksqVK2flSgAg7lKlSqXBgwfrgw8+UNu2beXj46Pq1atr1KhR+vLLL2nPBiBFoe0GgDjx8ZH+/NN027ZNevw4/P05c74Km2vVkrJnt06dQEw8fPhQs2bN0pQpU3T37l1JkouLi/r06aM+ffooc+bMVq4QAJKG6IzxjUajsmTJIl9fXx08eFDly5dPwAoBIH49fvxYPXv21PLlyyVJNWvW1E8//aTsfDACkEIQPgOIkYcPpe3bXwXOly6Fv9/F5VXYXLu2lD+/qb0GkBTcvHlTkydP1pw5c/TkyRNJUs6cOTVo0CB16dJFadKksXKFAJC0RGeMf/36deXKlUs2NjZ6+vSpUqdOnYAVAkD8MxqNWrJkiXr37i1/f39lypRJixYtUsOGDa1dGgDEO9puAHijoCDp8OFXYfPBg1Jw8Kv7bW2lypWlevVMtzJlJM4iQ1Jz7tw5jRs3TkuXLlXgf71iihcvrqFDh6pNmzays7OzcoUAkHyF9tIvVqwYwTOAZMlgMKhTp06qWLGi2rZtq3/++UeNGjXSF198IU9PTzk4OFi7RACIN4TPACK4dMkUNIf2bn70KPz9hQq9Cptr1JDSprVKmUCcHThwQJ6enlq3bp1CTwSqVq2aPDw85O7uLgPT9gEg3oX2ey5btqyVKwGA+OXm5qb9+/fLw8NDU6ZM0ZQpU7Rnzx6tXr1a+fLls3Z5ABAvCJ8B6NkzaccOadMm083HJ/z9GTNKdeqYwua6daXcua1SJmARRqNRmzZtkqenp3bt2mVe3qRJE3l4eKhSpUpWrA4AUp7Q8Ll06dJWrgQA4p+Dg4MmT56sOnXqqFOnTjpy5IhKly6tBQsW6OOPP7Z2eQBgcfR8BlKoixdNQfPGjabg+cWLV/fZ2kqVKr2a3Vy2LK00kPQFBgZq1apV8vLy0smTJyVJdnZ26tChgwYPHqwiRYpYuUIASH6iM8bPmTOnbty4od27d6tq1aoJWB0AWNf169fVpk0b7du3T5LUp08fjR8/njYcAJIVwmcghXj+XNq581XgfPFi+Ptz5pQ++sh0q1VLSpfOOnUClubv76/58+dr4sSJunbtmiQpTZo0+vzzz9W/f3+5urpauUIASL7eNsb38/OTi4uL+esMGTIkYHUAYH2BgYH6+uuv5enpKUkqU6aMVq1apQIFCli5MgCwDMJnIBm7dMkUNG/aJG3fbgqgQ9naStWqvQqcixWTaG+L5MTX11fTpk3T9OnT9eDBA0lS1qxZ9cUXX6hnz54EHACQAN42xt+9e7eqV6+uXLly6erVqwlcHQAkHhs3blTHjh11//59pU2bVvPnz1erVq2sXRYAxBk9n4Fk5MULadeuV4Gzt3f4+11dJXd3U9hcuzazm5E8XblyRRMmTNCCBQv0/L+/uOTPn19DhgxRp06dlDp1aitXCAAIdeLECUnSe++9Z+VKAMC63N3ddezYMbVt21Z79uxR69attWPHDk2cOJHxK4AkLZW1CwAQN//+K/34o9S8ufTOO1L9+tKUKabg2dZW+uADydNTOnFCun5dmjtXataM4BnJz/Hjx9WuXTsVKFBA06dP1/Pnz1W2bFmtXr1a58+fV48ePRi4A0AU7t27p379+ql8+fKqWbOmpk6dquDg4Dduc+jQIRUpUkQdOnSI9eOG9uAvUaJErPcBAMlFjhw5tH37dg0bNkySNGvWLFWqVEkXLlywcmUAEHvMfAaSGKNROn1a+u03af166eBB07JQ2bO/aqVRp46UPr31agXim9Fo1M6dO+Xp6anNmzebl9etW1ceHh6qVauWDPSTAYC3+uKLL3TkyBG1aNFCV69e1YwZM5Q6dWp179490vUfPnyoIUOGKCQkJE6PGxo+M/MZAExsbW01duxYVa9eXR06dNCxY8dUtmxZzZs3T61bt7Z2eQAQY4TPQBIQEGC6WOD69abblSvh7y9TRmrUyHQrU4bezUj+goODtW7dOnl6eurQoUOSpFSpUqlly5YaOnSoypQpY+UKASDpuHLlio4cOaLSpUtr7NixevDggSpVqqRffvklyvD5q6++kp+fX5we12g0MvMZAKLw4Ycfmttw7N69W23atNGOHTs0adIkzuYDkKTQdgNIpO7fl5YulVq1MrXTqFdPmjbNFDw7OJh6N8+eLd24IR05Iv3vf1LZsgTPSN5evnyp+fPnq2jRomrRooUOHTqk1KlTq2fPnvL29tbKlSsJngEghkJP586bN68kycXFRS4uLrp69aoCAwMjrL9s2TJt2bJFgwcPjtPj3rx5U0+ePJGNjY0KFiwYp30BQHLk6uqqbdu26auvvpLBYNDs2bNVsWJFeb9+cR8ASMSY+QwkIhcvSr/+amqpsW+fFPZM1qxZpYYNTbOb69SRorhoPJAsPXr0SLNnz9bkyZN1584dSVLGjBnVu3dv9e3bV1myZLFyhQCQdPn7+0uSHBwczMscHBxkNBr1/Plz2dnZmZd7e3vL09NTbdq0Ua1atfTdd9/F+nHPnj0rSSpQoIDs7e1jvR8ASM5sbW317bffqnr16mrfvr2OHz+usmXLasGCBWrVqpW1ywOAtyJ8BqzIaJSOH5fWrDGFzqdOhb+/RAlT2Ny4sVSunJSKcxWQwty+fVuTJ0/W7Nmz9fjxY0mmC7EMHDhQXbt2Vdq0aa1cIQAkfU5OTpKkgIAA87IXL17IYDDI0dHRvCwgIEADBw5UunTp1LZtW926dcu87tWrV5U7d+4YPe65c+ckSUWKFInrIQBAslevXj0dO3ZMn3zyiXbu3KnWrVtr7969GjduHH/AA5CoET4DCSw42DSr+ddfTbew/ZttbKQaNaSmTU2hcww/wwHJhre3t8aNG6clS5aYw5CiRYtq6NChatu2LQNsALCg/PnzSzL1fpZMFxP08/NTnjx5ws16/vfff80tOpo0aWJefuLECdWrV0/nz5+P0eOGhs+FCxeOS/kAkGJkz55dW7Zs0ciRI/X9999r6tSpOnjwoFavXq1cuXJZuzwAiBThM5AAXr6Utm0zhc3r1kn//vvqPkdHqX59qVkzU1sNFxfr1QlY26FDh+Tp6alff/1VRqNRklSlShV5eHioQYMGSsX0fwCwuPz586t48eI6cuSIhg8frmvXrkmSmjdvrps3b2rdunXKkyePatasqSlTppi3e/DggUaNGqUCBQqob9++MX5cwmcAiDlbW1uNHTtWlStXVocOHXTw4EGVKVNGP/30kz788ENrlwcAERiMoZ/urcxgMOjp06dyppEtkoknT6RNm0yB84YNpv+HypDBNLO5WTNT8Pzf2a5AimQ0GvXHH3/I09NTO3bsMC9v1KiRPDw8VKVKFesVBwAphK+vr0aPHq0DBw7I0dFRTZs2Vd++fXXkyBF17NhRNWrU0Jw5c8Jtc+PGDdWuXVvly5fX0qVLI93vm8b42bNn1+3bt3XgwAFVqFAhXo4LAJKzy5cvq2XLljpy5IgMBoO+/vprjRw5UjY2NtYuDQDMCJ8BC3ryRPr9d+nnn03B84sXr+57911TO41mzUytNcKcxQqkSEFBQVq9erW8vLx0/PhxSaaZHO3atdOQIUNUrFgxK1cIAIirqMb4jx49UoYMGSSZ2nykT5/eCtUBQNL34sULDRgwQLNnz5Yk1alTR8uXL1fmzJmtXBkAmBA+A3H0psC5QAGpeXNT4Fy+PBcMBCTp2bNnWrhwoSZMmGDuL+rs7Kzu3btrwIABypkzp3ULBABYTFRj/EOHDqlChQp69913zRcuBADE3k8//aQePXro2bNncnV11erVq1W5cmVrlwUA9HwGYuNNgXPBglKrVlLLllKJEpLBYL06gcTk/v37mjFjhqZNmyZfX19JUubMmdWvXz/16tVLLjQ8B4AUw9vbW5JUqFAhK1cCAMlD+/btVbp0abVo0ULnz5/XBx98IC8vL/Xv318GPpQCsCLCZyCaCJyB2Ll69aomTpyo+fPn69mzZ5KkfPnyafDgwercubMcHR2tXCEAIKH5+PhIMl3sEABgGcWKFdPhw4fVrVs3rVq1SgMHDtTevXu1cOFCpUuXztrlAUihCJ+BN/D3l9avl1avjjxwbtnSFDoTOAMRnTx5Ul5eXlqxYoWCg4MlSaVKldKXX36pFi1ayNaWX0EAkFKFhs8FChSwciUAkLykTZtWK1asUNWqVTVw4ED98ssvOnHihP7v//5PJUqUsHZ5AFIgPvkDrwkMlLZskZYtk9auNQXQoQoUeDXDuWRJAmfgdUajUXv27NEPP/ygjRs3mpfXqlVLHh4eqlu3Lqf9AQCY+QwA8chgMKhPnz56//331bJlS124cEEVKlTQrFmz1LlzZ2uXByCF4YKDgKSQEGn/fmn5ctMs5//a0UqS8uaV2rQxhc4EzkDkQkJCtH79enl6emr//v2STO/rH3/8sYYOHapy5cpZuUIAgDVENcbPli2b7t69q7///ltly5a1UnUAkPzdv39f7du31+bNmyVJXbp00bRp02h9ByDBED4jRTt1yhQ4L18uXb36anmWLFLr1tInn0gVKhA4A1EJCAjQTz/9pHHjxuncuXOSJAcHB3Xu3FmDBw/mdGoA/9/efUdFeebvH38PIKBig1hiQREN9ti7xt41ltijsXexSzTFjbvRgC323mMJ9oLGbogaS4iJ+sOIDexGVDCo9Pn9MV/ZddckouAzDNfrnJwZHsfh8lhyc3E/n1vSuRet8f/444+k2aMPHz4ke/bsBqUTEUkfEhMT+fLLL5kwYQJms5ny5cuzadMmChUqZHQ0EUkHVD5LuhMWBuvWWQrns2f/fd3FBdq2tRTO9euDxtGK/LlHjx6xaNEiZsyYwa1btwDIli0bgwYNwtvbmzx58hicUERErMGL1vi//vorZcuWxc3NjfD/vN1MRERS1f79++ncuTPh4eG4urqydu1aGjdubHQsEbFxqtckXXj0CDZsgJUr4Ycf/n3d0RGaNbMUzi1agO48Evlrd+/eZebMmcybN4/IyEgA8ubNy4gRI+jXr59O0RYRkb+lec8iIsZo0KABQUFBfPDBB5w6dYqmTZsyceJExo8fj52dndHxRMRGqXwWm5WQYDk4cNUq2LIFnj61XDeZoG5dS+Hcti3kyGFsTpG04NKlS0ydOpUVK1YQExMDgJeXF2PHjqVr1644OTkZnFBERNIKlc8iIsZxd3cnMDAQb29vFi9ezGeffcbJkydZtWqVxiCJSKpQ+Sw2JzjYssP5m2/g/6YBAFC8OHz0EXz4IeTLZ1w+kbQkKCgIX19fNm3aRGJiIgBVq1bFx8eHVq1aaYeEiIgk29WrVwEoXLiwwUlERNInZ2dnFi1aRJUqVRg8eDA7duygUqVKbN68mdKlSxsdT0RsjMpnsQn371vmOK9cCT/99O/rrq7QubOldK5YUQcHirwMs9nM/v378fX15cCBA0nXmzVrho+PD7Vq1cKkv0wiIvKKrl27BkDBggUNTiIikr717t2bd999l3bt2nHp0iWqVq3KkiVL6Ny5s9HRRMSGqHyWNCshAfbsgaVLYccOiIuzXHdwsMxx/ugjaN4cNA1A5OXEx8ezadMmfH19OX36NAD29vZ07tyZsWPHaheEiIikiGfls7u7u8FJRESkYsWKBAUF0blzZ/bv30+XLl04ceIEU6ZMIUOGDEbHExEbYDKbzWajQ8CLT8IWeZHQUFi2DJYvhxs3/n29fHlL4dy5M+TMaVg8kTTn6dOnLF++nGnTpnHlyhUAMmXKRN++fRkxYoR2pomIyCt70Ro/e/bsREZGEhwcTPHixQ1MJyIizyQkJPDZZ58xefJkAGrWrIm/vz9vv/22wclEJK1T+SxpQkwMbN8OS5bAvn3w7E+tqyt07w69eoE2ZYokz8OHD5k7dy6zZs3i3r17ALi5ueHt7c3gwYNxc3MzOKGIiKR1/73Gj4yMTDrQ6o8//sDFxcXAdCIi8t+2bt1K9+7d+eOPP3j77bfZsGEDNWrUMDqWiKRhKp/FqgUHW8ZqrFoF4eH/vt6gAfTpA61ba6yGSHJdv36dGTNmsGjRIh4/fgxAoUKFGDVqFL169SJTpkwGJxQREVvx32v8c+fOUbp0aVxdXbl//77B6URE5EVCQkJo06YNwcHBODg4MH36dIYMGaJzX0TklWjms1idJ0/A3x8WL4Zjx/59PW9e6NnTsstZh6OLJF9wcDB+fn6sWbOG+Ph4AMqUKYOPjw8dOnTAwUH/SxARkdSlec8iItbvnXfe4cSJE/Tu3Rt/f3+8vb05ceIEixYt0kYVEUk2NQ1iNS5cgPnzYcUKiIy0XLO3hxYtLLucmzSxHCYoIslz9OhRfH192bFjR9K1OnXq4OPjQ+PGjbWDQURE3hiVzyIiaYOLiwvr16+natWqjBkzhjVr1nD27Fk2bdpEkSJFjI4nImmIqjwxVFycZZbzvHlw8OC/rxcubCmce/QAnW8gknyJiYkEBATg6+vL0aNHAcutz23atMHHx4fKlSsbnFBERNIjlc8iImmHyWRixIgRlC9fng4dOnDmzBkqVarEunXraNKkidHxRCSNsDM6gKRPt27BF19AoULwwQeW4tnODlq2hN274eJFGDdOxbNIcsXGxrJy5UpKly5Nq1atOHr0KI6OjvTp04fz58+zadMmFc8iImIYlc8iImnPe++9x88//0zVqlWJiIigWbNmfPXVV1jJEWIiYuVeqXxesWIFXl5eLF269E9fExgYSKtWrShXrhxt27blxIkTrxxSbIPZbCmZP/gA3N3hH/+wlNC5csH48XDlimUXdJMmliJaRF5eVFQUM2bMwNPTkx49ehAcHEzWrFkZO3YsoaGhLF68GC8vL6NjiohIOqfyWUQkbcqXLx+HDx+mT58+mM1mxo0bR8eOHYmKijI6mohYuWRVfPHx8SxZsgQ/P7+/fN3ly5cZNGgQDx8+pH379ty+fZu+ffty48aN1woradOTJ7BoEZQqBfXrw6ZNkJAAtWrBunVw/Tp8+SUULGh0UpG05/fff+fTTz/F3d2dkSNHcuPGDfLkyYOvry/Xrl3D19eXt3ULgYiIWIlnXw/ky5fP4CQiIpJcTk5OLF68mAULFpAhQwY2bNhA9erVuXz5stHRRMSKJWvmc/Pmzbl+/ToFChQgNDT0T1+3fft24uLiGDp0KB06dEgqQgICAujfv//rZpY04sYNmDvXUjw/eGC55uIC3brBwIFQurSx+UTSsitXrjB16lSWL19OdHQ0YDmVesyYMXTr1g0nJyeDE4qIiDzPbDZz584dAH1jVEQkDevfvz+lSpWiXbt2nD17NmkOdOPGjY2OJiJWKFk7n6tUqYK/vz8tWrT4y9eFhIQA4OHhAYCnpycAFy9efJWMksacOAGdO1vmOX/1laV49vCAGTMshfS8eSqeRV7V6dOn6dSpE0WLFmX+/PlER0dTuXJlNm3aRHBwMH369FHxLCIiVunRo0c8ffoUUPksIpLW1ahRg6CgIKpUqcLDhw9p1qwZvr6+mgMtIv8jWeXzxIkTKVWq1N++7smTJwBJBYijo+Nz18X2xMXB+vVQrRpUrWp5npAAderAli2WAwSHD4ds2YxOKpL2mM1mDhw4QKNGjShfvjzffvstiYmJNGnShEOHDnH8+HHatm2Lvb290VFFRET+1O3btwHImjUrmTJlMjiNiIi8rnz58vH999/Tu3dvEhMT+fjjj+nUqROPHz82OpqIWJFUOdYtY8aMAMTGxgIQExMDoEWmDYqMBF9fy87mzp3h+HFwdIQePeD0aTh0CFq3BnViIsmXkJDAhg0bqFSpEg0aNGDfvn3Y29vTpUsXfvnlF3bv3k2dOnUwmUxGRxUREflbz8rnPHnyGJxERERSyrM50PPmzcPBwQF/f3+qVavGlStXjI4mIlYiVcrnZ2M2rl69CpA0H7po0aKp8enEADdvwtixUKAAfPyx5ePcueEf/4Br12D5cihb1uiUImlTdHQ0CxcupFixYnTo0IGgoCAyZszIkCFDuHjxImvWrOHdd981OqaIiEiyaN6ziIhtMplMDBw4kEOHDpE7d27Onj1LxYoV2bt3r9HRRMQKJOvAwT9z8+ZNtm3bRqFChWjWrBktW7Zk+fLlzJo1i0uXLrFjxw4cHR1p1qxZSnw6MdD58zB1KqxebRm1AVCyJIwebdn5rFGzIq8uIiKC+fPnM3PmTO7evQuAq6srQ4YMYciQIeTMmdPghCIiIq/u2c5nlc8iIrapZs2a/PTTT7Rr146TJ0/StGlTJk+ezJgxY3S3pkg6liI7n2/cuMHMmTPZtm0bAMWKFWPhwoW4ubnh7+9P7ty5WbRoEQUKFEiJTycGOHbMMj6jRAlYtsxSPNeuDTt3wpkzljEbKp5FXs3NmzcZM2YM7u7ujB8/nrt371KgQAG+/vprwsLC+OKLL1Q8i4hImqfyWUTE9uXPn5/vv/+eXr16kZiYiI+PD507d9YcaJF0zGS2kqNITSYTUVFRZM6c2ego8n/MZti1CyZPhqNHLddMJksJPXas5WBBEXl1v/32G1OmTGH16tXE/d+tBKVKlWLs2LF06tSJDBkyGJxQRETk9fznGr9bt2588803+Pr6MnbsWKOjiYhIKjKbzcyfP59hw4YRHx9PmTJl2Lp1Kx4eHkZHE5E3LFVmPkvalpgIW7ZAhQrQooWleHZ0hD59LGM3Nm9W8SzyOo4fP06bNm0oUaIEy5YtIy4ujlq1arFz507OnDlDt27dVDyLiIjN0c5nEZH0w2QyMWjQIA4ePEiuXLk4c+YMFStWZN++fUZHE5E3TOWzJElIAH9/ePddaNsWTp+GzJlhzBgIDYXFi8HLy+iUImmT2Wxm165dvPfee1SrVo2tW7diNpt5//33OXbsGIGBgTRv3lyz0ERExGapfBYRSX9q1arFTz/9RKVKlXjw4AFNmjRh+vTpWMlN+CLyBqh8FuLjYc0aKFUKOnaEc+cga1b45BNL6eznB/oaQeTVxMXF8c033/Duu+/SvHlzAgMDyZAhA7169SI4OJitW7dSrVo1o2OKiIikOpXPIiLpU4ECBQgMDKRHjx4kJiYyatQoevbsSXR0tNHRROQN0MzndCwuzlI6f/klXLpkuZY9O4wYAd7eluci8moeP37MkiVLmD59OteuXQPAxcWFAQMGMHz4cPLly2dwQhERkdT3bI3v4OCAs7MzAOHh4bi5uRmcTERE3jSz2cysWbMYOXIkiYmJVKlShS1btuibkiI2TuVzOpSQAOvWwT/+AZcvW665ucGoUTB4sGXXs4i8mvDwcGbPns2cOXN48OABALlz52bYsGEMHDiQ7PqujoiIpCPP1vgPHjzA3d0dBwcHYmNjNWZKRCQd27dvHx07duThw4fkzZuXrVu3UqlSJaNjiUgqUfmcjpjNloMEP/sMgoMt13Llssx0HjAAXFyMzSeSloWGhjJt2jSWLl3K06dPAfD09GTMmDF89NFHSbu9RERE0pNna/yLFy9Srlw5cufOzZ07d4yOJSIiBrt06RKtWrXi/PnzODk5sXTpUrp27Wp0LBFJBZr5nA6YzbB7N1SsCO3aWYrn7Nlh0iS4cgVGj1bxLPKqfv31V7p27UqRIkWYM2cOT58+pUKFCvj7+3PhwgX69++v4llERNK9+/fvA2jchoiIAFCkSBGOHz9OixYtiImJ4cMPP8THx4eEhASjo4lIClP5bOMCA6F2bWjWDH7+2VIyf/opXL0K48aBNpqLJJ/ZbObw4cM0bdqUsmXLsnbtWhISEmjYsCH79+/n1KlTtG/fHnt7e6OjioiIWAWVzyIi8t+yZs3K1q1bGTduHAB+fn60atWKyMhIg5OJSEpS+WyjgoOhRQt47z04cgScnGDkSMtO53/+U4cJiryKhIQENm/eTNWqValbty7fffcddnZ2dOzYkaCgIPbu3Uv9+vU1x1JEROS/qHwWEZEXsbe3Z9KkSaxduxZnZ2d27dpFlSpVCAkJMTqaiKQQlc825vZt6NcPSpeGgACwt4f+/S0HC06bBjlzGp1QJO2JiYlhyZIllChRgnbt2nHy5EmcnZ0ZOHAgISEhrF+/nvLlyxsdU0RExGqpfBYRkb/SuXNnjhw5Qv78+blw4QKVK1dmz549RscSkRSg8tlGREXBF19A0aKweDEkJkLr1vD//h8sWAD58hmdUCTtiYyMxNfXl0KFCtG3b19CQkLIkSMHn376KWFhYcybNw9PT0+jY4qIiFi98PBwAN566y2Dk4iIiLWqUKECp06donr16kRGRtKsWTOmT5+O2Ww2OpqIvAaVz2lcfLylbC5aFP7xD3j8GKpUscx63rIFvLyMTiiS9ty+fRsfHx/c3d35+OOPuXPnDvnz52f69OmEhYXxz3/+k1y5chkdU0REJM3QzmcREXkZefLk4eDBg/Tu3ZvExERGjRpFjx49iI6ONjqaiLwilc9p2KFDUK6cZczGnTtQuDB8+y38+CPUqmV0OpG0JyQkhL59+1KoUCH8/Px49OgRJUqUYMWKFVy+fJkRI0aQJUsWo2OKiIikOSqfRUTkZTk5ObF48WJmzZqFvb09q1at4r333uPWrVtGRxORV6DyOQ26dg06dIB69eDcOciRA2bMsBwy2KED6KwzkeQ5efIk7dq1o1ixYixZsoTY2Fhq1KjB9u3bOXv2LB999BGOjo5GxxQREUmzVD6LiEhymEwmhg4dyp49e8iRIwcnT56kYsWKnDx50uhoIpJMKp/TkOho+Oc/oVgx2LAB7Oxg8GC4dAmGDwcnJ6MTiqQdZrOZ7777jrp161KlShU2b96M2WymZcuWHDlyhCNHjtCyZUvs7PTPpIiIyOtS+SwiIq+ifv36nDp1ihIlSnD79m1q167NN998Y3QsEUkGtSppgNkMW7dCiRLw+efw9CnUrg0//wxz5oCrq9EJRdKO+Ph41q5dS7ly5WjatCmHDx/GwcGBjz76iHPnzrF9+3Zq1KhhdEwRERGbovJZRERelaenJz/++CMtW7YkJiaGbt26MX78eBITE42OJiIvwWS2kmNDTSYTUVFRZM6c2egoViUkBIYOhb17LR/nywdTp0LHjhqvIZIcT548YdmyZUybNo3Q0FAAMmfOTL9+/RgxYgQFChQwNqCIiIgNMplMREREkD17dgDu3r2rQ3tFROSVJCYm8umnnzJ58mQAWrduzerVq3FxcTE4mYj8FZXPViomBnx94csvITYWHB1h9GgYNw7076rIy7t//z5z585l9uzZhIeHA5AzZ068vb0ZNGgQrrp1QEREJNWYTCauXr2Kh4cHAHFxcTg4OBicSkRE0rLVq1fTp08fYmNjeffdd9m+fTvu7u5GxxKRP6Hy2Qr98AP06we//Wb5uHFjy3iNIkWMzSWSloSFhTF9+nSWLFnCkydPAPDw8GD06NH07NmTjBkzGpxQRETE9plMJoKCgqhQoQLZs2fn4cOHRkcSEREb8OOPP9K6dWt+//13cufOzZYtW6hWrZrRsUTkBTTz2Yo8eAB9+ljmOf/2G+TKBevWwe7dKp5FXtbZs2fp1q0bnp6ezJo1iydPnlC2bFnWrVtHSEgIgwYNUvEsIiLyBj148ADQvGcREUk51apV4+TJk5QpU4a7d+9St25d1qxZY3QsEXkBlc9WwGyG9euheHFYutRyrW9fSwHdqZNmO4v8HbPZTGBgIM2bN6dMmTJ88803JCQkUK9ePfbs2cPPP/9Mp06ddJuviIiIAVQ+i4hIaihYsCBHjx7l/fffJyYmhg8//JBPPvlEBxGKWBmVzwa7exfatYPOneH33y0F9A8/wKJFkCOH0elErFtiYiJbt26levXqvPfee+zatQuTyUT79u05deoUBw4coFGjRpj0HRwRERHDREREAJBDi1sREUlhLi4ubN68mY8//hiASZMm8cEHHxAVFWVwMhF5RuWzgb79FkqWhC1bwMEB/vEPOH0aatY0OpmIdYuJiWHZsmWULFmSNm3acPz4cZycnOjfvz8hISH4+/tTsWJFo2OKiIgI8McffwCQNWtWg5OIiIgtsrOzY/LkyaxatQpHR0e2bNlCrVq1uH79utHRRATQPegGuHcPBg+GDRssH7/7LqxcaXkUkT/36NEjFi1axIwZM7h16xYA2bJlY9CgQXh7e5MnTx6DE4qIiMh/e7b7LEuWLAYnERERW/bs7J82bdrwyy+/UKlSJbZu3UrVqlWNjiaSrmnn8xu2aZNlt/OGDZbdzhMmwMmTKp5F/srdu3cZP3487u7ujBkzhlu3bpE3b16mTJnCtWvXmDRpkopnERERK/Vs57PKZxERSW3Vq1d/7iDCOnXqsHbtWqNjiaRrKp/fkD/+gI8+gg8+sOx8Ll0aTpywjNpwdDQ6nYh1unTpEgMGDKBgwYJMnjyZyMhIvLy8WLp0KVeuXGH06NG6hVdERMTKaeeziIi8Sc8OImzVqhUxMTF07dpVBxGKGEjl8xtw8iSUKwerVoGdHYwfD6dOQfnyRicTsU5BQUF06NABLy8vFi5cSExMDFWrVmXLli0EBwfTq1cvnJycjI4pIiIiL0E7n0VE5E1zcXFhy5Yt+Pj4AP8+iPDx48cGJxNJf1Q+p6KEBPjyS6heHS5fBnd3+P57yzX1ZiLPM5vN7Nu3jwYNGlCxYkU2bNhAYmIizZo14/vvv+fYsWO0bt0aOzv9syUiIpKWaOeziIgYwc7Ojq+++oqVK1cmHURYs2ZNHUQo8oapxUklN29CvXrw6aeWErpjR/j1V6hZ0+hkItYlPj6eb7/9lgoVKtCoUSMOHDiAvb09H374IWfOnCEgIIDatWtjMpmMjioiIiKvQOWziIgYqXv37hw6dIicOXMmHUR44sQJo2OJpBsqn1PBvn1QtiwEBoKLC6xYAevWQfbsBgcTsSJPnz5l3rx5eHl50alTJ06fPk2mTJkYNmwYly9fZvXq1ZQuXdromCIiIvKaNHZDRESMVr16dU6dOkXp0qW5e/cu7733HuvWrTM6lki6oPI5BSUkwBdfQOPGEB5uKaBPn7YcNKhNmyIWDx8+5F//+hcFCxZk8ODBXLlyBTc3N7744guuXbvG119/TcGCBY2OKSIiIilEO59FRMQa/PdBhF26dOGLL77AbDYbHU3Epql8TiH37kHTpvCPf4DZDH37wrFjUKSI0clErMP169cZOXIkBQoU4LPPPuPevXsUKlSI2bNnc+3aNT7//HPc3NyMjikiIiIpTDufRUTEWmTJkoXNmzczZswYAP7xj3/QrVs3oqOjDU4mYrtMZiv5Fo/JZCIqKorMmTMbHSXZTpyAdu0sc54zZoQFC6B7d6NTiViH4OBg/Pz8WLNmDfHx8QCUKVMGHx8fOnTogIODg8EJRUREJLWYTCayZ89OREQE58+fp1ixYkZHEhERAWDx4sUMGjSI+Ph4qlevztatW8mZM6fRsURsjsrn17RqFfTrBzEx4OUFGzdCqVJGpxIx3tGjR/H19WXHjh1J1+rUqYOPjw+NGzfWAYIiIiLpgMlkwt7enoSEBG7cuEG+fPmMjiQiIpLkwIEDtGvXjsjISDw8PAgICKB48eJGxxKxKRq78YoSEmD0aMs855gYeP99OHVKxbOkb4mJiezYsYOaNWtSs2ZNduzYgclkom3btpw4cYJDhw7RpEkTFc8iIiLpSEJCAqCxGyIiYn3q16/Pjz/+SOHChbl69SrVqlVj//79RscSsSna+fwKIiKgc2f47jvLx599Zpn1bKcqX9Kp2NhY1q1bh5+fH8HBwQA4OjrSvXt3Ro8ejZeXl8EJRURExAj/+Q3nhIQE7LRgFhERK3Tv3j3atGnD0aNHsbe3Z968efTr18/oWCI2QeVzMl28CC1aQEiIZb7zihXQoYPRqUSMERUVxeLFi5k+fTo3btwAIGvWrAwYMIDhw4fz9ttvG5xQREREjPSsfM6cOTNRUVEGpxEREflzMTEx9O7dmzVr1gAwatQofH19sbe3NziZSNqm8jkZfvwRWraE+/ehQAHYtg3KlTM6lcib9/vvvzNr1izmzZvHw4cPAciTJw8jRoygf//+ZMuWzeCEIiIiYg2elc958uTh9u3bBqcRERH5a2azmX/+859MmDABgPfff581a9ZYdVclYu1UPr+kbdugUyeIjoZKlWDHDsid2+hUIm/WlStXmDp1KsuXLyc6OhqAd955hzFjxtCtWzecnJwMTigiIiLW5Fn5XLRoUUJCQgxOIyIi8nLWrVtHz549iYmJoVy5cuzYsUOH5oq8Ig1dewnz50PbtpbiuXlzOHRIxbOkL6dPn6ZTp04ULVqU+fPnEx0dTeXKldm0aRPBwcH06dNHxbOIiIj8KR02KCIiaUnnzp05ePAgOXPm5PTp01SuXJmff/7Z6FgiaZLK579gNsO4cTBoECQmQt++sHUrWOnmbJEUZTabOXDgAI0aNaJ8+fJ8++23JCYm0qRJEw4ePMjx48dp27at5l+JiIjI31L5LCIiaU316tU5ceIEJUqU4NatW9SqVYtt27YZHUskzVH5/CcSEqBXL/jqK8vHEyfCwoXg4GBsLpHUlpCQwIYNG6hUqRINGjRg37592Nvb06VLF3755Rd2795N3bp1nzu9XkREROSvuLi4GB1BREQk2Tw8PDh27BgNGzbkyZMntGnThmnTpmElE2xF0gSVzy8QFwddu8KKFWBvD0uXwmefgbo2sWXR0dEsXLiQYsWK0aFDB4KCgsiYMSNDhgzh4sWLrFmzhnfffdfomCIiIpIGaeeziIikVdmyZSMgIIABAwZgNpsZPXo0AwYMIC4uzuhoImmC9vH+l+ho6NgRtm+HDBlg/XrLvGcRWxUREcH8+fOZOXMmd+/eBcDV1ZUhQ4YwZMgQcubMaXBCERERSetUPouISFqWIUMG5s2bh5eXFyNHjmTRokVcuXKFDRs2kD17dqPjiVg1lc//4ckTaN0a9u0DZ2fYtAmaNTM6lUjquHnzJl9//TULFy7kjz/+AKBAgQKMGjWK3r176/ZYERERSTEqn0VEJK0zmUwMHz4cT09POnfuzP79+6levTo7d+6kcOHCRscTsVoau/F/nj6FVq0sxXPmzBAQoOJZbNP58+fp1asXHh4eTJ06lT/++INSpUqxatUqLl++zLBhw1Q8i4iISIpS+SwiIraiZcuWHDlyhHz58nH+/HmqVKnCsWPHjI4lYrVUPgMxMfDBB3DggKV43rMH6tUzOpVIyvrxxx9p3bo1JUqUYPny5cTFxVGrVi127tzJmTNn6NatGxkyZDA6poiIiNigzJkzGx1BREQkxZQtW5aTJ09SoUIFwsPDqVevHv7+/kbHErFK6b58jouDTp1g1y7ImNGy47lGDaNTiaQMs9lMQEAAtWvXpnr16mzbtg2A999/n2PHjhEYGEjz5s0x6TRNERERSUUODpr2JyIitiVv3rx8//33vP/++8TExNCxY0d8fX0xm81GRxOxKum6fE5IgG7dYOtWcHKyHDL43ntGpxJ5fXFxcaxevZoyZcrQokULfvjhBzJkyECvXr0IDg5m69atVKtWzeiYIiIikk7Y29sbHUFERCTFZc6cmU2bNjF8+HAAPv74Y/r3709cXJyxwUSsSLrdgmA2w9Ch8O23kCGD5XDBBg2MTiXyeh4/fsySJUuYPn06165dA8DFxYUBAwYwfPhw8uXLZ3BCERERSY9UPouIiK2yt7dnxowZFC5cmOHDh7N48WKuXbuGv78/WbNmNTqeiOHSbfk8eTLMnw8mE6xdC82bG51I5NWFh4cze/Zs5syZw4MHDwDInTs3w4YNY+DAgWTPnt3YgCIiIpKuqXwWERFbN3ToUAoWLEjnzp3Zs2cPNWvWJCAggAIFChgdTcRQ6XLsxooV8MknluezZlkOGxRJi0JDQxk6dCju7u5MnDiRBw8e4OnpyYIFCwgNDWXcuHEqnkVERMRwmvksIiLpQatWrfj+++/JnTs3Z8+epWrVqpw+fdroWCKGSnfl83ffQZ8+luc+PjBkiLF5RF7Fr7/+SteuXSlSpAhz5szh6dOnVKhQAX9/fy5cuED//v1xdnY2OqaIiIikEffu3cPb25vKlStTt25dZs2aRUJCwgtfu2LFCho3bkyZMmVo2LAhS5cu/dv3185nERFJLypWrMiJEycoWbIkt27dolatWuzatcvoWCKGSVfl89mz0L695aDBDz+ESZOMTiTy8sxmM4cPH6Zp06aULVuWtWvXkpCQQMOGDdm/fz+nTp2iffv2+uJOREREkm3YsGHs2bOHBg0akDdvXubOnfvCUnnjxo1MnjwZZ2dnevbsiclkws/PD39//798f61PREQkPSlYsCBHjhyhfv36PH78mJYtWzJv3jyjY4kYIt2Uzw8eQOvWEBUFdevC0qVgl25+9ZKWJSQksHnzZqpWrUrdunX57rvvsLOzo2PHjgQFBbF3717q16+PyWQyOqqIiIikQaGhoQQFBVGuXDkmTZrE7NmzAdi0adP/vDYsLAxPT098fX0ZMWIEPj4+AJw4ceIvP4fKZxERSW+yZ8/Orl276NmzJ4mJiQwePJjRo0eTmJhodDSRNypd1K/x8dCxI1y5Ah4esGEDODoanUrkr8XExLBkyRJKlChBu3btOHnyJM7OzgwcOJCQkBDWr19P+fLljY4pIiIiadzFixcB8PDwAMDV1RVXV1fCwsKIi4t77rWjRo1i165dFCtWDICgoCAAihQp8pefQ+WziIikR46OjixdupR//etfAEybNo327dvz5MkTg5OJvDnp4uQPHx/Yvx8yZ4Zt28DNzehEIn8uMjKSBQsW8PXXX3Pnzh0AcuTIweDBgxk6dCi5cuUyOKGIiIjYksePHwPg5OSUdM3JyQmz2czTp0/JkCHDC3/e0qVLWbp0KXny5KFz585/+TlUPouISHplMpn45JNP8PDwoGfPnmzevJmbN2+yfft2fX0v6YLNl8+rV8P06ZbnK1dC6dLG5hH5M7dv3+brr79mwYIFPHr0CID8+fMzcuRI+vTpQ5YsWQxOKCIiIrYoU6ZMAMTGxiZdi46OxmQykTFjxv95fUJCAhMnTmT9+vXkzZuXJUuWkD179r/8HA4ONv9lh4iIyF/q0qUL+fPnp3Xr1pw4cYKqVas+dzeRiK2y6bEb585Bv36W559+Cu3aGZtH5EVCQkLo27cvhQoVws/Pj0ePHlGiRAlWrFjB5cuXGTFihIpnERERSTWenp6AZfYzQEREBA8fPqRgwYL/s+s5ISGBESNGsH79ekqXLo2/v3/Sz/8r2vksIiICtWvX5scff6Rw4cJcvXqV6tWr8/333xsdSyRV2ewWhCdPLHOeo6OhcWP44gujE4k87+TJk/j6+rJlyxbMZjMANWrUwMfHh+bNm2OnEzFFRETkDfD09KRUqVIEBQUxfvx4rl27BkDbtm25efMm27Zto1ChQjRr1oylS5eyZ88e7OzsqFy5Mhs2bABI+vE/o/JZRETEwsvLi+PHj9OqVSuOHz9Ow4YNWbZsGR9++KHR0URShcn8rPUymMlkIioqisyZM6fI+/XpA0uXwttvwy+/gMboiDUwm83s2bMHX19fDh8+nHS9ZcuW+Pj4UKNGDePCiYiISLoVHh7OxIkTOX78OBkzZqR169YMHTqUoKAgunfvTp06dVi4cCE1a9bk3r17//Pzn/34fzOZTADs3r2bJk2apPqvQ0REJK14+vQp3bt3Z+PGjQBMnDiRTz/9NOn/nSK2wibLZ39/y65nk8ly0GC9eikQUOQ1xMfH4+/vj5+fH7/++itgmX3YtWtXxowZQ8mSJQ1OKCIiIpLynn0BvXfvXho2bGhwGhEREeuSmJjIuHHj8PPzA6BHjx4sXLgQR0dHg5OJpBybG7tx5w4MHGh5Pn68imcx1pMnT1i2bBnTpk1LmqOYOXNm+vXrx4gRIyhQoICxAUVERETeAI3dEBER+V92dnb4+vri4eHB4MGDWbFiBdeuXWPz5s1ky5bN6HgiKcKmdj6bzdCqFezcCeXKwYkT8F9npIi8Effv32fu3LnMnj2b8PBwAHLmzIm3tzeDBg3C1dXV4IQiIiIiqe/Zzufvv/+e2rVrG5xGRETEeu3evZsOHToQFRVF6dKl2bVrF/nz5zc6lshrs6kTzVautBTPjo6W5yqe5U0LCwtj2LBhuLu7M2HCBMLDw/Hw8GDu3LmEhYXx6aefqngWERGRdEc7n0VERP5a06ZNCQwMJE+ePJw9e5aqVaty9uxZo2OJvDabKZ9v34ZhwyzPJ06E0qWNzSPpy9mzZ+nWrRuenp7MmjWLJ0+eULZsWdatW0dISAiDBg0iY8aMRscUERERMYTKZxERkb9Xrlw5jh8/TvHixbl58yY1a9bkwIEDRscSeS02Uz6PHAmPHkHFijB6tNFpJD0wm80EBgbSvHlzypQpwzfffENCQgL16tVjz549/Pzzz3Tq1AkHB5sbrS4iIiKSLCqfRUREXk7BggU5evQotWvX5tGjRzRt2pTVq1cbHUvkldlE+bxvH6xfD3Z2sHAhaG0rqSkxMZGtW7dSvXp13nvvPXbt2oXJZKJ9+/acOnWKAwcO0KhRo6QZhyIiIiLpncpnERGRl5cjRw727NlDhw4diIuLo3v37kyaNAkrObZNJFmSXT7fu3cPb29vKleuTN26dZk1axYJCQkvfO24cePw8vJ67r8ZM2a8duj/FB0NgwZZng8ZAuXLp+jbiySJiYlh2bJllCxZkjZt2nD8+HGcnJzo378/ISEh+Pv7U7FiRaNjioiIiFgdlc8iIiLJ4+zszLp16xj9f7f3f/LJJwwcOJD4+HiDk4kkT7LnAQwbNoygoCDatWtHWFgYc+fOxdnZmX79+v3Pa8+ePUvWrFnp2bNn0rWULuemTIFLl+Dtt+Gf/0zRtxYB4NGjRyxatIgZM2Zw69YtALJly8agQYPw9vYmT548BicUERERsW4aQyYiIpJ8dnZ2TJkyhYIFC+Lt7c3ChQu5efMm69evJ3PmzEbHE3kpJnMy9uyHhobSuHFjypUrx/r163nw4AHVqlWjUKFC7Nmz57nXPnnyhAoVKlC5cmVmzJiBs7MzmTJl+vMgJhNRUVHJ+stz5w4UKQKPH8PatdC580v/VJG/dffuXWbOnMm8efOIjIwEIG/evIwYMYJ+/fqRNWtWgxOKiIiIWLdnY8jOnz9PsWLFDE4jIiKSdm3ZsoUuXboQHR1NxYoV2blzJ7lz5zY6lsjfStbYjYsXLwLg4eEBgKurK66uroSFhREXF/fca//f//t/JCYm8ssvv1CtWjXKly+Pt7c3UVFRKRQdJkywFM+VK0OnTin2tpLOXbp0iQEDBlCwYEEmT55MZGQkXl5eLF26lCtXrjB69GgVzyIiIiLJoLEbIiIir6dNmzYcPHgQNzc3fvrpJ6pVq8aFCxeMjiXyt5JVPj9+/BgAJyenpGtOTk6YzWaePn363GvDw8PJly8f9evXZ/LkydSpU4c9e/bg5+eXArEhOBiWLLE8nzoVdLabvK6ffvqJ9u3b884777Bw4UJiYmKoWrUqW7ZsITg4mF69ej33Z19EREREXo7KZxERkddXrVo1jh07RuHChbl69SrVq1fn6NGjRscS+UvJGr72bGxGbGxs0rXo6GhMJhMZM2Z87rVNmzaladOmSR/Xr1+fypUrExgY+Dp5k/j4QGIivP8+1KqVIm8p6ZDZbGbfvn34+vpy8ODBpOvNmjXDx8eHWrVqJd0uKiIiIiKvRuWziIhIynjnnXf48ccfadmyJSdPnqR+/fqsWbOGdu3aGR1N5IWStfPZ09MTsMx+BoiIiODhw4cULFiQDBkyPPfa3bt3M2HCBM6fPw9YDm0D/nLu88s6fhx27gR7e/D1fe23k3QoPj6eb7/9lgoVKtC4cWMOHjyIvb09H374IWfOnCEgIIDatWureBYRERFJASqfRUREUk6uXLk4ePAgLVu2JCYmhvbt2zNz5kyjY4m8ULJ2Pnt6elKqVCmCgoIYP348165dA6Bt27bcvHmTbdu2UahQIZo1a4bZbGb9+vUEBgbSokULDh8+DEDXrl1fO/SXX1oeu3UDL6/XfjtJR54+fcry5cuZNm0aV65cASzfEOnbty8jRoygYMGCBicUERERsT0ODsn6skNERET+RubMmdmyZQtDhw5l/vz5DB8+nLCwMKZOnYqdXbL2moqkKpPZbDYn5yeEh4czceJEjh8/TsaMGWndujVDhw4lKCiI7t27U6dOHRYuXAjAhg0bWLZsGTdu3CBXrlx069aNHj16vDiIyURUVBSZM2f+y8//669QtqxlxvNvv8E77yQnvaRXDx8+ZO7cucyaNYt79+4B4Obmhre3N4MHD8bNzc3ghCIiIiK259ldZL///js5c+Y0OI2IiIjtMZvN+Pn58fHHHwPQvn17Vq1ahbOzs8HJRCySXT6nlpctnzt2BH9/y+P69W8onKRZ169fZ8aMGSxatCjpwMxChQoxatQoevXqlSJjYERERETkxZ6Vz/fv38fV1dXgNCIiIrZr7dq19OjRg7i4OGrWrMm2bdv0/16xCmmqfL5wAYoXB7PZsgO6TJk3GFDSlODgYPz8/FizZg3x8fEAlClTBh8fHzp06KBbP0VERETegGflc0REBNmyZTM4jYiIiG07dOgQbdq0ITIyEi8vL3bv3o2Hh4fRsSSdS1NDYKZOtRTPLVuqeJYXO3r0KK1ataJkyZKsXLmS+Ph46tSpw+7du/nll1/o0qWLimcRERGRN0wHDoqIiKS+unXrcuTIEQoUKMCFCxeoVq0aQUFBRseSdC7N7HyOiIB8+eDJEwgMhFq13mw+sV6JiYkEBATg6+vL0aNHAcufpzZt2uDj40PlypUNTigiIiKSPj3b+fzkyRMyZsxocBoREZH04ebNmzRr1owzZ86QOXNmNm7cSJMmTYyOJelUmtn5/M03luK5ZEmoWdPoNGINYmNjWblyJaVLl6ZVq1YcPXoUR0dH+vTpw/nz59m0aZOKZxEREREroDvPRERE3px8+fLxww8/0KBBAx4/fkzLli1ZuXKl0bEknUoT5bPZDAsWWJ4PGAD/t4FC0qmoqChmzJiBp6cnPXr0IDg4mKxZszJ27FhCQ0NZvHgxXl5eRscUERERkf+jsRsiIiJvVtasWQkICKBr167Ex8fTo0cPJk+ejJUMQJB0JE2M3ThyxDJmI1MmuHULdFZJ+vT7778za9Ys5s2bx8OHDwHIkycPw4cPZ8CAATrERkRERMTKPBu7YSVfcoiIiKQ7iYmJfPzxx0yZMgWAwYMHM3PmTH1jWN6YNHH/27Ndz507q3hOj65cucLUqVNZvnw50dHRABQtWpQxY8bQrVs3nJ2dDU4oIiIiIn/Gzi5N3GwpIiJik+zs7PDz8yNfvnyMGDGCuXPncvv2bdasWaM+Rd4Iq9/5fP8+5M0LsbFw6hRUrGhQQHnjTp8+ja+vLxs2bCAxMRGASpUq4ePjQ+vWrfVdOhERERErZzKZyJAhA7GxsUZHERERSff8/f3p1q0bsbGx1KpVi23btpEjRw6jY4mNs/ptCBs3WornsmVVPKcHZrOZAwcO0KhRI8qXL8+3335LYmIiTZo04eDBg5w4cYJ27dqpeBYRERFJI7RuExERsQ4dOnRgz549ZM2alR9++IGaNWty/fp1o2OJjbP68nndOstjly7G5pDUlZCQwIYNG6hUqRINGjRg37592Nvb06VLF3755Rd2795N3bp1k+YGioiIiEja4OCQJib9iYiIpAt16tThhx9+IG/evAQHB1OtWjXOnTtndCyxYVY9duPmTShQAMxmCAsDd3cDA0qqiI6OZuXKlUydOpVLly4BkDFjRnr37s3IkSPx8PAwOKGIiIiIvCqTyUS2bNmIiIgwOoqIiIj8h2vXrtGkSRPOnz9PtmzZ2LZtG++9957RscQGWfXO5w0bLMVzjRoqnm1NREQEkydPplChQgwYMIBLly7h6urK559/TlhYGLNnz1bxLCIiImIDNHZDRETE+ri7u3PkyBFq1KhBZGQkjRo1YuPGjUbHEhtk1eXz9u2Wx3btjM0hKefmzZuMGTMGd3d3xo8fz927dylQoABff/01YWFhfPHFF+TMmdPomCIiIiKSQlQ+i4iIWCdXV1f27dtH69atiY2NpUOHDsyePdvoWGJjrHbsRkQE5MwJ8fFw6RJ4ehqbT17P+fPnmTJlCt988w1xcXEAlCpVirFjx9KpUycyZMhgcEIRERERSWkmk4ncuXNz584do6OIiIjIn0hISGDo0KHMnz8fAB8fHyZPnqxztyRFWO3O5+++sxTPxYureE7LfvzxR1q3bk2JEiVYvnw5cXFx1KpVi507d3LmzBm6deum4llERETEhmnns4iIiHWzt7dn7ty5/Otf/wLA19eXjz76KGnzoMjrsNryeccOy2OrVsbmkOQzm80EBARQu3ZtqlevzrZt2wB4//33OXbsGIGBgTRv3lzfQRMRERFJBxwcHIyOICIiIn/DZDLxySefsGzZMuzt7Vm9ejUtWrTgjz/+MDqapHFWWT6bzXDggOV506bGZpGXFxcXx+rVqylTpgwtWrTghx9+IEOGDPTq1Yvg4GC2bt1KtWrVjI4pIiIiIm+Qdj6LiIikHT179mT79u1kypSJvXv3UqdOHe7evWt0LEnDrHLmc3AwlCwJzs6W2c9OTkank7/y+PFjlixZwvTp07l27RoALi4uDBgwgOHDh5MvXz6DE4qIiIiIEUwmE0WKFOHixYtGRxEREZFkOHnyJM2bNyc8PJzChQvz3XffUbRoUaNjSRpklffAHTxoeaxZU8WzNQsPD2f27NnMmTOHBw8eAJA7d26GDRvGwIEDyZ49u7EBRURERMRw2vksIiKS9lSuXJljx47RpEkTrly5QvXq1QkICKBy5cpGR5M0xirHbhw6ZHmsW9fYHPJioaGhDB06FHd3dyZOnMiDBw/w9PRkwYIFhIaGMm7cOBXPIiIiIgKofBYREUmrihYtyrFjxyhfvjzh4eHUrVuXXbt2GR1L0hirK5/NZjh82PK8Xj1Do8h/+fXXX+natStFihRhzpw5PH36lAoVKuDv78+FCxfo378/zs7ORscUEREREStiZ2d1X3KIiIjIS8qdOzeHDx+mUaNGPHnyhFatWrFixQqjY0kaYnUrwcuX4cEDy7iNChWMTiNms5nDhw/TtGlTypYty9q1a0lISKBhw4bs37+fU6dO0b59e+1oEREREZEXcnCwykl/IiIi8pKyZMnCjh076NatGwkJCfTs2RM/Pz+s5Bg5sXJWVz7/9JPlsWxZyJDB0CjpWkJCAps3b6Zq1arUrVuX7777Djs7Ozp27EhQUBB79+6lfv36mEwmo6OKiIiIiBXTJgUREZG0z9HRkRUrVjB69GgAfHx8GD16NImJiQYnE2tnddsQnpXPFSsamyO9iomJYfXq1UyZMoWQkBAAnJ2d6dmzJ6NGjcLT09PghCIiIiKSlqh8FhERsQ12dnZMmTKF3LlzM2bMGKZPn87du3dZtmwZjo6ORscTK6XyWQCIjIxkwYIFfP3119y5cweAHDlyMHjwYIYOHUquXLkMTigiIiIiaZHKZxEREdsyevRocufOTa9evVizZg3h4eFs3LgRFxcXo6OJFbKq8jkxEYKCLM9VPr8Zt2/f5uuvv2bBggU8evQIgPz58zNy5Ej69OlDlixZDE4oIiIiImmZymcRERHb061bN9566y0++OAD9uzZQ/369QkICOCtt94yOppYGaua+XzxIkRFQcaMUKyY0WlsW0hICH379qVQoUL4+fnx6NEjSpQowYoVK7h8+TIjRoxQ8SwiIiIir03ls4iIiG1q2rQpBw4cwNXVlZMnT1KzZk3CwsKMjiVWxqrK5+Bgy2Pp0qBDsVPHyZMnadeuHcWKFWPJkiXExsZSo0YNtm/fztmzZ/noo480p0dEREREUozKZxEREdtVtWpVjhw5QoECBbhw4QLVq1fn3LlzRscSK2JV5fOVK5bHd94xNoetMZvNfPfdd9StW5cqVaqwefNmzGYzLVu25MiRIxw5coSWLVtiZ2dVfxxERERExAaofBYREbFtxYsX59ixY5QoUYJbt25Rq1Ytjhw5YnQssRJW1TZevmx5LFLE2By2Ij4+nrVr11KuXDmaNm3K4cOHcXBw4KOPPuLcuXNs376dGjVqGB1TRERERGyYymcRERHblz9/fn744QeqV69OREQEDRs2ZMeOHUbHEitgleVz0aLG5kjrnjx5wpw5cyhatChdu3bl119/JXPmzIwYMYIrV66wYsUKSpYsaXRMEREREUkHVD6LiIikD66uruzbt48WLVoQHR1NmzZtWLZsmdGxxGBWNVlZO59fz/3795k7dy6zZ88mPDwcgJw5c+Lt7c2gQYNwdXU1OKGIiIiIpDcqn0VERNKPTJkysWXLFvr27cuKFSvo3bs3d+/e5eOPP8ZkMhkdTwxgVeXznTuWR5XPyRMWFsb06dNZsmQJT548AcDDw4PRo0fTs2dPMmbMaHBCEREREUmvVD6LiIikLw4ODixbtozcuXPj6+vL+PHjuXv3LtOnT9d5Y+mQVZXPAK6ulv/k7509exY/Pz/WrVtHQkICAGXLlsXHx4cPPvgABwer++0VERERkXRGa1IREZH0x2Qy8dVXX5E7d25GjhzJzJkz+f3331mxYgWOjo5Gx5M3yOpWgtr1/NfMZjM//PADvr6+7Nq1K+l6vXr18PHxoWHDhrqNQURERESshnY+i4iIpF8jRowgV65c9OjRg3Xr1nH//n02bdqEi4uL0dHkDbG6ve4qn18sMTGRrVu3Ur16dd577z127dqFyWSiffv2nDp1igMHDtCoUSMVzyIiIiJiVXR7rYiISPrWtWtXdu7cSebMmdm7dy/16tXj3r17RseSN8TqVoL58hmdwLrExMSwbNkySpYsSZs2bTh+/DhOTk7079+fkJAQ/P39qVixotExRUREREReSDufRUREpHHjxhw8eBA3NzdOnTpFzZo1CQ0NNTqWvAFWVz5nymR0Auvw6NEjpk6dSuHChenduze//fYb2bJlY9y4cYSGhrJgwQKKaJu4iIiIiFg5lc8iIiICULlyZY4ePYq7uzshISFUr16dM2fOGB1LUpnVzXxO7+Xz3bt3mTlzJvPmzSMyMhKAvHnzMmLECPr160fWrFkNTigiIiIi8vJUPouIiMgzXl5eHDt2jCZNmnDu3Dlq167Njh07qFWrltHRJJVY3c7njBmNTmCMS5cuMWDAAAoWLMjkyZOJjIzEy8uLpUuXcuXKFUaPHq3iWURERETSHAcHq9vvIiIiIgbKly8fgYGB1KxZk8jISBo2bMi2bduMjiWpxOrK5/S28/mnn36iffv2vPPOOyxcuJCYmBiqVq3Kli1bCA4OplevXjg5ORkdU0RERETklWjns4iIiPy3HDlysHfvXlq1akVMTAxt27Zl+fLlRseSVGB15XN62PlsNpvZu3cv9evXp1KlSmzcuBGz2UyzZs34/vvvOXbsGK1bt9bJ4CIiIiKS5ql8FhERkRfJmDEjmzZtolevXiQmJtKrVy+mTp1qdCxJYVZ3D5wtl8/x8fFs3LgRPz8/Tp8+DVgW4507d2bs2LGULl3a4IQiIiIiIilLGypERETkzzg4OLBkyRLeeust/Pz8GDNmDOHh4UyePBmTyWR0PEkBVlc+2+LYjadPn7J8+XKmTZvGlStXAMiUKRN9+/ZlxIgRFCxY0OCEIiIiIiKpQzufRURE5K+YTCZ8fX1xc3PDx8cHX19fwsPDWbBggc6OsAFW9ztoSzufHzx4wLx585g1axb37t0DwM3NDW9vbwYPHoybm5vBCUVEREREUpfKZxEREXkZY8eOxc3NjX79+rF06VIePHjA2rVrcXZ2NjqavAaVz6ng+vXrzJgxg0WLFvH48WMAChUqxKhRo+jVqxeZbHF7t4iIiIjIC2jHkoiIiLys3r174+rqSqdOndiyZQvNmzdn69atZMmSxeho8oqsbgBbWu5lg4OD6dGjB4ULF2bGjBk8fvyYMmXKsGbNGi5evMiQIUNUPIuIiIhIuqKdzyIiIpIcbdq04bvvviNLliwcPHiQevXqJU0UkLTH6srntLjz+ejRo7Rq1YqSJUuycuVK4uPjqVOnDrt37+aXX36hS5cu2vEhIiIiIumSymcRERFJrrp163Lo0CHeeustfvrpJ2rVqsW1a9eMjiWvQOXzK0pMTGTHjh3UrFmTmjVrsmPHDkwmE23btuXEiRMcOnSIJk2a6GROEREREUnXVD6LiIjIq6hQoQJHjhyhQIECXLhwgRo1anD+/HmjY0kyWV35bO1TKWJjY1m5ciWlS5emVatWHD16FEdHR/r06cP58+fZtGkTlStXNjqmiIiIiIhVUPksIiIir8rLy4ujR49SvHhxbty4Qa1atTh58qTRsSQZrK58ttadz1FRUcyYMQNPT0969OhBcHAwWbNmZezYsVy9epXFixfj5eVldEwREREREatiZ2d1X3KIiIhIGlKgQAECAwOpVKkS9+/fp169euzfv9/oWPKSrGolaDJBhgxGp3je77//zqeffoq7uzsjR47kxo0b5MmTh6+++opr167h6+tL3rx5jY4pIiIiImKVdPaJiIiIvK633nqLAwcO0KBBAx4/fkzz5s3ZuHGj0bHkJVhV+Zwpk6WAtgZXrlxh0KBBFCxYkC+//JKHDx9StGhRFi1axNWrV/Hx8SFbtmxGxxQRERERsWoauyEiIiIpIUuWLOzcuZMPPviA2NhYOnTowKJFi4yOJX/Dqspnaxi5cfr0aTp16kTRokWZP38+0dHRVKpUiY0bN3L+/Hn69u2Ls7Oz0TFFRERERNIElc8iIiKSUpycnFi/fj39+/fHbDbTv39/Jk+ejNlsNjqa/AmrugfOqPLZbDZz8OBBfH192bdvX9L1Jk2aMHbsWOrUqYPJWrZki4iIiIikISqfRUREJCXZ29szf/583NzcmDRpEuPHjyc8PJwpU6borAkrlK7L54SEBDZv3oyvry9BQUGA5Q9wx44dGTt2LO++++6bDSQiIiIiYmNUPouIiEhKM5lMfPnll7i5uTFq1CimT5/O/fv3WbJkic6bsDJW9bvxpqZZREdHs3LlSqZOncqlS5cAyJgxI71792bkyJF4eHi8mSAiIiIiIjZO5bOIiIiklpEjR/LWW2/Rq1cvVq5cycOHD1m/fj0ZrWG2rwBWVj5nypS67x8REcH8+fOZOXMmd+/eBcDV1ZUhQ4YwZMgQcubMmboBRERERETSGZXPIiIikpq6d+9Ojhw56NChA9u3b6dJkyZs376dbNmyGR1NSCcHDt68eZMxY8bg7u7O+PHjuXv3LgUKFODrr78mLCyML774QsWziIiIiEgqUPksIiIiqa1ly5bs2bOHrFmzEhgYSJ06dZI2noqxrKp8TumxG+fPn6dXr154eHgwdepU/vjjD0qVKsWqVau4fPkyw4YNw8XFJWU/qYiIiIiIJFH5LCIiIm9C7dq1OXz4MLly5eKXX36hZs2ahIaGGh0r3bOq8jmlxm78+OOPtG7dmhIlSrB8+XLi4uKoVasWO3fu5MyZM3Tr1o0MGTKkzCcTEREREZE/pfJZRERE3pRy5cpx5MgRChUqxKVLl6hRowbnzp0zOla6ZlXl8+vsfDabzQQEBFC7dm2qV6/Otm3bAHj//fc5duwYgYGBNG/eHJPJlEJpRURERETk76h8FhERkTepaNGiHD16lJIlS3Lr1i1q167NiRMnjI6VbllV+fwqM5/j4uJYvXo1ZcqUoUWLFvzwww9kyJCBXr16ERwczNatW6lWrVrKhxURERERkb+l8llERETetLx58xIYGEjVqlV5+PAh9evX58CBA0bHSpesqnxOztiNx48fM3PmTIoUKUL37t05d+4cLi4ujB49mqtXr7J06VKKFy+eemFFRERERORvOTg4GB1BRERE0iFXV1f27dtHgwYNePz4Mc2aNWPr1q1Gx0p3rKp8fpmxG+Hh4UyYMAF3d3eGDx/OtWvXyJ07N5MmTeL69etMmTKFfPnypX5YERERERH5W3Z2VvUlh4iIiKQjLi4u7Ny5k7Zt2xIbG8sHH3zAqlWrjI6VrljVSvCvxm6EhoYydOhQ3N3dmThxIg8ePMDT05MFCxYQGhrKuHHjyJ49+xvLKiIiIiIif09jN0RERMRITk5OfPvtt/To0YOEhAQ++ugjZs+ebXSsdCPZ5fO9e/fw9vamcuXK1K1bl1mzZpGQkPDC1wYGBtKqVSvKlStH27Zt/3a494vK519//ZWuXbtSpEgR5syZw9OnT6lQoQL+/v5cuHCB/v374/w6JxWKiIiIiKRzqbnGV/ksIiIiRnNwcGDp0qUMHz4cAG9vbyZOnIjZbDY2WDqQ7PJ52LBh7NmzhwYNGpA3b17mzp3L0qVL/+d1ly9fZtCgQTx8+JD27dtz+/Zt+vbty40bN/70vZ+Vz2azmUOHDtGkSRPKli3L2rVrSUhIoGHDhuzfv59Tp07Rvn17LWRFRERERFJAaq7xtWYXERERa2BnZ8f06dP54osvAJgwYQIjR44kMTHR4GS2LVnlc2hoKEFBQZQrV45JkyYlbVHftGnT/7x2+/btxMXFMXToUMaPH0/fvn2JiYkhICDgT9/f2TmBTZs2UaVKFerVq8eePXuws7OjY8eOBAUFsXfvXurXr4/JZErmL1NERERERF4ktdf4Kp9FRETEWphMJj7//HNmzpwJwNdff03v3r2Jj483OJntStbR0xcvXgTAw8MDsJwa6erqSlhYGHFxcWTIkCHptSEhIc+91tPT87n3eJHPPivH779fASzzWLp164a3tzeFCxcG4PHjx8mJKyIiIiJWIHPmzEZHkL+Q2mv8uLg4reNFRETEqvTu3ZuMGTMyYMAAVqxYwYMHD1i+fDlOTk5GR0szXnaNn6zy+dmi8T9/I5ycnDCbzTx9+vS5hemTJ0+ee62jo+Nz11/kWfEMEBMTw5IlS1iyZElyIoqIiIiIldEsPeuW2mv8xo0bp3hmERERkZS0fft23NzcjI6RprzsGj9Z5XOmTJkAiI2NTboWHR2NyWQi43+dFvjs42evjYmJee49XjWwiIiIiIikHK3xRURERCS1JGvm87Pb6kJDQwGIiIjg4cOHFCxY8LkdEf/52qtXrz73c4oWLfo6eUVEREREJAVpjS8iIiIiqSVZO589PT0pVaoUQUFBjB8/nmvXrgHQtm1bbt68ybZt2yhUqBDNmjWjZcuWLF++nFmzZnHp0iV27NiBo6MjzZo1S5VfiIiIiIiIJJ/W+CIiIiKSWpK18xlg4cKFNG7cmP3793P9+nUGDBhA7969uXHjBjNnzmTbtm0AFCtWjIULF+Lm5oa/vz+5c+dm0aJFFChQIMV/ESIiIiIi8uq0xhcRERGR1GAyaxCbiIiIiIiIiIiIiKSwZO98Tmn37t3D29ubypUrU7duXWbNmkVCQoLRseQ1rFixAi8vL5YuXWp0FHlFZ86coXv37pQvX55q1aoxcuRI7t69a3QseUUHDhygQ4cOlC9fntq1azNp0iSePn1qdCx5DTExMbRo0QIvLy9u3LhhdBx5BSdPnsTLy+u5/2rXrm10LJEUozW+7dEaP+3TGt+2aI1ve7TGT/u0xn+xZM18Tg3Dhg0jKCiIdu3aERYWxty5c3F2dqZfv35GR5Nkio+PZ8WKFUyfPt3oKPIaHj58SJ8+fYiLi6Njx47cuHGDgIAAbt26xfr1642OJ8kUFhbGsGHDcHNzo0OHDpw5c4aVK1diMpkYN26c0fHkFU2aNImLFy8aHUNew9mzZwFo164d+fPnB8DFxcXISCIpSmt826E1vm3QGt+2aI1vm7TGT/u0xn8xQ8vn0NBQgoKCKFeuHJMmTeLBgwdUq1aNTZs2aWGaBjVv3pzr169ToECBpJPPJe25fPky+fPnp06dOnh7e2M2m6latSq//PILsbGxODo6Gh1RkiFPnjxs27YNFxcXnjx5Qnx8PEFBQTg7OxsdTV7R/v37Wb9+PU5OTsTExBgdR17Rs4Vp3759yZIlC2+99ZbBiURSjtb4tkVrfNugNb5t0Rrf9miNbxu0xn8xQ8vnZ9/R8fDwAMDV1RVXV1fCwsKIi4sjQ4YMRsaTZKpSpQrTpk3j0KFDzJkzx+g48ooqVqzI5s2bkz4OCQnh0aNHuLu7a1GaBjk5OeHp6QlY/o5GRERQuXJlBg8ebHAyeRV37tzhk08+oWHDhkRGRnLy5EmjI8krerYwbd26NdHR0bi7u/PVV19RoUIFg5OJvD6t8W2L1vi2QWt826I1vm3RGt92aI3/YobOfH78+DFg+YfzGScnJ8xms2YVpUETJ06kVKlSRseQFPTbb7/Rp08fEhMTGT58uNFx5DUkJiYyevRoOnbsyMmTJ/nkk0+MjiTJlJiYyJgxY3BycuJf//qX0XHkNURHR5MpUya8vLzw8fFh2LBh3Lp1iyFDhhAVFWV0PJHXpjW+bdEa3/ZojW87tMZP+7TGtx1a4/85Q3c+Z8qUCYDY2Nika9HR0ZhMJjJmzGhULBEBAgMDGT58OE+fPuWzzz6jWbNmRkeS12BnZ0f79u354IMPOHToENu3b+fzzz8nS5YsRkeTl7Rw4UJOnjzJF198QWRkJNHR0QDcunWL7Nmza5ZYGuLs7MyOHTueu3b27FkOHjzI2bNnqVatmkHJRFKG1vgi1ktrfNuiNX7apzW+7dAa/88ZuvP52W0iz2aHRURE8PDhQwoWLKjb8UQM9N133zFw4EASExOZM2cOH374odGR5BUdOnSIFi1asGLFCgDi4uKIiYkhQ4YMz+1IE+t37NgxACZMmECjRo04c+YMAN26dWPv3r1GRpNkun79OpMmTeKbb75JuvbHH38AkDlzZqNiiaQYrfFFrJPW+LZDa3zboTW+7dAa/88ZuvPZ09OTUqVKERQUxPjx47l27RoAbdu2NTKWSLp2/fp1xo4dS3x8PFWrVuXChQtcuHABgF69eukQizSmRIkS3L17l5kzZ3L79m2Cg4OJjIykS5cumu+XxgwdOpQHDx4kfTx79mwuXbrEhAkTqFKlioHJJLly5MjBzp07iYiIICQkhMePH3Pq1ClKly5N6dKljY4n8tq0xhexPlrj2xat8W2H1vi2Q2v8P2do+QyWWwwmTpzI/v37yZgxIwMGDKB3795GxxJJt/z9/ZNO1z1y5AhHjhxJ+rFOnTppYZrG5M6dm2XLluHn54e/vz85cuRg0KBBDBw40OhokkyVK1d+7uM1a9YAULt2bfLly2dEJHlFLi4urFixAl9fXwICAjCZTLRo0YLx48djMpmMjieSIrTGF7EuWuPbFq3xbYfW+LZDa/w/ZzKbzWajQ4iIiIiIiIiIiIiIbTF05rOIiIiIiIiIiIiI2CaVzyIiIiIiIiIiIiKS4lQ+i4iIiIiIiIiIiEiKU/ksIiIiIiIiIiIiIilO5bOIiIiIiIiIiIiIpDiVzyIiIiIiIiIiIiKS4lQ+i4iIiIiIiIiIiEiKU/ksIiIiIiIiIiIiIilO5bOIiIiIiIiIiIiIpDiVzyIiqezjjz/Gy8uLli1bGh1FRERERERe07Vr1/Dy8sLLy4u2bdsaHUdExKqpfBYRSWXnz58HoFixYgYnERERERGR13Xu3Lmk5++++66BSURErJ/KZxGRVBQXF8fly5cBKF68uMFpRERERETkdZ09ezbpucpnEZG/pvJZRCQVXb58mbi4OODf5fO8efPw8vKiVKlSbN261cB0IiIiIiKSXP+587lMmTKApZCuVasWXl5elCtXjoCAAKPiiYhYFQejA4iI2LJnIzcA3nnnHT7//HO+/fZbXF1dmT17NhUrVjQwnYiIiIiIJIfZbCY4OBiArFmz4uHhQUBAAOPHjyc6OppChQoxe/Zs3nnnHYOTiohYB+18FhFJRb/99hsAOXLk4NNPP+Xbb7+lSJEi+Pv7JxXPN27cYPbs2Zw+fdrIqCIiIiIi8jeuXLlCVFQUAKVLl2bWrFmMHDmS6Oho6tWrx8aNG58rnitUqICXlxe///67UZFFRAylnc8iIqno2c7nhw8fcvDgQUqWLMmqVatwcXFJes2RI0eYM2cOJUuWNCqmiIiIiIi8hP+c9/zTTz9x9OhR7Ozs8Pb2ZsCAAZhMpqQfv3XrFlFRUWTPnp1cuXIZEVdExHDa+SwikoouXLgAwFtvvQVYdkrcuXMn6cdXrlzJhAkTABg4cCCNGjV68yFFREREROSl/Oe855iYGADefvttPvzww+eK5/3791O3bl0AIiIi8PLyomnTpm82rIiIFVD5LCKSSm7fvk1ERAQAn3zyCSVLluTp06cMHz6cp0+fAlC5cmXy5cuHi4sL06dP55///KeBiUVERERE5K/8Z/ncunVrAG7evMmYMWMwm81JP5YlS5akOxtLly5N9+7d6dWr1xvNKiJiDVQ+i4ikkv88bLB06dJMmzaNTJkycfHixaSSuXjx4kRGRlK0aFGaN29OlSpVjIorIiIiIiJ/ISEhIelMl+zZszNp0iSqVq0KwKFDh5g7d27Sa6tUqUKBAgUA6NChA5988gnt27d/86FFRAym8llEJJU8W5hmzpyZ/Pnz4+Hhwfjx4wHYtGkT27ZtS5oDV7RoUSOjioiIiIjI37h48WLSHYwlSpTA3t6eadOmkTNnTgDmzJnDoUOHnns98NwBhCIi6Y3KZxGRVDJo0CAuXLjAzz//nDT/rX379ly4cIELFy7w/vvvc+/ePQCuX7/O7t27jYwrIiIiIiJ/4T9HbpQoUQKwnO0yY8YMHBwcMJvNjBkzhtDQUGJjYwkLCwOgSJEihuQVEbEGJvN/DiUSEZE36smTJ3Tv3p1z585RsGBB9uzZY3QkERERERF5TVFRUVSoUAGA5s2bU6FCBbp27WpwKhGRN8/B6AAiIulZpkyZ2Lhxo9ExREREREQkBbm4uNC9e3c2bNhAQEAAmTJlMjqSiIghtPNZRERERERERERERFKcZj6LiIiIiIiIiIiISIpT+SwiIiIiIiIiIiIiKU7ls4iIiIiIiIiIiIikOJXPIiIiIiIiIiIiIpLiVD6LiIiIiIiIiIiISIpT+SwiIiIiIiIiIiIiKU7ls4iIiIiIiIiIiIikOJXPIiIiIiIiIiIiIpLiVD6LiIiIiIiIiIiISIpT+SwiIiIiIiIiIiIiKU7ls4iIiIiIiIiIiIikOJXPIiIiIiIiIiIiIpLi/j/5jd7dtvRIlwAAAABJRU5ErkJggg==
#
# '
# text/plain: <Figure size 1800x600 with 2 Axes>
# metadata: {}
# output_type: display_data
# tabbable: null
# tooltip: null
# e1d670f3f9fb467bb214057129777dc7:
# model_module: '@jupyter-widgets/controls'
# model_module_version: 2.0.0
# model_name: SliderStyleModel
# state:
# _model_module: '@jupyter-widgets/controls'
# _model_module_version: 2.0.0
# _model_name: SliderStyleModel
# _view_count: null
# _view_module: '@jupyter-widgets/base'
# _view_module_version: 2.0.0
# _view_name: StyleView
# description_width: initial
# handle_color: null
# f38de7846e9d42daa339e1189dd6fcd7:
# model_module: '@jupyter-widgets/base'
# model_module_version: 2.0.0
# model_name: LayoutModel
# state:
# _model_module: '@jupyter-widgets/base'
# _model_module_version: 2.0.0
# _model_name: LayoutModel
# _view_count: null
# _view_module: '@jupyter-widgets/base'
# _view_module_version: 2.0.0
# _view_name: LayoutView
# align_content: null
# align_items: null
# align_self: null
# border_bottom: null
# border_left: null
# border_right: null
# border_top: null
# bottom: null
# display: null
# flex: null
# flex_flow: null
# grid_area: null
# grid_auto_columns: null
# grid_auto_flow: null
# grid_auto_rows: null
# grid_column: null
# grid_gap: null
# grid_row: null
# grid_template_areas: null
# grid_template_columns: null
# grid_template_rows: null
# height: null
# justify_content: null
# justify_items: null
# left: null
# margin: null
# max_height: null
# max_width: null
# min_height: null
# min_width: null
# object_fit: null
# object_position: null
# order: null
# overflow: null
# padding: null
# right: null
# top: null
# visibility: null
# width: null
# f6a4919cc0504674a7ca7f6de8473b6f:
# model_module: '@jupyter-widgets/base'
# model_module_version: 2.0.0
# model_name: LayoutModel
# state:
# _model_module: '@jupyter-widgets/base'
# _model_module_version: 2.0.0
# _model_name: LayoutModel
# _view_count: null
# _view_module: '@jupyter-widgets/base'
# _view_module_version: 2.0.0
# _view_name: LayoutView
# align_content: null
# align_items: null
# align_self: null
# border_bottom: null
# border_left: null
# border_right: null
# border_top: null
# bottom: null
# display: null
# flex: null
# flex_flow: null
# grid_area: null
# grid_auto_columns: null
# grid_auto_flow: null
# grid_auto_rows: null
# grid_column: null
# grid_gap: null
# grid_row: null
# grid_template_areas: null
# grid_template_columns: null
# grid_template_rows: null
# height: null
# justify_content: null
# justify_items: null
# left: null
# margin: null
# max_height: null
# max_width: null
# min_height: null
# min_width: null
# object_fit: null
# object_position: null
# order: null
# overflow: null
# padding: null
# right: null
# top: null
# visibility: null
# width: null
# ff0f6e7fb87e4797835013ad57a1d95f:
# model_module: '@jupyter-widgets/output'
# model_module_version: 1.0.0
# model_name: OutputModel
# state:
# _dom_classes: []
# _model_module: '@jupyter-widgets/output'
# _model_module_version: 1.0.0
# _model_name: OutputModel
# _view_count: null
# _view_module: '@jupyter-widgets/output'
# _view_module_version: 1.0.0
# _view_name: OutputView
# layout: IPY_MODEL_a51b014f6f254aaea8830531a0d85123
# msg_id: ''
# outputs:
# - data:
# image/png: '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
#
# '
# text/plain: <Figure size 900x600 with 1 Axes>
# metadata: {}
# output_type: display_data
# tabbable: null
# tooltip: null
# version_major: 2
# version_minor: 0
# ---
# %% [markdown]
# # The Diamond OLG Model
#
# [](https://econ-ark.org/materials/diamondolg#launch)
# %% [markdown]
# ### Convergence of OLG Economy to Steady State
# %%
# Some initial setup
from HARK.ConsumptionSaving.ConsIndShockModel import (
PerfForesightConsumerType,
init_perfect_foresight,
)
import seaborn as sns
from matplotlib import rc, rcParams
import ipywidgets as widgets
from ipywidgets import interact, fixed
from matplotlib import pyplot as plt
import numpy as np
import scipy as sp
plt.style.use("seaborn-darkgrid")
palette = plt.get_cmap("Dark2")
# For plots
rc("font", weight="bold")
rcParams["text.latex.preamble"] = r"\usepackage{sfmath} \boldmath"
sns.set_style("white", {"axes.edgecolor": "black"})
years_per_gen = 30
# %% jupyter={"source_hidden": true}
# Define a function that plots something given some inputs
def plot1(Epsilon, DiscFac, PopGrowth, YearsPerGeneration, Initialk):
"""Inputs:
Epsilon: Elasticity of output with respect to capital/labour ratio
DiscFac: One period discount factor
YearPerGeneration: No. of years per generation
PopGrowth: Gross growth rate of population in one period"""
#
kMax = 0.1
# Define some parameters
Beta = DiscFac**YearsPerGeneration
xi = PopGrowth**YearsPerGeneration
Q = (1 - Epsilon) * (Beta / (1 + Beta)) / xi
kBar = Q ** (1 / (1 - Epsilon))
# Create an agent that will solve the consumption problem
PFagent = PerfForesightConsumerType(**init_perfect_foresight)
PFagent.cycles = 1 # let the agent live the cycle of periods just once
PFagent.T_cycle = 2 # Number of periods in the cycle
# Income only in the first period
PFagent.assign_parameters(PermGroFac=[0.000001])
PFagent.LivPrb = [1.0]
PFagent.DiscFac = Beta
# Hark seems to have trouble with log-utility
# so pass rho = 1 + something small.
PFagent.CRRA = 1.001
PFagent.solve()
# Plot the OLG capital accumulation curve and 45 deg line
plt.figure(figsize=(9, 6))
kt_range = np.linspace(0, kMax, 300)
# Analitical solution plot
ktp1 = Q * kt_range**Epsilon
plt.plot(kt_range, ktp1, "b-", label="Capital accumulation curve")
plt.plot(kt_range, kt_range, "k-", label="45 Degree line")
# Plot the path
kt_ar = Initialk
ktp1_ar = 0.0
for i in range(3):
# Compute Next Period's capital using HARK
wage = (1 - Epsilon) * kt_ar**Epsilon
c = PFagent.solution[0].cFunc(wage)
a = wage - c
k1 = a / xi
plt.arrow(
kt_ar,
ktp1_ar,
0.0,
k1 - ktp1_ar,
length_includes_head=True,
lw=0.01,
width=0.0005,
color="black",
edgecolor=None,
)
plt.arrow(
kt_ar,
k1,
k1 - kt_ar,
0.0,
length_includes_head=True,
lw=0.01,
width=0.0005,
color="black",
edgecolor=None,
)
# Update arrow
kt_ar = k1
ktp1_ar = kt_ar
# Plot kbar and initial k
plt.plot(kBar, kBar, "ro", label=r"$\bar{k}$")
plt.plot(Initialk, 0.0005, "co", label="$k_0$")
plt.title("Convergence of OLG Economy to Steady State", fontsize=20, y=1.05)
plt.legend()
plt.xlim(0, kMax)
plt.ylim(0, kMax)
plt.axvline(0, color="black")
plt.axhline(0, color="black")
plt.xlabel("$k_t$", fontsize=12)
plt.ylabel("$k_{t+1}$", fontsize=12)
plt.show()
return None
# %% jupyter={"source_hidden": true}
# Define some widgets to control the plot
# Define a slider for Epsilon
Epsilon_widget1 = widgets.FloatSlider(
min=0.2,
max=0.4,
step=0.01,
value=0.33,
continuous_update=False,
readout_format=".3f",
description=r"Capital Share $\epsilon$",
style={"description_width": "initial"},
)
# Define a slider for the discount factor
DiscFac_widget1 = widgets.FloatSlider(
min=0.94,
max=0.99,
step=0.001,
value=0.96,
continuous_update=False,
readout_format=".3f",
description=r"Discount Factor $\beta$",
style={"description_width": "initial"},
)
# Define a slider for pop. growth
PopGrowth_widget1 = widgets.FloatSlider(
min=0.98,
max=1.05,
step=0.001,
value=1.01,
continuous_update=False,
readout_format=".3f",
description=r"Pop. Growth $\Xi$",
style={"description_width": "initial"},
)
# Define a slider for initial k
Initialk_widget1 = widgets.FloatSlider(
min=0.01,
max=0.1,
step=0.01,
value=0.08,
continuous_update=True,
readout_format=".3f",
description="Init. capital ratio",
style={"description_width": "initial"},
)
# %%
# Make the widget
interact(
plot1,
Epsilon=Epsilon_widget1,
DiscFac=DiscFac_widget1,
PopGrowth=PopGrowth_widget1,
YearsPerGeneration=fixed(years_per_gen),
Initialk=Initialk_widget1,
)
# %% [markdown]
# ### Gross and Net Per Capita Output as a Function of k
# %% jupyter={"source_hidden": true}
# Define a function that plots something given some inputs
def plot2(Epsilon, PopGrowth, YearsPerGeneration):
"""Inputs:
Epsilon: Elasticity of output with respect to capital/labour ratio
DiscFac: One period discount factor
YearPerGeneration: No. of years per generation
PopGrowth: Gross growth rate of population in one period"""
kMax = 5.5
# Define some parameters
xi = PopGrowth**YearsPerGeneration
Xi = xi - 1
Xi ** (1 / (Epsilon - 1))
# Plot the production function and depreciation/dilution curves
kt_range = np.linspace(0, kMax, 500)
plt.figure(figsize=(18, 6))
plt.suptitle("Gross and Net Per Capita Output as a Function of $k$", fontsize=20)
plt.subplot(1, 2, 1)
plt.plot(kt_range, kt_range**Epsilon, "b-", label="$f(k)$")
plt.plot(kt_range, Xi * kt_range, "k-", label="$Xi * k$")
plt.legend()
plt.xlim(0, kMax)
plt.ylim(0, 3)
plt.xlabel("$k_t$")
plt.axvline(0, color="black")
plt.axhline(0, color="black")
plt.subplot(1, 2, 2)
plt.plot(
kt_range, kt_range**Epsilon - Xi * kt_range, "k-", label="$f(k) - Xi * k$"
)
plt.legend()
plt.xlim(0, kMax)
plt.ylim(0, 1)
plt.xlabel("$k_t$", fontsize=12)
plt.axvline(0, color="black")
plt.axhline(0, color="black")
plt.show()
return None
# %% jupyter={"source_hidden": true}
# Define some widgets to control the plot
# Define a slider for Epsilon
Epsilon_widget2 = widgets.FloatSlider(
min=0.2,
max=0.4,
step=0.01,
value=0.33,
continuous_update=False,
readout_format=".3f",
description=r"Capital Share $\epsilon$",
style={"description_width": "initial"},
)
# Define a slider for pop. growth
PopGrowth_widget2 = widgets.FloatSlider(
min=0.98,
max=1.05,
step=0.001,
value=1.01,
continuous_update=False,
readout_format=".3f",
description=r"Pop. Growth $\Xi$",
style={"description_width": "initial"},
)
# %%
# Make the widget
interact(
plot2,
Epsilon=Epsilon_widget2,
PopGrowth=PopGrowth_widget2,
YearsPerGeneration=fixed(years_per_gen),
)
# %% [markdown]
# ### Pay As You Go (PAYG) Social Security system
#
# Suppose that initially this economy had no Social Security system and we are interested in the effects of introducing a Pay-As-You-Go Social Security system that is expected to remain a constant size from generation to generation from now on: $z_{2,t+1} = − z_{1,t+1}$ while $z_{1,t+1} = z_{1,t}$, so that taxes are greater than transfers when young and transfers are greater than taxes when old.
# %% jupyter={"source_hidden": true}
# Define a function that plots something given some inputs
def plot3(Epsilon, DiscFac, z, Rfree, YearsPerGeneration):
"""Inputs:
Epsilon: Elasticity of output with respect to capital/labour ratio
DiscFac: One period discount factor
z: Social Security Transfers
Rfree: Interest rate
YearPerGeneration: No. of years per generation
"""
# unchanged parameters
kMax = 0.1
PopGrowth = 1
# Define some parameters
Beta = DiscFac**YearsPerGeneration
xi = PopGrowth**YearsPerGeneration
# Before introduction of PAYG System
Q = (1 - Epsilon) * (Beta / (1 + Beta)) / xi
kBar_old = Q ** (1 / (1 - Epsilon))
# After introduction of PAYG System
transfers = z * (1 + (Rfree * Beta)) / (Rfree * (1 + Beta))
solution = sp.optimize.root(lambda k: k - Q * (k**Epsilon) + transfers, kBar_old)
kBar_new = solution.x
# Create an agent that will solve the consumption problemFac
PFagent = PerfForesightConsumerType(**init_perfect_foresight)
PFagent.cycles = 1 # let the agent live the cycle of periods just once
PFagent.T_cycle = 2 # Number of periods in the cycle
# Income only in the first period
PFagent.assign_parameters(PermGroFac=[0.000001])
PFagent.LivPrb = [1.0]
PFagent.DiscFac = Beta
# Hark seems to have trouble with log-utility
# so pass rho = 1 + something small.
PFagent.CRRA = 1.001
PFagent.solve()
# Plot the OLG capital accumulation curve and 45 deg line
plt.figure(figsize=(9, 6))
kt_range = np.linspace(0, kMax, 300)
# Analitical solution plot
ktp1 = Q * kt_range**Epsilon
ktp1_PAYG = Q * kt_range**Epsilon - transfers
plt.plot(kt_range, ktp1, "b-", label="Capital accumulation curve")
plt.plot(kt_range, ktp1_PAYG, "g-", label="Capital accumulation curve with PAYG")
plt.plot(kt_range, kt_range, "k-", label="45 Degree line")
# Plot the path
kt_PAYG = kBar_old
ktp1_PAYG = kBar_old
while np.abs(kBar_new - kt_PAYG) > 0.0015: # Create as many arrows as necessary
# Compute Next Period's capital using HARK
wage = (1 - Epsilon) * kt_PAYG**Epsilon
net_wage = (1 - z) * wage # Agent has to pay transfers
PFagent.solution[0].cFunc(net_wage)
k1 = (wage * (Beta / (1 + Beta)) - transfers) / (
xi
) # capital accumulation differs
plt.arrow(
kt_PAYG,
k1,
k1 - kt_PAYG,
0.0,
length_includes_head=True,
lw=0.01,
width=0.0005,
color="black",
edgecolor=None,
)
plt.arrow(
kt_PAYG,
ktp1_PAYG,
0.0,
k1 - ktp1_PAYG,
length_includes_head=True,
lw=0.01,
width=0.0005,
color="black",
edgecolor=None,
)
# Update arrow
kt_PAYG = k1
ktp1_PAYG = kt_PAYG
# Plot kbar and initial k
plt.plot(kBar_old, kBar_old, "co", label=r"$\bar{k}$")
plt.plot(kBar_new, kBar_new, "ro", label=r"$\bar{k}_{PAYG}$")
plt.title(
"Convergence of OLG Economy After Intro of PAYG Social Security System",
fontsize=20,
y=1.05,
)
plt.legend()
plt.xlim(0, kMax)
plt.ylim(0, kMax)
plt.xlabel("$k_t$", fontsize=12)
plt.ylabel("$k_{t+1}$", fontsize=12)
plt.axvline(0, color="black")
plt.axhline(0, color="black")
plt.show()
return None
# %% jupyter={"source_hidden": true}
# Define a slider for Epsilon
Epsilon_widget3 = widgets.FloatSlider(
min=0.2,
max=0.4,
step=0.01,
value=0.33,
continuous_update=False,
readout_format=".3f",
description=r"Capital Share $\epsilon$",
style={"description_width": "initial"},
)
# Define a slider for the discount factor
DiscFac_widget3 = widgets.FloatSlider(
min=0.95,
max=0.99,
step=0.001,
value=0.96,
continuous_update=False,
readout_format=".3f",
description=r"Discount Factor $\beta$",
style={"description_width": "initial"},
)
# Define a slider for the discount factor
Rfree_widget3 = widgets.FloatSlider(
min=1.0,
max=1.1,
step=0.0001,
value=1.03,
continuous_update=True,
readout_format=".3f",
description="$Rfree$",
)
# Define a slider for z
z_widget3 = widgets.FloatSlider(
min=0,
max=0.02,
step=0.005,
value=0.015,
continuous_update=True,
readout_format=".3f",
description="$z$",
)
# %%
# Make the widget
interact(
plot3,
Epsilon=Epsilon_widget3,
DiscFac=DiscFac_widget3,
Rfree=Rfree_widget3,
z=z_widget3,
YearsPerGeneration=fixed(years_per_gen),
)
|
993,854 | e42230af564c27ce25271e5bc3fe73d4df9404df | email = "peterukpong.pou@gmail.com"
name = "Peter Ukpong"
hng_id = "HNG-02417"
lang = "Python"
print("Hello World, this is "+ name +" with HNGi7 ID " + hng_id + " using "+ lang + " for stage 2 task email " + email) |
993,855 | 401e52015764a103bc5f24cec9d9ee23bd4935fc | class Task:
def __init__(self,
name,
description,
status,
date_stop,
user_id,
task_id = 0,
date_start = ""
):
self.name = name
self.description = description
self.status = status
self.date_stop = date_stop
self.user_id = user_id
self.task_id = task_id
self.date_start = date_start
def __str__(self):
return "| %2d | %20s | %20s | %20s | %20s | %20s | %2d |" % \
(self.task_id, self.name, self.description, self.status, self.date_start, self.date_stop, self.user_id)
|
993,856 | c78fb4cf9816ff2215c8fc454240151710f7762e | arr = [(1,3), (2,4)]
if __name__ == '__main__':
for x,y in arr:
print(x)
print(y) |
993,857 | d0541b4de09d89e172ea2468552bb2e6b7523641 |
from . import base
from .sensors_packet_pb2 import sensors_packet as pkt
class Producer(base.SensorProducer):
event_name = 'temperature'
schema = {
'type': 'object',
'properties': {
'temperature': {
'type': 'number'
},
},
'required': ['temperature']
}
@staticmethod
def pack(*, request, temperature):
return pkt(
sensor_temperature=pkt.SensorTemperaturePayload(
temperature=temperature,
)
)
|
993,858 | 6da00f1cd9a3421685ec555ef505c419ea7250a4 | # Generated by Django 2.2.1 on 2019-08-30 08:31
from django.db import migrations, models
class Migration(migrations.Migration):
dependencies = [
('maps', '0014_auto_20190830_0657'),
]
operations = [
migrations.AlterField(
model_name='map',
name='bbox',
field=models.CharField(blank=True, max_length=100, null=True, verbose_name='extent'),
),
]
|
993,859 | ec79cb4774052bd83ec16a5f17e7fdacf416aaca | import paho.mqtt.client as mqtt
import myMqtt as my
import pymysql as sql
brkIp = my.brkIp
brkPort = my.Port
brkKeepAlive = my.brkKeepAlive
tls_arr = my.tls_arr
conn = sql.connect(host = '127.0.0.1', user = 'root', password = 'ziumks', db = 'jeju', charset = 'utf8')
def on_connect(client, userdata, flags, rc):
print("connected with result code " + str(rc))
client.subscribe("cdns/T12/#")
def on_message(client, userdata, msg):
data = str(msg.payload.decode('utf8'))
if 'period' in eval(data).keys():
print("check period")
qry = 'update response set period = (%s)'
param = data['period']
try:
with conn.cursor() as cursor:
cursor.execute(qry, param)
conn.commit()
except TypeError as e:
print(e)
if 'abc' not in eval(data).keys():
pass
def cdnsCallback(client, userdata, msg):
data = str(msg.payload.decode('utf8')
myTopic = msg.topic
deviceNo = myTopic.split('/')[2]
if 'Q' in eval(data).keys():
sendTopic = 'cdnc/' + myTopic[1] + '/' + myTpic[2]
qry= 'select MAX(time) from jeju_ack where deviceNo = %s'
with conn.cursor(sql.cursors.DictCursor) as cursor:
cursor.execute(qry, deviceNo)
rows = cursor.fetchone()
now = time.time()
result = {'t':now, 'cdn':'ON', ackt : row['time']}
pass
def setCallback(client, userdata, msg): # managing max, min of hm, hx, cm, cx, tm, tx, t, nt (noti_time)
data = str(msg.payload.decode('utf8'))
myTopic = msg.topic
deviceNo = myTopic.split('/')[2]
data = json.dumps(data)
qry = 'insert into setting (deviceNo, nt, hm, hx, cm, cx, tm, tx, t) values (%s, %s, %s, %s, %s, %s, %s, %s, %s)'
param = (deviceNo, data['nt'], data['hm'], data['hx'], data['cm'], data['cx'], data['tm'], data['tx'], data['t'])
with conn.cursor() as cursor:
cursor.execute(qry, param)
cursor.commit()
except TypeError as e:
print(e)
pass
def recvData():
client = mqtt.Client()
client.on_connect = on_connect
client.on_message = on_message
# client.message_callback_add("cdns/S002/#", cdnsCallback)
client.message_callback_add("set/S002/#", setCallback)
# client.message_callback_add("")
client.connect(brkIp, brkPort, brkKeepAlive)
client.loop_forever()
|
993,860 | accb129371acf79771eeb6d4d0ae1b217d702812 | """Custom serializers for oscarapi."""
import warnings
from django.core.urlresolvers import reverse, NoReverseMatch
from rest_framework import serializers
from oscarapi import serializers as oscarapi_serializers
from oscar.core.loading import get_class, get_model
Selector = get_class("partner.strategy", "Selector")
ShippingAddress = get_model("order", "ShippingAddress")
class ProductSerializer(oscarapi_serializers.ProductSerializer):
price = serializers.SerializerMethodField()
def get_price(self, obj):
request = self.context.get("request")
strategy = Selector().strategy(request=request, user=request.user)
ser = oscarapi_serializers.PriceSerializer(
strategy.fetch_for_product(obj).price, context={"request": request}
)
return ser.data
class OrderSerializer(oscarapi_serializers.OrderSerializer):
def get_payment_url(self, obj):
try:
reverse("api-payment", args=(obj.pk,))
return obj.pk
except NoReverseMatch:
msg = (
"You need to implement a view named 'api-payment' "
"which redirects to the payment provider and sets up the "
"callbacks."
)
warnings.warn(msg)
return msg
class CheckoutSerializer(oscarapi_serializers.CheckoutSerializer):
def create(self, validated_data):
try:
basket = validated_data.get("basket")
order_number = self.generate_order_number(basket)
request = self.context["request"]
if validated_data.get("shipping_address"):
shipping_address = ShippingAddress(**validated_data["shipping_address"])
else:
shipping_address = None
return self.place_order(
order_number=order_number,
user=request.user,
basket=basket,
shipping_address=shipping_address,
shipping_method=validated_data.get("shipping_method"),
shipping_charge=validated_data.get("shipping_charge"),
billing_address=validated_data.get("billing_address"),
order_total=validated_data.get("total"),
guest_email=validated_data.get("guest_email") or "",
)
except ValueError as e:
raise exceptions.NotAcceptable(e.message)
|
993,861 | 7a5bb64becd53c68fce3975381b5c12faf6fe9e7 | #coding:utf-8
import numpy as np
import pickle
import jieba
with open('newkey.txt','w',encoding='utf8') as newkey,open('newvalue.txt','w',encoding='utf8') as newvalue,open('test.txt','w',encoding='utf8') as testFile ,\
open('testoutput.txt','w',encoding='utf8') as testoutputFile ,open('wordlist.json', 'wb') as wordlist:
newSet=set()
conversationDictionary = np.load('conversationDictionary.npy').item()
allLen=len(conversationDictionary.items())
testLen=allLen*0.01
trainLen=allLen-testLen
i=0
for key,value in conversationDictionary.items():
key=' '.join(jieba.cut(key))
value=' '.join(jieba.cut(value))
if i<trainLen:
newkey.write(key+'\n')
newvalue.write(value+'\n')
newSet = newSet | set(key.split()) | set(value.split())
else:
testFile.write(key+'\n')
testoutputFile.write(value+'\n')
i += 1
newDict=dict()
i=0
for word in newSet:
newDict[word]=i
i=i+1
print('wordnum:%d\n' % i)
print('trainlen:%d\n' % trainLen)
print('testlen:%d\n' % testLen)
pickle.dump(newDict, wordlist)
|
993,862 | 03312319f71eb9612dc9f36d8065ee4fc73b2f6c | print("hello world")
print("Done the work")
print("Edited again") |
993,863 | e77ce799d4e1811ad7f62de94be5e9861e1df66a | import os
import random
import string
from configparser import ConfigParser
import netifaces
from scale.logger import create_logger
from scale.network.node import Node
class VPNManager:
def __init__(self, config):
self.logger = create_logger('VPN')
self.config = config
self.nodes: list[Node] = []
self.iface = self.config.network['interface']
pass
# Read the interfaces from the host
def get_interfaces(self):
return netifaces.interfaces()
def bootstrap(self):
# Check if the VPN is running
if (len(self.config.network['passKey']) == 0):
self.logger.info('No passkey found. Preparing one')
self.config.network['passKey'] = self.generate_pass_key(32)
self.config.save()
if (len(self.config.network['publicKey']) < 1 or len(self.config.network['privateKey']) < 1):
self.logger.info('No keys found. Generating...')
self.generate_keys()
self.generate_wg_config()
self.logger.info('Bootstrapped VPN...')
def check_if_wg_running(self):
return self.get_interfaces().__contains__(self.iface)
def generate_wg_config(self) -> None:
config_path = os.path.join(
'/etc/wireguard', '{}.conf'.format(self.iface))
self.logger.info('Generating WG config [{}]'.format(config_path))
with open(config_path, 'w') as f:
iconfig = ConfigParser()
iconfig.optionxform = str
iconfig.add_section('Interface')
iconfig.set('Interface', 'PrivateKey',
self.config.network['privateKey'])
iconfig.set('Interface', 'Address ', '10.0.1.1/24')
iconfig.set('Interface', 'ListenPort ', str(
self.config.network['discoveryPort'] - 1))
for node in self.nodes:
iconfig.add_section('Interface')
iconfig.set('Interface', 'PublicKey', node.public_key)
# TODO: Add support for multiple interfaces~
for iface in node.interfaces:
if iface.name == self.iface:
iconfig.set('Interface', 'Address', iface.ip)
iconfig.write(f)
pass
def generate_pass_key(self, length):
return ''.join(random.choice(string.ascii_letters) for _ in range(length))
def generate_keys(self):
privateKey = os.popen('wg genkey').read().replace('\n', '')
publicKey = os.popen('echo {} | wg pubkey'.format(
privateKey)).read().replace('\n', '')
self.config.network['privateKey'] = privateKey
self.config.network['publicKey'] = publicKey
self.logger.info('Private key: {}'.format(privateKey))
self.logger.info('Public key: {}'.format(publicKey))
self.config.save()
pass
def connect(self):
# Restart with the new config
if self.check_if_wg_running():
self.stop()
self.generate_wg_config()
exit_code = os.system('wg-quick up {}'.format(self.iface))
if (exit_code == 0):
self.logger.info('WireGuard started')
else:
self.logger.fatal('WireGuard failed to start')
def stop(self):
exit_code = os.system(
'wg-quick down {}'.format(self.iface))
if (exit_code == 0):
self.logger.info('WireGuard stopped')
else:
self.logger.fatal('WireGuard failed to stop')
|
993,864 | c0da824b8560a155a94e9d371528a4d9a5226490 | import subprocess
import glob
import os
path = "/media/edison/STORAGE/Files/Music/"
songs = glob.glob(path + "*.mp4")
name = "/media/edison/STORAGE/Files/Music/Juice\ WRLD\ –\ Lucid\ Dreams.mp4"
# subprocess.call("cd venv", shell=True)
# subprocess.call("cd media/edison/STORAGE/Files/Screen\ Record")
# subprocess.call("pwd", cwd="/media/edison/STORAGE/Files/")
# subprocess.call("pwd", cwd="/media/edison/STORAGE/Files/Screen Record")
# subprocess.call("ffmpeg -i "+ name +" -f mp3 -vn exported/gu.mp3", cwd="/media/edison/STORAGE/Files/Music", shell=True)
# subprocess.call("ls", cwd="/media/edison/STORAGE/Files/Music", shell=True)
# print (songs[1])
# print(songs)
for e in songs:
# newStr = e.replace(path, "")
song_name = e.replace(path, "").replace(" ", "")
newStr = e.replace(" ", "\ ")
chars = set('()')
if any((c in chars) for c in newStr):
print(song_name + " " + "was found.")
os.chdir(path)
print(song_name)
os.rename(e.replace(path, ""), song_name.replace("(", "").replace(")", ""))
subprocess.call("ffmpeg -i " + newStr + " -f mp3 -vn exported/"+ song_name +".mp3", cwd="/media/edison/STORAGE/Files/Music", shell=True)
print(songs.index(e))
|
993,865 | d85c4aea96f53f8721c946921be7aed5b4e453ad | import pickle
d = {}
count = 0
name = "jesus"
with open("game_data_" + name + ".txt", "r") as f:
for line in f:
count += 1
if (count % 1000000 == 0):
print(count)
temp = line.strip().split(" ")
if (len(temp) < 3):
print("test" + line + "test")
d[int(temp[0])] = [int(temp[1]), int(temp[2])]
print("Data processed, dumping.")
pickle.dump(d, open("game_data_" + name + ".p", "wb")) |
993,866 | cd825716fa3fece7c9a33a709a8ff5708da80ef4 | GOOGLE_PLACES_API_KEY = 'AIzaSyCLNuufBC6PSot71KVsulAMHEppB4192CI' |
993,867 | fbdfc8d5da55aae72f872661eb7ab03bb8ff6304 | #!/usr/bin/env python
#coding=utf-8
"""def spam(text):
print text,"spam"
def spamSum(a,b):
print a+b
print "ffff"
X = 88
def f():
global x
x = 99
message = "First version 22"
def printer():
print "reloaded:", message
def tester():
print "It's Christmas in Heaven.."
if __name__ == "__main__":
tester()"""
print "I'm:",__name__
def tester():
print "It's Christmas in Heaven.."
def minmax(test,*args):
res = args[0]
for arg in args[1:]:
if test(arg,res):
res = arg
return res
def lessthan(x,y):return x<y
def grtrthan(x,y):return x>y
if __name__ == "__main__":
print minmax(lessthan,4,2,1,5,6,3)
print minmax(grtrthan,4,2,1,5,6,3)
|
993,868 | 50d1276122d7d86cc84a980c015acacff1d676cd | import itertools
n, m = map(int, input().split())
a = list(map(int, input().split()))
a.sort()
for i in sorted(list(set(itertools.combinations_with_replacement(a, m)))):
print(*i)
|
993,869 | be79be05173ab97b3fd1fd667b6cf74be00d2cd1 | a,b=list(map(str,input().split()))
for i in range(1,int(b)+1):
print(a)
|
993,870 | 0d335122ecf48dcfb339d044d35c9b87f0e9f2be | # -*- coding: utf-8 -*-
"""
Created on Mon Oct 12 19:21:32 2020
@author: gaurav baba
"""
import pandas as pd
import numpy as np
import matplotlib.pyplot as plt
import seaborn as sns
%matplotlib inline
train=pd.read_csv("C:/daase/titanic_train.csv")
train.head()
train.isnull()
sns.heatmap(train.isnull(), yticklabels=False, cbar=False, cmap='viridis')
sns.set_style('whitegrid')
sns.countplot(x='Survived', data=train)
sns.set_style('whitegrid')
sns.countplot(x='Survived', hue='Sex', data=train, palette='RdBu_r')
sns.set_style('whitegrid')
sns.countplot(x='Survived', hue='Pclass', data=train, palette='rainbow')
sns.distplot(train['Age'].dropna(), kde=False, color='darkred', bins=40)
train['Age'].hist(bins=30, color='darkred', alpha=0.3)
sns.countplot(x='SibSp', data=train)
train['Fare'].hist(color='green', bins=40, figsize=(8,4))
plt.figure(figsize=(12,7))
sns.boxplot(x='Pclass', y='Age', data=train, palette='winter')
def impute_age(cols):
Age=cols[0]
Pclass=cols[1]
if pd.isnull(Age):
if Pclass==1:
return 37
elif Pclass==2:
return 29
else:
return 24
else:
return Age
train['Age']=train[['Age', 'Pclass']].apply(impute_age, axis=1)
sns.heatmap(train.isnull(), yticklabels=False, cbar=False, cmap='viridis')
train.drop('Cabin', axis=1, inplace=True)
train.head()
sns.heatmap(train.isnull(), yticklabels=False, cbar=False, cmap='viridis')
train.dropna(inplace=True)
train.info()
pd.get_dummies(train['Embarked'], drop_first=True).head()
sex=pd.get_dummies(train['Sex'], drop_first=True)
embark=pd.get_dummies(train['Embarked'], drop_first=True)
train.drop(['sex', 'Embarked', 'Name', 'Ticket'], axis=1, implace=True)
train.haed()
train=pd.concat([train, sex, embark], axis=1)
train.head()
train.drop('Survived', axis=1).head()
train['Survived'].head()
from sklearn.model_selection import train_test_split
x_train, x_test, y_train, y_test=train_test_split(train.drop('Survived', axis=1), train['Survived'], test_size=0.30, random_state=101)
from sklearn.linear_model import LogisticRegression
logmodel=LogisticRegression()
logmodel.fit(x_train, y_train)
Predictions=logmodel.predict(x_test)
from sklearn.metrics import confusion_matrix
accuracy=confusion_matrix(y_test, predictions)
accuracy
from sklearn.metrics import accuracy_score
accuracy=accuracy_score(y_test, predictions)
accuracy
predictions
|
993,871 | d8769ba2d3ea8fd11c3fe822d9e87ad40be8f2a2 | import numpy as np
import pandas as pd
from sklearn.cluster import KMeans
from sklearn import metrics
import matplotlib.pyplot as plt
'''K-means算法在手写体数字图像数据上的使用示例'''
'''
# 读取训练集和测试集
digits_train = pd.read_csv('optdigits.tra', header=None)
digits_test = pd.read_csv('optdigits.tes', header=None)
# 分离出64维度像素特征和1维度数字目标
x_train = digits_train[np.arange(64)]
y_train = digits_train[64]
x_test = digits_test[np.arange(64)]
y_test = digits_test[64]
# 初始化KMeans模型，并设置聚类中心数量为10
kmeans = KMeans(n_clusters=10)
kmeans.fit(x_train)
# 逐条判断每个测试图像所属的聚类中心
y_pred = kmeans.predict(x_test)
# 因为该图片数据自身带有正确的类别信息，使用ARI进行K-means聚类性能评估
print('The ARI of KMeans is', metrics.adjusted_rand_score(y_test, y_pred))
'''
'''利用轮廓系数评价不同类簇数量的K-means聚类实例'''
from sklearn.metrics import silhouette_score
plt.figure(1)
plt.subplot(3, 2, 1)
x1 = np.array([1, 2, 3, 1, 5, 6, 5, 5, 6, 7, 8, 9, 7, 9])
x2 = np.array([1, 3, 2, 2, 8, 6, 7, 6, 7, 1, 2, 1, 1, 3])
x = np.array(list(zip(x1, x2))).reshape(len(x1), 2)
# 在一号子图做出原始信号点阵分布
plt.xlim([0, 10])
plt.ylim([0, 10])
plt.title('Instances')
plt.scatter(x1, x2)
colors = ['b', 'g', 'r', 'c', 'm', 'y', 'k', 'b']
markers = ['o', 's', 'D', 'v', '^', 'p', '*', '+']
clusters = [2, 3, 4, 5, 8]
subplot_counters = 1
sc_scores = []
for t in clusters:
subplot_counters += 1
plt.subplot(3, 2, subplot_counters)
kmeans_model = KMeans(n_clusters=t).fit(x)
print(kmeans_model.labels_)
for i, l in enumerate(kmeans_model.labels_):
plt.plot(x1[i], x2[i], color=colors[l], marker=markers[l], ls='None')
plt.xlim([0, 10])
plt.ylim([0, 10])
sc_score = silhouette_score(x, kmeans_model.labels_,metric='euclidean')
sc_scores.append(sc_score)
plt.title('K=%s, silhouette coefficient=%0.03f' % (t, sc_score))
'''“肘部”观察法示例'''
from scipy.spatial.distance import cdist
cluster1 = np.random.uniform(0.5, 1.5, (2, 10))
cluster2 = np.random.uniform(5.5, 6.5, (2, 10))
cluster3 = np.random.uniform(3.0, 4.0, (2, 10))
# 绘制30个数据样本的分布图像，30行2列
x = np.hstack((cluster1, cluster2, cluster3)).T
plt.figure(2)
plt.subplot(1, 2, 1)
plt.title('Instances')
plt.scatter(x[:, 0], x[:, 1])
plt.xlabel('x1')
plt.ylabel('x2')
# 测试9种不同聚类中心数量下，每组情况的聚类质量
K = range(1, 10)
meandistortions = []
for k in K:
kmeans = KMeans(n_clusters=k)
kmeans.fit(x)
meandistortions.append(sum(np.min(cdist(x, kmeans.cluster_centers_,
'euclidean'), axis=1))/x.shape[0])
plt.subplot(1, 2, 2)
plt.plot(K, meandistortions, 'bx-')
plt.xlabel('k')
plt.ylabel('Average Dispersion')
plt.title('Selecting k with the Elbow Method')
plt.show()
|
993,872 | ceb12818c79f4700b4f7deee6158412e1fc73b85 | #! coding:utf-8
# tools.py
import unittest
from operator import attrgetter
from copy import deepcopy
import random
import numpy as np
class Individuale(object):
def __init__(self, pop):
self._x = np.asarray(pop)
self._fitness_values = None
self._fitness_valid = False
@property
def x(self):
return self._x
@x.setter
def x(self, v):
self._x = np.asarray(v)
@x.deleter
def x(self):
self._x = None
@property
def fitness_values(self):
return self._fitness_values
@fitness_values.setter
def fitness_values(self, v):
self._fitness_values = v
self._fitness_valid = True
@fitness_values.deleter
def fitness_values(self):
self._fitness_values = None
self._fitness_valid = False
@property
def fitness_valid(self):
return self._fitness_valid
@fitness_valid.setter
def fitness_valid(self, v):
v = bool(v)
self._fitness_valid = v
@fitness_valid.deleter
def fitness_valid(self):
self._fitness_valid = False
def __str__(self):
return "{} {} {}".format(self.x, self.fitness_values, self.fitness_valid)
def __repr__(self):
return "x:{}, fittness.values:{} fitness_valid{}".format(self.x, self.fitness_values, self.fitness_valid)
def converter(min=None, max=None, n=None, type=None):
"""実パラメータ空間の値を，指定された整数範囲に射影する
:min, max: パラメータ空間の最小値，最大値
:n: 離散化階数
:type: 射影先の型. int:[0, n-1], float:[0.0, 1.0], bool:[0,1]
"""
if type == bool:
return int, bool, [0, 1]
else:
x1, x2 = min, max
if type == int:
y1, y2 = 0, n - 1
init = range(n)
elif type == float:
y1, y2 = 0.0, 1.0
init = np.linspace(0, 1, n).tolist()
a = float(y1 - y2) / float(x1 - x2)
b = y1 - float(y1 - y2) / float(x1 - x2) * x1
def encoder(x):
if not min <= x <= max:
raise Warning("Input Value is Cliped => {:.2f}<={:.2f}<={:.2f}".format(x1, x, x2))
pass
return a * x + b
def decoder(y):
if not y1 <= y <= y2:
raise Warning("Input Value is Cliped => {:.2f}<={:.2f}<={:.2f}".format(y1, y, y2))
pass
return (y - b) / a
return encoder, decoder, init
def decoder(decs):
"""遺伝子の実数型から，パラメータ空間の型へ戻すためのデコーダー
:decs: <list> tools.convrterから取得したdecoderのリスト
"""
assert isinstance(decs, list)
def closure(ys):
"""
格遺伝子の値をデコードする
:ys: <list> 遺伝子の実数値.
"""
ret = []
for dec, y in zip(decs, ys):
ret.append(dec(y))
return ret
return closure
def encoder(encd):
"""未実装"""
assert isinstance(encd, list)
def closure(xlist):
"""
格遺伝子の値をデコードする
:xlist: <list> 遺伝子の実数値.
"""
ret = []
for enc, x in zip(encd, xlist):
ret.append(enc(x))
return ret
return closure
def population(params, n=100):
"""
p1 = tools.paramgen(
name="HightPower.f1", min=1000, max=20000, delta=1000, type=int)
"""
pops = []
for i in range(n):
pop = []
for param in params:
pop.append(np.random.choice(param))
individuale = Individuale(pop)
pops.append(individuale)
# print("No.{} : {} : {}".format(i, individuale, individuale.x))
return pops
def param_generater(**p):
"""initial population value generater"""
if p["type"] == int:
return np.arange(p["min"], p["max"] + p["delta"], p["delta"], np.int)
elif p["type"] == float:
return np.arange(p["min"], p["max"] + p["delta"], p["delta"], np.float)
elif p["type"] == bool:
return np.array([0, 1])
else:
raise TypeError
def evalute(pop):
# return pop.x.sum()
return sum(pop.x)
def select(individuals, n):
"""Individuales Selector
same selBest()
"""
# return selBest(individuals, n)
return individuals[:n]
def selBest(individuals, n):
"""Individuales Selector"""
return sorted(individuals, key=attrgetter("fitness_values"), reverse=True)[:n]
def selTournamentDCD(individuals, k):
if k % 4 != 0:
raise ValueError
def tourn1d(ind1, ind2):
if ind1.fitness_values > ind1.fitness_values:
return ind1
elif ind1.fitness_values < ind1.fitness_values:
return ind2
# same value
if random.random() <= 0.5:
return ind1
else:
return ind2
individuals_1 = random.sample(individuals, len(individuals))
individuals_2 = random.sample(individuals, len(individuals))
chosen = []
for i in xrange(0, k, 4):
chosen.append(tourn1d(individuals_1[i], individuals_1[i + 1]))
chosen.append(tourn1d(individuals_1[i + 2], individuals_1[i + 3]))
chosen.append(tourn1d(individuals_2[i], individuals_2[i + 1]))
chosen.append(tourn1d(individuals_2[i + 2], individuals_2[i + 3]))
chosen = map(clone, chosen)
return chosen
def mate(ind1, ind2):
raise StandardError
def cxOnePoint(ind1, ind2):
"""Executes a one point crossover on the input :term:`sequence` individuals.
The two individuals are modified in place. The resulting individuals will
respectively have the length of the other.
:param ind1: The first individual participating in the crossover.
:param ind2: The second individual participating in the crossover.
:returns: A tuple of two individuals.
This function uses the :func:`~random.randint` function from the
python base :mod:`random` module.
"""
size = min(len(ind1.x), len(ind2.x))
cxpoint = random.randint(1, size - 1)
ind1.x[cxpoint:], ind2.x[cxpoint:] = ind2.x[cxpoint:], ind1.x[cxpoint:]
return ind1, ind2
def mutate(ind1):
"""mutation base method"""
raise StandardError
def mutUniformInt(individual, low, up, idps):
"""Mutate an individual by replacing attributes, with probability *indpb*,
by a integer uniformly drawn between *low* and *up* inclusively.
:param individual: :term:`Sequence <sequence>` individual to be mutated.
:param low: The lower bound or a :term:`python:sequence` of
of lower bounds of the range from wich to draw the new
integer.
:param up: The upper bound or a :term:`python:sequence` of
of upper bounds of the range from wich to draw the new
integer.
:param indpb: Independent probability for each attribute to be mutated.
:returns: A tuple of one individual.
"""
size = len(individual.x)
for i, xl, xu in zip(xrange(size), low, up):
individual.x[i] = random.randint(xl, xu)
return individual,
def clone(individuals):
"""object copy"""
return deepcopy(individuals)
class TestTools(unittest.TestCase):
def setUp(self):
pass
def test_Individuales(self):
pop = [1, 2, 3, 4, 5]
ind = Individuale(pop)
self.assertEqual(ind.x.all(), np.asarray(pop).all())
self.assertIsNone(ind.fitness_values)
self.assertFalse(ind.fitness_valid)
ind.fitness_values, ind.fitness_valid = 999, True
self.assertIsNotNone(ind.fitness_values)
self.assertTrue(ind.fitness_valid)
del ind.fitness_values, ind.fitness_valid
self.assertIsNone(ind.fitness_values)
self.assertFalse(ind.fitness_valid)
def test_population(self):
p1_enc, p1_dec, p1_init = converter(min=100, max=20000, n=10, type=int) # 0-9
self.assertEqual(p1_enc(100), 0)
self.assertAlmostEqual(p1_enc(20000), 10 - 1)
self.assertEqual(len(p1_init), 10)
p2_enc, p2_dec, p2_init = converter(min=0.1, max=1, n=10, type=int) # 0-9
self.assertEqual(p2_enc(0.1), 0)
self.assertAlmostEqual(p2_enc(1), 10 - 1)
self.assertEqual(len(p2_init), 10)
p3_enc, p3_dec, p3_init = converter(min=10000, max=200000, n=10, type=float) # 0-9
self.assertEqual(p3_enc(10000), 0)
self.assertAlmostEqual(p3_enc(200000), 1)
self.assertEqual(len(p3_init), 10)
p4_enc, p4_dec, p4_init = converter(type=bool) # 0-9
self.assertEqual(p4_enc(True), 1)
self.assertEqual(p4_enc(False), 0)
self.assertEqual(p4_enc(1), True)
self.assertEqual(p4_enc(0), False)
self.assertEqual(len(p4_init), 2)
# p1_enc(100) # 0
# p1_enc(1000) # 9
# p1_dec(0) # 100
# p1_dec(1) # 1000
# map([p1_enc, p2_enc, p3_enc, p4_enc], ind.x)
NPOP = 100
# population init
pops = population([p1_init, p2_init, p3_init, p4_init], n=NPOP)
self.assertIsInstance(pops, list)
self.assertEqual(len(pops), NPOP)
self.assertIsInstance(pops[0], Individuale)
class TestTools2(unittest.TestCase):
def setUp(self):
p1_enc, p1_dec, p1_init = converter(min=100, max=20000, n=10, type=int) # 0-9
p2_enc, p2_dec, p2_init = converter(min=0.1, max=1, n=10, type=int) # 0-9
p3_enc, p3_dec, p3_init = converter(min=10000, max=200000, n=10, type=int) # 0-9
p4_enc, p4_dec, p4_init = converter(type=bool) # 0-9
NPOP = 100
self.pops = population([p1_init, p2_init, p3_init, p4_init], n=NPOP)
self.low = [0, 0, 0, 0]
self.up = [9, 9, 9, 1]
def test_select(self):
pops = self.pops
# selection : selBest()
selected_pops = selBest(pops, 11)
self.assertEqual(len(selected_pops), 11)
self.assertIsInstance(selected_pops[0], Individuale)
self.assertTrue(selected_pops[0].fitness_values >= selected_pops[1].fitness_values)
self.assertTrue(selected_pops[0].fitness_values >= selected_pops[3].fitness_values)
self.assertTrue(selected_pops[0].fitness_values >= selected_pops[5].fitness_values)
# selection : select()
selected_pops = select(pops, 11)
self.assertEqual(len(selected_pops), 11)
self.assertIsInstance(selected_pops[0], Individuale)
def test_selTournamentDCD(self):
pops = self.pops
# selection : selTournamentDCD()
selected_pops = selTournamentDCD(pops, 4 * 25)
self.assertEqual(len(selected_pops), 4 * 25)
self.assertIsInstance(selected_pops[0], Individuale)
with self.assertRaises(ValueError):
selTournamentDCD(pops, 5)
selTournamentDCD(pops, 7)
selTournamentDCD(pops, 15)
selTournamentDCD(pops, 17)
def test_evalution(self):
pops = self.pops
# evalution : Reset
for p in pops:
del p.fitness_values
del p.fitness_valid
# select
selected_pops = selTournamentDCD(pops, 16)
# evalution
for p in pops:
p.fitness_values = evalute(p)
p.fitness_valid = True
# select
selected_pops = selTournamentDCD(pops, 16)
self.assertEqual(len(selected_pops), 16)
self.assertIsInstance(selected_pops[0], Individuale)
def test_clone(self):
pops = self.pops
# copy
offsprint = selTournamentDCD(pops, len(pops))
offsprint_cp = clone(offsprint)
# offsprint_cp = list(map(clone, offsprint))
for offsp, offsp_cp in zip(offsprint, offsprint_cp):
self.assertEqual(offsp.fitness_values, offsp_cp.fitness_values)
self.assertEqual(offsp.fitness_valid, offsp_cp.fitness_valid)
self.assertNotEqual(id(offsp), id(offsp_cp))
self.assertNotEqual(offsp, offsp_cp)
offsp_cp.fitness_values = 10
offsp.fitness_values = 1000000000
self.assertNotEqual(offsp.fitness_values, offsp_cp.fitness_values)
self.assertEqual(offsp.fitness_valid, offsp_cp.fitness_valid)
self.assertNotEqual(id(offsp), id(offsp_cp))
self.assertNotEqual(offsp, offsp_cp)
def test_mate(self):
pops = self.pops
mate = cxOnePoint
offspring = selTournamentDCD(pops, len(pops))
offspring = list(map(clone, offspring))
for ind1, ind2 in zip(offspring[::2], offspring[1::2]):
mate(ind1, ind2)
self.assertIsInstance(ind1, Individuale)
self.assertIsInstance(ind2, Individuale)
del ind1.fitness_values, ind1.fitness_valid
del ind2.fitness_values, ind2.fitness_valid
for ind in offspring:
self.assertIsNone(ind.fitness_values)
self.assertFalse(ind.fitness_valid)
def test_mutate(self):
pops = self.pops
low = self.low
up = self.up
indpb = 0.02
offspring = selTournamentDCD(pops, len(pops))
offspring = list(map(clone, offspring))
print("-- mutation start --")
for ind1, ind2 in zip(offspring[::2], offspring[1::2]):
mutUniformInt(ind1, low, up, indpb)
mutUniformInt(ind2, low, up, indpb)
del ind1.fitness_values, ind1.fitness_valid
del ind2.fitness_values, ind2.fitness_valid
for v, l, u in zip(ind1.x, low, up):
self.assertTrue(l <= v <= u)
for v, l, u in zip(ind2.x, low, up):
self.assertTrue(l <= v <= u)
for ind in offspring:
self.assertIsNone(ind.fitness_values)
self.assertFalse(ind.fitness_valid)
class TestTools3(unittest.TestCase):
def setUp(self):
# Init Individual Value
PMAX = 32
p1 = converter(min=1000, max=10000, n=PMAX, type=int) # 0-9
p2 = converter(min=10, max=20, n=PMAX, type=float) # 0-9
p3 = converter(type=bool) # 0-9
convs = [p1, p2, p3]
init_params, decs, encs = list(), list(), list()
for enc, dec, init in convs:
encs.append(enc)
decs.append(dec)
init_params.append(init)
self.encs = encs
self.decs = decs
def test_encoder(self):
"""パラメータ値を離散値へエンコード"""
_encoder = encoder(self.encs)
self.assertAlmostEqual(_encoder([1000, 20, 0]), [0, 1.0, False])
self.assertAlmostEqual(_encoder([10000, 20, 0]), [32 - 1, 1.0, False])
self.assertAlmostEqual(_encoder([1000, 10, 0]), [0, 0.0, False])
self.assertAlmostEqual(_encoder([1000, 20, 0]), [0, 1.0, False])
self.assertAlmostEqual(_encoder([1000, 20, 1]), [0, 1.0, True])
with self.assertRaises(Warning):
y1 = [999, 10, 9]
ans1 = _encoder(y1)
def test_decoder(self):
"""離散値をパラメータ値へデコード"""
_decoder = decoder(self.decs)
self.assertAlmostEqual(_decoder([32 - 1, 1.0, 1]), [10000, 20, True])
# self.assertAlmostEqual(_decoder([0, 1.0, 1]), [1000., 20., True])
with self.assertRaises(Warning):
self.assertEqual(_decoder([32, 2, 1]), [10000, 20, True])
if __name__ == '__main__':
unittest.main()
|
993,873 | fe325bc347b4b7b5a024444a89f1029f14970619 | import numpy as np
from gmpy2 import mpz
from .curve import Point
def to_np(msg_cipher):
return np.array([[(u.x, u.y) for u in t] for t in msg_cipher],
dtype=np.int).reshape(-1)
def from_np(np_cipher):
return [
tuple(Point(*(int(d) for d in u)) for u in e)
for e in np_cipher.reshape(-1, 2, 2)
]
|
993,874 | 9918cb47ce73bb1b962a114a636a310d6d6e0938 | from __future__ import absolute_import, division
import numpy as np
import math,sys,os,re
from scipy.sparse.csgraph import connected_components,csgraph_from_dense
from itertools import izip
from Bio import SeqIO
from ghost.util.distance import hamming
import networkx as nx
import collections
__author__ = "CDC/OID/NCHHSTP/DVH bioinformatics team"
#REQUIREMENTS FROM INPUT
#needs each read id to end with "_XXXX" where XXXX is the frequency of that read
#needs file name to end with "XX.fas" where XX is a valid HCV genotype (1b,5a,etc.)
#better legend
#number of dots for each color
#genotype
#min hamming
def parseInput(input1,input2): #get sequences from 2 files
ali=True
seqs={}
input_handle = open(input1)
input_handle2 = open(input2)
f1counter=0
f2counter=0
fbothcounter=0
for record in SeqIO.parse(input_handle, "fasta"): # for FASTQ use "fastq", for fasta "fasta"
if len(record.seq) > 0 and len(record.seq) < 50000:
f1counter+=1
freq=re.findall('_(\d*)$',record.id)
seqs[record.seq]="1_"+str(f1counter)+"_"+freq[0]
input_handle.close()
for record in SeqIO.parse(input_handle2, "fasta"): # for FASTQ use "fastq", for fasta "fasta"
if len(record.seq) > 0 and len(record.seq) < 50000:
if record.seq not in seqs:
f2counter+=1
freq=re.findall('_(\d*)$',record.id)
seqs[record.seq]="2_"+str(f2counter)+"_"+freq[0]
else:
fbothcounter+=1
seqs[record.seq]="x_"+str(fbothcounter)
input_handle2.close()
if ali:
newseqs=align(seqs)
try:
haploSize = len(newseqs.keys()[0])
except IndexError:
sys.exit("Either mafft isn't loaded or both of your fasta files are empty. Exiting")
haploNum = len(newseqs)
return(newseqs,haploSize,haploNum,f1counter,f2counter,fbothcounter)
else:
haploSize = len(seqs.keys()[0])
haploNum = len(seqs)
return(seqs,haploSize,haploNum,f1counter,f2counter,fbothcounter)
def align(seqs):
f=open("tmp.fas","w")
for seq in seqs:
f.write(">"+seqs[seq]+"\n")
f.write(str(seq)+"\n")
f.close()
os.system('mafft --quiet --auto --thread 20 --preservecase tmp.fas > align.fas')
outs={}
input_handle = open('align.fas')
for record in SeqIO.parse(input_handle, "fasta"): # for FASTQ use "fastq", for fasta "fasta"
if len(record.seq) > 0 and len(record.seq) < 50000:
if collections.Counter(record.seq).most_common(1)[0][0]!="-":
outs[record.seq]=record.id
input_handle.close()
# rmstr="rm align.fas tmp.fas"
# os.system(rmstr)
# for item in outs:
# print(item)
# print(outs[item])
return outs
#@profile
def calcOrderedFrequencies(haploNum,haploSize,seqs): #calculate the ordered frequencies
freqCount = np.zeros((haploSize, 5))
productVector = np.zeros((haploSize, 5))
for read in seqs:
for pos in range(haploSize):
if read[pos] == 'A':
freqCount[pos, 0] = freqCount[pos, 0] + 1
elif read[pos] == 'C':
freqCount[pos, 1] = freqCount[pos, 1] + 1
elif read[pos] == 'G':
freqCount[pos, 2] = freqCount[pos, 2] + 1
elif read[pos] == 'T':
freqCount[pos, 3] = freqCount[pos, 3] + 1
elif read[pos] == '-':
freqCount[pos, 4] = freqCount[pos, 4] + 1
freqRel = np.divide(freqCount, haploNum, dtype = float)
for pos in range(haploSize):
for i in range(5):
freqPos = freqRel[pos, i]
if freqPos > 0:
logFreqRel = math.log(freqPos, 2)
productVector[pos, i] = -1*(np.multiply(freqPos, logFreqRel, dtype = float))
return np.sum(productVector, axis = 1)
#@profile
def orderPositions(hVector,seqs,haploSize): #order positions
invH = np.multiply(-1, hVector, dtype = float)
ordH = np.argsort(invH)
# reorder the sequences by entropy
ordSeqs = {}
for i in seqs:
newOne = ''
for p in range(haploSize):
newOne = ''.join([newOne, i[ordH[p]]])
ordSeqs[newOne]=seqs[i]
return ordSeqs
#@profile
def calcDistances(haploNum,haploSize,asdf): #Calculate hamming distances between sequences
ordSeqs=asdf.keys()
# for seq in ordSeqs:
# print(seq,asdf[seq])
colors=[]
compNum = haploSize
compList = range(haploNum)
t = 0
adjMatrix = np.zeros((haploNum, haploNum))
kStepList = []
while compNum > 1:
t = t + 1
# Check each query sequence
for r1 in range(haploNum-1):
haplotype1 = ordSeqs[r1]
for r2 in range(r1+1, haploNum):
if compList[r1] != compList[r2]:
haplotype2 = ordSeqs[r2]
tempDist = 0
for a, b in izip(haplotype1, haplotype2):
if a != b:
tempDist = tempDist + 1
if tempDist > t:
break
if tempDist == t:
adjMatrix[r1][r2] = 1
asdf[ordSeqs[r1]]
kStepList.append([asdf[ordSeqs[r1]], asdf[ordSeqs[r2]], t])
if asdf[ordSeqs[r1]] not in colors:
colors.append(asdf[ordSeqs[r1]])
if asdf[ordSeqs[r2]] not in colors:
colors.append(asdf[ordSeqs[r2]])
# Calculate components
sparseMatrix = csgraph_from_dense(adjMatrix)
connected = connected_components(sparseMatrix, directed=False, connection='weak', return_labels=True)
compNum = connected[0]
compList = connected[1]
# print(compNum)
# print(compList)
# print(t)
# print("=")
if t/haploSize>.42:
if sum(compList)==1:
offender=ordSeqs[np.argmax(compList)]
del asdf[offender]
altwrapper(asdf,colors,haploNum-1,haploSize)
sys.exit()
else:
sys.exit("Your output is going to look really messed up. Exiting")
return kStepList,colors
def kstepToNX(kStepList,colors): #go from a matrix to a weighted networkx graph
labels=False
g=nx.Graph()
for node in colors:
splitnode=node.split("_")
if splitnode[0]=="2":
g.add_node(node,color='blue',shape='point',width=.1)
if splitnode[0]=="1":
g.add_node(node,color='red',shape='point',width=.1)
if splitnode[0]=="x":
g.add_node(node,color='green',shape='point',width=.1)
for row in kStepList:
if labels:
g.add_edge(row[0],row[1],len=row[2],color='gray',label=str(row[2]))
else:
g.add_edge(row[0],row[1],len=row[2],color='gray')
# g.add_edge(row,col,len=matrix[row,col]*numRows,arrowsize=".01",color='gray',alpha='.5')
return g
def splitseqs(seqs): #get sequences from 2 files
firstFile=[]
secondFile=[]
# for key in seqs:
# print(seqs[key])
for key in seqs:
if seqs[key].startswith("1"):
firstFile.append(key)
elif seqs[key].startswith("2"):
secondFile.append(key)
elif seqs[key].startswith("x"):
firstFile.append(key)
secondFile.append(key)
return (firstFile,secondFile)
def calcDistanceMatrix(seqs1,seqs2): #calculate distance matrix from the 1-step list
hdist=hamming(seqs1,seqs2,ignore_gaps=False)
l=len(seqs1)
w=len(seqs2)
arr=np.zeros([l,w])
for id in range(len(hdist)):
item=hdist[id]
arr[:,id]=item[:,0]
return arr
def altwrapper(seqs,colors,haploNum,haploSize):
# print("-------------altwrapper-------------")
# print("-------------altwrapper-------------")
# print("-------------altwrapper-------------")
aligned=align(seqs)
hVector=calcOrderedFrequencies(haploNum,haploSize,seqs)
ordSeqs=orderPositions(hVector,seqs,haploSize)
kStepList,colors=calcDistances(haploNum,haploSize,ordSeqs)
g=kstepToNX(kStepList,colors)
nx.drawing.nx_agraph.write_dot(g,'tmp.dot')
f=open("tmp.dot","r")
try:
g=open(sys.argv[3],"w")
except IndexError:
g=open("tmper.dot","w")
lines=f.readlines()
f.close()
lineco=0
for line in lines:
lineco+=1
if lineco==2:
g.write("\tgraph [outputorder=edgesfirst];\n")
g.write(line)
g.close()
try:
grapit="neato -Tpng "+sys.argv[3]+" -O"
except IndexError:
grapit="neato -Tpng tmper.dot -o tmp.png"
os.system(grapit)
# os.system("eog tmp.png")
if __name__=="__main__":
checkMinHamm=False
gen1=re.findall('_(\d[a-z]).fas',sys.argv[1])[0]
gen2=re.findall('_(\d[a-z]).fas',sys.argv[2])[0]
if gen1!=gen2:
print("WARNING: your two files are not the same genotype! Your output will likely be bad")
(seqs,haploSize,haploNum,f1counter,f2counter,both)=parseInput(sys.argv[1],sys.argv[2])
(seqs1,seqs2)=splitseqs(seqs)
array=calcDistanceMatrix(seqs1,seqs2)
minHamm=np.amin(array)/len(seqs1[0])
if checkMinHamm==True:
if minHamm>.42:
sys.exit("Your files do not appear to be the same genotype - manual curation likely required. Exiting")
hVector=calcOrderedFrequencies(haploNum,haploSize,seqs)
ordSeqs=orderPositions(hVector,seqs,haploSize)
kStepList,colors=calcDistances(haploNum,haploSize,ordSeqs)
g=kstepToNX(kStepList,colors)
nx.drawing.nx_agraph.write_dot(g,'tmp.dot')
f=open("tmp.dot","r")
try:
g=open(sys.argv[3],"w")
except IndexError:
g=open("tmper.dot","w")
lines=f.readlines()
f.close()
lineco=0
l=len(lines)
for line in lines:
lineco+=1
if lineco==2:
g.write("\tgraph [outputorder=edgesfirst];\n")
if lineco!=l:
g.write(line)
else:
g.write("{ rank = sink;\n")
g.write(" Legend [shape=none, margin=0, label=<\n")
g.write(""" <TABLE BORDER="0" CELLBORDER="1" CELLSPACING="0" CELLPADDING="4">\n""")
g.write(" <TR>\n")
g.write(""" <TD COLSPAN="2"><B>Genotype = %s, Min H-Dist = %f</B></TD>\n""" %(gen1,minHamm))
g.write(" </TR>\n <TR>\n")
g.write(" <TD>%s</TD>\n" %sys.argv[1])
g.write(""" <TD><FONT COLOR="red">%i</FONT></TD>\n"""%f1counter)
g.write(" </TR>\n <TR>\n")
g.write(" <TD>%s</TD>\n" %sys.argv[2])
g.write(""" <TD><FONT COLOR="blue">%i</FONT></TD>\n"""%f2counter)
g.write(" </TR>\n <TR>\n")
g.write(" <TD>Both Files</TD> ")
g.write(""" <TD><FONT COLOR="green">%i</FONT></TD>\n"""%both)
g.write(" </TR>\n </TABLE>\n >];\n }\n} ")
g.close()
try:
grapit="neato -Tpng "+sys.argv[3]+" -O"
except IndexError:
grapit="neato -Tpng tmper.dot -o tmp.png"
os.system(grapit)
# os.system("eog tmp.png") |
993,875 | f9ba92e5b5f8baaa2c551a7e03ecc0ded4277c45 | class Node(object):
"""
Node object which is being used to test the functions.
Delete before submitting.
Args:
data (:anything): Data which is contained in the LinkedList node.
next_node (:obj type Node): Pointer to the subsequent node in the linkedList.
Attributes:
data (:anything): Data which is contained in the LinkedList node.
next_node (:obj type Node): Pointer to the subsequent node in the linkedList.
"""
def __init__(self, data = None, next_node = None):
self.data = data
self.next = next_node
def __str__(self):
return str(self.data)
def set_next(self, new_next):
self.next = new_next
def has_cycle(head):
"""
Detects whether there is a cycle in a linked list.
Args:
head (:obj type Node): A pointer to the head Node of a linkedlist. Contains 'None' if the list is empty.
A Node is defined as:
class Node(object):
def __init__(self, data = None, next_node = None):
self.data = data
self.next = next_node
Returns:
cycle (:boolean): Contains a True if the linkedlist contains a cycle and a False otherwise.
"""
# We specify that an empty list won't have a cycle.
if head is None:
return False
assert isinstance(head, Node)
pass
|
993,876 | b7623deac8e147d827fd847ae810b1d93d6a967b | k=int(input())
def s(n):
m=str(n)
l=list(m)
l=map(int,l)
return n/sum(l)
p=1
q=0
for i in range(k):
print(p)
if s(p+10**q)>s(p+10**(q+1)):
q+=1
p+=10**q |
993,877 | e59bb0714649209b9d26b6af39dbd7aead3fea05 | from datetime import datetime
from flask import abort, current_app as app, request, g
from flask_restplus import Resource
from werkzeug.exceptions import BadRequest
from ...database.models import Activity
from ..restplus import api
from ..serializers import activity_request, activity_response
import logging
log = logging.getLogger(__name__)
ns = api.namespace('activities', description='Groups related items')
@ns.route('/')
class ActivityCollection(Resource):
@api.expect(activity_request)
@api.marshal_with(activity_response)
@api.response(201, 'Activity created')
def post(self):
"""
Creates a new activity
"""
data = request.json
now = datetime.utcnow().isoformat()
data['created_time'] = now
# Our type is already registered within the DB, so generate a
# model object that looks like what we'll be interacting with
try:
activity = Activity(data)
except KeyError:
raise BadRequest("payload validation failed: {}".format(data))
activity.save()
log.debug("Wrote activity: " + str(activity._to_dict()))
return activity._to_dict(), 201
@api.marshal_list_with(activity_response, envelope='activities')
@api.response(200, 'Ok')
def get(self):
"""
Retrieve a list of activities
"""
return [a._to_dict() for a in Activity().get_all()]
@ns.route('/<string:id>')
class ActivityResource(Resource):
@api.marshal_with(activity_response)
@api.response(200, 'Ok')
@api.response(404, 'Activity not found')
def get(self, id):
"""
Retrieve a single activity
"""
activity = Activity().get(id)
if not activity:
abort(404, "Activity not found")
return activity
@api.expect(activity_request)
@api.marshal_with(activity_response)
@api.response(200, 'Ok')
@api.response(404, 'Activity not found')
def put(self, id):
"""
Update an activity, e.g. to add a new item
"""
activity = Activity().get(id)
if not activity:
abort(404, "Activity not found")
return activity._update(request.json)
|
993,878 | 0b609ce1c96ba138a24e12f976b889bd77c0eee7 | try:
from StringIO import StringIO
except ImportError:
from io import StringIO
import unittest
import sys
import time
from serial import Serial
PORT = '/dev/ttyACM0'
class TestSettings(unittest.TestCase):
def setUp(self):
self.held, sys.stdout = sys.stdout, StringIO()
def test_00_echo(self):
ser=Serial(PORT, 2000000, timeout=3)
time.sleep(0.1)
ser.write('AT\r')
time.sleep(0.1)
response = ser.readline()
ser.close()
self.assertEqual(response, 'AT\r\n')
def test_01_version(self):
ser=Serial(PORT, 2000000, timeout=3)
time.sleep(0.1)
ser.write('ATV\r')
time.sleep(0.1)
response = ser.readline()
ser.close()
self.assertEqual(response, 'ChaosDuino v0.1.3 for PCB rev3 running...\r\n')
def test_02_LED(self):
ser=Serial(PORT, 2000000, timeout=3)
time.sleep(0.1)
ser.write('ATLR0\r')
ser.write('ATLG0\r')
ser.write('ATLB0\r')
j = 0
for i in range(50):
if j & 1 == 1:
ser.write('ATLB1\r')
if j & 1 == 0:
ser.write('ATLB0\r')
if j & 2 == 0:
ser.write('ATLG1\r')
if j % 4 != 0:
ser.write('ATLG0\r')
if j % 8 == 0:
ser.write('ATLR1\r')
if j % 8 != 0:
ser.write('ATLR0\r')
j+=1
time.sleep(0.1)
ser.write('ATLR0\r')
ser.write('ATLG0\r')
ser.write('ATLB0\r')
ser.close()
def test_03_ATE(self):
ser=Serial(PORT, 2000000, timeout=3)
time.sleep(0.1)
ser.write('ATE?\r')
time.sleep(0.1)
response = ser.readline()
self.assertEqual(response, 'ATE0\r\n')
ser.write('ATE1\r')
time.sleep(0.1)
ser.write('ATE?\r')
time.sleep(0.1)
response = ser.readline()
self.assertEqual(response, 'ATE1\r\n')
ser.close()
def test_04_ATP(self):
ser=Serial(PORT, 2000000, timeout=3)
time.sleep(0.1)
ser.write('ATP?\r')
time.sleep(0.1)
response = ser.readline()
self.assertEqual(response, 'ATP0\r\n')
for i in range(3):
ser.write('ATP{}\r'.format(i))
time.sleep(0.1)
ser.write('ATP?\r')
time.sleep(0.1)
response = ser.readline()
self.assertEqual(response, 'ATP{}\r\n'.format(i))
ser.write('ATP0\r')
time.sleep(0.1)
ser.write('ATP?\r')
time.sleep(0.1)
response = ser.readline()
self.assertEqual(response, 'ATP0\r\n')
ser.close()
def test_05_ATM(self):
ser=Serial(PORT, 2000000, timeout=3)
time.sleep(0.1)
ser.write('ATM?\r')
time.sleep(0.1)
response = ser.readline()
self.assertEqual(response, 'ATM1\r\n')
for i in range(8):
ser.write('ATM{}\r'.format(i))
time.sleep(0.1)
ser.write('ATM?\r')
time.sleep(0.1)
response = ser.readline()
self.assertEqual(response, 'ATM{}\r\n'.format(i))
ser.write('ATM1\r')
time.sleep(0.1)
ser.write('ATM?\r')
time.sleep(0.1)
response = ser.readline()
self.assertEqual(response, 'ATM1\r\n')
ser.close()
def test_06_OK(self):
ser=Serial(PORT, 2000000, timeout=3)
time.sleep(0.1)
ser.write('ATOK?\r')
time.sleep(0.1)
response = ser.readline()
self.assertEqual(response, 'ATOK0\r\n')
ser.close()
def test_07_POOL(self):
ser=Serial(PORT, 2000000, timeout=3)
time.sleep(0.1)
ser.write('ATPOOL?\r')
time.sleep(0.1)
response = ser.readline()
self.assertEqual(response, '10000\r\n')
ser.close()
def test_08_BIP39(self):
ser=Serial(PORT, 2000000, timeout=3)
time.sleep(0.1)
ser.write('BIP39W?\r')
time.sleep(0.1)
response = ser.readline()
self.assertEqual(response, 'BIP39W24\r\n')
for i in [15, 18, 21, 24]:
ser.write('BIP39W{}\r'.format(i))
time.sleep(0.1)
ser.write('BIP39W?\r')
time.sleep(0.1)
response = ser.readline()
self.assertEqual(response, 'BIP39W{}\r\n'.format(i))
ser.close()
# class TestData(unittest.TestCase):
# def setUp(self):
# self.held, sys.stdout = sys.stdout, StringIO()
# def test_dual_classes(self):
# jam1 = Jam({'issue': {'rmamba': 0.5}})
# jam2 = Jam({'issue': {'ledi_mambix': 1.5}})
# self.assertEqual(jam1.jam, {'issue': {'rmamba': 0.5}})
# self.assertFalse(jam1.modified, 'Should not be modified!!!')
# self.assertEqual(jam2.jam, {'issue': {'ledi_mambix': 1.5}})
# self.assertFalse(jam2.modified, 'Should NOT be modified!!!')
# class TestEntropy(unittest.TestCase):
# def setUp(self):
# self.held, sys.stdout = sys.stdout, StringIO()
# def test_dual_classes(self):
# jam1 = Jam({'issue': {'rmamba': 0.5}})
# jam2 = Jam({'issue': {'ledi_mambix': 1.5}})
# self.assertEqual(jam1.jam, {'issue': {'rmamba': 0.5}})
# self.assertFalse(jam1.modified, 'Should not be modified!!!')
# self.assertEqual(jam2.jam, {'issue': {'ledi_mambix': 1.5}})
# self.assertFalse(jam2.modified, 'Should NOT be modified!!!')
|
993,879 | 156aabf8a03ee0197c6094733614cae345cd3070 | # -*- coding: utf-8 -*-
from django.shortcuts import render
from django.contrib.auth.decorators import login_required
from django.contrib.auth import logout
from django.contrib.auth.models import User, Group
from django.http import HttpResponseRedirect, HttpResponseForbidden, HttpResponse
import datetime
from logapp.models import *
# Remember to change these when the experiment terms change in the
# Prolog code.
term2string = {
1 : "males",
2 : "females",
3 : "adults",
4 : "mothers",
5 : "fathers",
6 : "parents",
7 : "students",
8 : "teachers",
9 : "doctors",
}
# Remember to change these when the experiment terms change in the
# Prolog code (test2015b)
artificialterm2string = {
1 : "bloogs",
2 : "zeefs",
3 : "kloojarbs",
4 : "flarkjerbs",
5 : "greegarfs",
6 : "habzoops",
7 : "plarfeebs",
8 : "meefjarbs",
9 : "troofklejoobs",
}
def root_view(request):
return HttpResponseRedirect('/syllog/index.html')
def getValue(request, key, bValid):
if not bValid:
return (None, False)
else:
if key in request.GET:
value = request.GET[key]
return (value, True)
else:
return (None, False)
def log(request):
if request.method == "GET":
bValid = True
(usernumber_str, bValid) = getValue(request, u"uid", bValid)
(timelapse_str, bValid) = getValue(request, u"timelapse", bValid)
(syllnumber_str, bValid) = getValue(request, u"syllnumber", bValid)
(figure_str, bValid) = getValue(request, u"figure", bValid)
(correctness_str, bValid) = getValue(request, u"corr", bValid)
(term_S_str, bValid) = getValue(request, u"S", bValid)
(term_M_str, bValid) = getValue(request, u"M", bValid)
(term_P_str, bValid) = getValue(request, u"P", bValid)
(conclusion_truth_value_str, bValid) = getValue(request, u"conclusion", bValid)
if not bValid:
return render(request, 'log.html')
else:
if correctness_str == "true":
bCorrect = True
else:
bCorrect = False
if conclusion_truth_value_str == "1":
bConclusionTruthValue = True
else:
bConclusionTruthValue = False
newobj = Answer(usernumber=int(usernumber_str),
timelapse=int(timelapse_str),
syllnumber=int(syllnumber_str),
figure=int(figure_str),
correctness=bCorrect)
newobj.save()
newobj2 = Answer2(answer=newobj,
conclusion_truth_value=bConclusionTruthValue,
term_S=int(term_S_str),
term_M=int(term_M_str),
term_P=int(term_P_str))
newobj2.save()
return render(request, 'log.html')
else:
return render(request, 'log.html')
def log3(request):
"""This is the endpoint for most version of Syllog."""
if request.method == "GET":
bValid = True
(usernumber_str, bValid) = getValue(request, u"uid", bValid)
(timelapse_str, bValid) = getValue(request, u"timelapse", bValid)
(propkind_str, bValid) = getValue(request, u"propkind", bValid)
(figure_str, bValid) = getValue(request, u"figure", bValid)
(correctness_str, bValid) = getValue(request, u"corr", bValid)
(term_S_str, bValid) = getValue(request, u"S", bValid)
(term_M_str, bValid) = getValue(request, u"M", bValid)
(term_P_str, bValid) = getValue(request, u"P", bValid)
(term_Q1_str, bValid) = getValue(request, u"q1", bValid)
(term_Q2_str, bValid) = getValue(request, u"q2", bValid)
(term_Q3_str, bValid) = getValue(request, u"q3", bValid)
(conclusion_truth_value_str, bValid) = getValue(request, u"conclusion", bValid)
(validity_answer, bValid) = getValue(request, u"validity", bValid)
if not bValid:
return render(request, 'log.html')
else:
if correctness_str == "true":
bCorrect = True
else:
bCorrect = False
if conclusion_truth_value_str == "1":
bConclusionTruthValue = True
else:
bConclusionTruthValue = False
newobj3 = Answer3(usernumber=int(usernumber_str),
timelapse=int(timelapse_str),
figure=int(figure_str),
quantor_1=term_Q1_str,
quantor_2=term_Q2_str,
quantor_3=term_Q3_str,
term_S=int(term_S_str),
term_M=int(term_M_str),
term_P=int(term_P_str),
propkind=propkind_str,
student_answer=validity_answer,
correctness=bCorrect,
conclusion_truth_value=bConclusionTruthValue)
newobj3.save()
return render(request, 'log.html')
else:
return render(request, 'log.html')
def log5(request):
"""log5 is a version of log3 without conclusion. It is meant for
the version that has artificial words as terms in the syllogism."""
if request.method == "GET":
bValid = True
(usernumber_str, bValid) = getValue(request, u"uid", bValid)
(timelapse_str, bValid) = getValue(request, u"timelapse", bValid)
(propkind_str, bValid) = getValue(request, u"propkind", bValid)
(figure_str, bValid) = getValue(request, u"figure", bValid)
(correctness_str, bValid) = getValue(request, u"corr", bValid)
(term_S_str, bValid) = getValue(request, u"S", bValid)
(term_M_str, bValid) = getValue(request, u"M", bValid)
(term_P_str, bValid) = getValue(request, u"P", bValid)
(term_Q1_str, bValid) = getValue(request, u"q1", bValid)
(term_Q2_str, bValid) = getValue(request, u"q2", bValid)
(term_Q3_str, bValid) = getValue(request, u"q3", bValid)
(validity_answer, bValid) = getValue(request, u"validity", bValid)
if not bValid:
return render(request, 'log.html')
else:
if correctness_str == "true":
bCorrect = True
else:
bCorrect = False
bConclusionTruthValue = False
newobj5 = Answer5(usernumber=int(usernumber_str),
timelapse=int(timelapse_str),
figure=int(figure_str),
quantor_1=term_Q1_str,
quantor_2=term_Q2_str,
quantor_3=term_Q3_str,
term_S=int(term_S_str),
term_M=int(term_M_str),
term_P=int(term_P_str),
propkind=propkind_str,
student_answer=validity_answer,
correctness=bCorrect,
conclusion_truth_value=bConclusionTruthValue)
newobj5.save()
return render(request, 'log.html')
else:
return render(request, 'log.html')
def log4(request):
"""This is the endpoint for the versions of Syllog that allow a
student first to try to prove the syllogism valid or invalid,
using the rules: Trans, Subst, Contra, Ex, and (optionally) Mut.
The following front-end endpoints were meant for this view:
- proof
- test6
- test8
"""
if request.method == "GET":
bValid = True
(usernumber_str, bValid) = getValue(request, u"uid", bValid)
(timelapse_str, bValid) = getValue(request, u"timelapse", bValid)
(figure_str, bValid) = getValue(request, u"figure", bValid)
(term_Q1_str, bValid) = getValue(request, u"q1", bValid)
(term_Q2_str, bValid) = getValue(request, u"q2", bValid)
(term_Q3_str, bValid) = getValue(request, u"q3", bValid)
(correctness_str, bValid) = getValue(request, u"corr", bValid)
(term_S_str, bValid) = getValue(request, u"S", bValid)
(term_M_str, bValid) = getValue(request, u"M", bValid)
(term_P_str, bValid) = getValue(request, u"P", bValid)
(conclusion_truth_value_str, bValid) = getValue(request, u"conclusion", bValid)
(validity_answer, bValid) = getValue(request, u"validity", bValid)
(rule_clicks_str, bValid) = getValue(request, u"rule_clicks", bValid)
if not bValid:
return render(request, 'log.html')
else:
if correctness_str == "true":
bCorrect = True
else:
bCorrect = False
if conclusion_truth_value_str == "1":
bConclusionTruthValue = True
else:
bConclusionTruthValue = False
try:
rule_clicks = int(rule_clicks_str)
except:
rule_clicks = -1
propkind_str = "verbal"
newobj4 = Answer4(usernumber=int(usernumber_str),
timelapse=int(timelapse_str),
figure=int(figure_str),
quantor_1=term_Q1_str,
quantor_2=term_Q2_str,
quantor_3=term_Q3_str,
term_S=int(term_S_str),
term_M=int(term_M_str),
term_P=int(term_P_str),
propkind=propkind_str,
student_answer=validity_answer,
correctness=bCorrect,
conclusion_truth_value=bConclusionTruthValue,
rule_clicks=rule_clicks)
newobj4.save()
return render(request, 'log.html')
else:
return render(request, 'log.html')
def getCorrectness(result, usernumber, date, figure, syllnumber):
if date not in result:
return (0,0)
if usernumber not in result[date]:
return (0,0)
if figure not in result[date][usernumber]:
return (0,0)
if syllnumber not in result[date][usernumber][figure]:
return (0,0)
correct = 0
wrong = 0
if "correct" in result[date][usernumber][figure][syllnumber]:
correct = result[date][usernumber][figure][syllnumber]["correct"]
if "wrong" in result[date][usernumber][figure][syllnumber]:
wrong = result[date][usernumber][figure][syllnumber]["wrong"]
return (correct, wrong)
def dump(request):
if not request.user.is_authenticated:
return HttpResponseForbidden()
else:
import csv
answer_list = list(Answer.objects.all())
result = {} # date --> { usernumber --> { figure --> {syllnumber --> { "correct" : correct-count, "wrong" : wrong-count } } } }
for answer in answer_list:
date = u"%s" % answer.date
result.setdefault(date, {})
result[date].setdefault(answer.usernumber, {})
result[date][answer.usernumber].setdefault(answer.figure, {})
result[date][answer.usernumber][answer.figure].setdefault(answer.syllnumber, {})
result[date][answer.usernumber][answer.figure][answer.syllnumber].setdefault("correct", 0)
result[date][answer.usernumber][answer.figure][answer.syllnumber].setdefault("wrong", 0)
if answer.correctness:
result[date][answer.usernumber][answer.figure][answer.syllnumber]["correct"] += 1
else:
result[date][answer.usernumber][answer.figure][answer.syllnumber]["wrong"] += 1
# Create the HttpResponse object with the appropriate CSV header.
response = HttpResponse("", content_type='text/csv')
response['Content-Disposition'] = 'attachment; filename=logsyllog-data.csv'
writer = csv.writer(response)
first_row = ['date', 'usernumber']
for figure in [1,2,3,4]:
for syllnumber in [1,2,3,4,5,6,7,8,9,10,11,12]:
first_row.append('F:%d/S:%d:C' % (figure, syllnumber))
first_row.append('F:%d/S:%d:W' % (figure, syllnumber))
writer.writerow(first_row)
for date in sorted(result):
for usernumber in sorted(result[date]):
next_row = []
next_row.append(date)
next_row.append('%d' % usernumber)
for figure in [1,2,3,4]:
for syllnumber in [1,2,3,4,5,6,7,8,9,10,11,12]:
(correct,wrong) = getCorrectness(result, usernumber, date, figure, syllnumber)
next_row.append('%d' % correct)
next_row.append('%d' % wrong)
writer.writerow(next_row)
del next_row
return response
def dump2(request):
if not request.user.is_authenticated:
return HttpResponseForbidden()
else:
import csv
answer_list = Answer2.objects.all().order_by("id")
result = {} # date --> { order --> { mydict } }
order_dict = {} #usernumber --> order-int
order_int = 0
for answer2 in answer_list:
answer = answer2.answer
date = u"%s" % answer.date
usernumber = answer.usernumber
figure = answer.figure
syllnumber = answer.syllnumber
timelapse = answer.timelapse
correctness = answer.correctness
conclusion_truth_value = answer2.conclusion_truth_value
term_S = term2string[answer2.term_S]
term_M = term2string[answer2.term_M]
term_P = term2string[answer2.term_P]
if usernumber in order_dict:
order = order_dict[usernumber]
else:
order = order_int
order_int += 1
order_dict[usernumber] = order
result.setdefault(date, {})
result[date].setdefault(order, [])
result[date][order].append({
"usernumber" : usernumber,
"figure" : figure,
"syllnumber" : syllnumber,
"timelapse" : timelapse,
"correctness" : correctness,
"conclusion_truth_value" : conclusion_truth_value,
"term_S" : term_S,
"term_M" : term_M,
"term_P" : term_P,
})
# Create the HttpResponse object with the appropriate CSV header.
response = HttpResponse("", content_type='text/csv')
response['Content-Disposition'] = 'attachment; filename=logsyllog-data.csv'
writer = csv.writer(response)
first_row = ['date', 'usernumber', 'figure', 'syllnumber', 'term_S', 'term_M', 'term_P', 'correctness', 'conclusion', 'timelapse (s)', 'totaltime (s)', 'lastcorrectinarow', 'answercount']
writer.writerow(first_row)
for date in sorted(result):
for order in sorted(result[date]):
usernumber = 0
last_correct_in_a_row = 0
totaltime = 0
answercount = 0
for mydict in result[date][order]:
answercount += 1
next_row = []
next_row.append(date)
next_row.append('%d' % mydict['usernumber'])
usernumber = mydict['usernumber']
next_row.append('%d' % mydict['figure'])
next_row.append('%d' % mydict['syllnumber'])
next_row.append('%s' % mydict['term_S'])
next_row.append('%s' % mydict['term_M'])
next_row.append('%s' % mydict['term_P'])
next_row.append('%s' % mydict['correctness'])
next_row.append('%s' % mydict['conclusion_truth_value'])
next_row.append('%.3f' % ((1.0 * mydict['timelapse']) / 1000.0))
totaltime += mydict['timelapse']
writer.writerow(next_row)
del next_row
if mydict["correctness"]:
last_correct_in_a_row += 1
else:
last_correct_in_a_row = 0
next_row = [date, '%d' % usernumber, '', '', '', '', '', '', '', '', '%.3f' % ((1.0 * totaltime) / 1000.0), '%d' % last_correct_in_a_row, '%d' % answercount]
writer.writerow(next_row)
del next_row
return response
# Amended from: http://docs.python.org/2/library/datetime.html
class GMT1(datetime.tzinfo):
def utcoffset(self, dt):
return datetime.timedelta(hours=1) + self.dst(dt)
def dst(self, dt):
# DST starts last Sunday in March
d = datetime.datetime(dt.year, 4, 1) # ends last Sunday in October
self.dston = d - datetime.timedelta(days=d.weekday() + 1)
d = datetime.datetime(dt.year, 11, 1)
self.dstoff = d - datetime.timedelta(days=d.weekday() + 1)
if self.dston <= dt.replace(tzinfo=None) < self.dstoff:
return datetime.timedelta(hours=1)
else:
return datetime.timedelta(0)
def tzname(self,dt):
return "GMT +1"
session_start_dict = {
"2014-02-04" : 12,
"2014-02-05" : 12,
"2014-02-06" : 12,
"2014-02-07" : 10,
}
def fake_timezone1(last_time, time, date):
if date not in session_start_dict:
return time
else:
session_start = session_start_dict[date]
hour = int(time[0:2])
if hour < session_start:
hour += 6 # Assume UTC-0500
new_time = "%02d:%s" % (hour, time[3:])
return new_time
else:
return time
def fake_timezone2(last_time, time, date):
if date not in session_start_dict:
return time
elif last_time == None:
return time
else:
hour_minute = time[0:5]
last_hour_minute = last_time[0:5]
if hour_minute < last_hour_minute:
hour = int(time[0:2])
hour += 1
new_time = "%02d:%s" % (hour, time[3:])
return new_time
else:
return time
def dump3(request):
bAggregateSessions = False
return dump3_super(request, bAggregateSessions)
def dump3_aggregatesess(request):
bAggregateSessions = True
return dump3_super(request, bAggregateSessions)
def dump3_super(request, bAggregateSessions):
if not request.user.is_authenticated:
return HttpResponseForbidden()
else:
import csv
answer_list = Answer3.objects.all().order_by("id")
result = {} # date --> { order --> { mydict } }
order_dict = {} #usernumber --> order-int
order_int = 0
mytz = GMT1()
for answer3 in answer_list:
usernumber = answer3.usernumber
# Convert aware datetime into our own timezone
mydate = answer3.date
date = (u"%s" % mydate)[0:10]
time = (u"%s" % mydate)[11:19]
timelapse = answer3.timelapse
figure = answer3.figure
quantor_1 = answer3.quantor_1
quantor_2 = answer3.quantor_2
quantor_3 = answer3.quantor_3
term_S = term2string[answer3.term_S]
term_M = term2string[answer3.term_M]
term_P = term2string[answer3.term_P]
correctness = answer3.correctness
conclusion_truth_value = answer3.conclusion_truth_value
student_answer = answer3.student_answer
propkind = answer3.propkind
if student_answer == "valid":
if correctness:
syllogism_validity = "valid"
else:
syllogism_validity = "invalid"
else:
if correctness:
syllogism_validity = "invalid"
else:
syllogism_validity = "valid"
if usernumber in order_dict:
order = order_dict[usernumber]
else:
order = order_int
order_int += 1
order_dict[usernumber] = order
result.setdefault(date, {})
result[date].setdefault(order, [])
result[date][order].append({
"time" : time,
"usernumber" : usernumber,
"timelapse" : timelapse,
"figure" : figure,
"quantor_1" : quantor_1,
"quantor_2" : quantor_2,
"quantor_3" : quantor_3,
"term_S" : term_S,
"term_M" : term_M,
"term_P" : term_P,
"correctness" : correctness,
"syllogism_validity" : syllogism_validity,
"conclusion_truth_value" : conclusion_truth_value,
"student_answer" : student_answer,
"propkind" : propkind,
})
# Create the HttpResponse object with the appropriate CSV header.
response = HttpResponse("", content_type='text/csv')
response['Content-Disposition'] = 'attachment; filename=logsyllog-data3.csv'
writer = csv.writer(response)
if not bAggregateSessions:
first_row = ['date', 'time', 'usernumber', 'figure', 'quantor_1', 'quantor_2', 'quantor_3', 'term_S', 'term_M', 'term_P', 'conclusion_truth_value', 'syllogism_validity', 'answer_is_correct', 'answer', 'propkind', 'timelapse (s)', 'totaltime (s)', 'lastcorrectinarow', 'answercount', 'correctanswercount']
else:
first_row = ['date', 'usernumber', 'totaltime (s)', 'lastcorrectinarow', 'answercount', 'correctanswercount']
writer.writerow(first_row)
for date in sorted(result):
for order in sorted(result[date]):
usernumber = 0
last_correct_in_a_row = 0
totaltime = 0
answercount = 0
correctanswercount = 0
for mydict in result[date][order]:
answercount += 1
next_row = []
next_row.append(date)
next_row.append('%s' % mydict['time'])
next_row.append('%d' % mydict['usernumber'])
usernumber = mydict['usernumber']
next_row.append('%d' % mydict['figure'])
next_row.append('%s' % mydict['quantor_1'])
next_row.append('%s' % mydict['quantor_2'])
next_row.append('%s' % mydict['quantor_3'])
next_row.append('%s' % mydict['term_S'])
next_row.append('%s' % mydict['term_M'])
next_row.append('%s' % mydict['term_P'])
next_row.append('%s' % mydict['conclusion_truth_value'])
next_row.append('%s' % mydict['syllogism_validity'])
next_row.append('%s' % mydict['correctness'])
next_row.append('%s' % mydict['student_answer'])
next_row.append('%s' % mydict['propkind'])
next_row.append('%.3f' % ((1.0 * mydict['timelapse']) / 1000.0))
totaltime += mydict['timelapse']
if not bAggregateSessions:
writer.writerow(next_row)
del next_row
if mydict["correctness"]:
last_correct_in_a_row += 1
correctanswercount += 1
else:
last_correct_in_a_row = 0
if not bAggregateSessions:
next_row = [date, '', '%d' % usernumber, '', '', '', '', '', '', '', '', '', '', '', '', '', '%.3f' % ((1.0 * totaltime) / 1000.0), '%d' % last_correct_in_a_row, '%d' % answercount, '%d' % correctanswercount]
else:
next_row = [date, '%d' % usernumber, '%.3f' % ((1.0 * totaltime) / 1000.0), '%d' % last_correct_in_a_row, '%d' % answercount, '%d' % correctanswercount]
writer.writerow(next_row)
del next_row
return response
def getSyllogismName(figure, quantor_1, quantor_2, quantor_3):
return u"%d:%s%s%s" % (figure, quantor_1[0], quantor_2[0], quantor_3[0])
def dump3_syllogism_based_validonly(request):
bDoValidSyllogismsOnly = True
start_date = None
end_date = None
return dump3_syllogism_based_super(request, bDoValidSyllogismsOnly, start_date, end_date)
def dump3_syllogism_based_validonly_dates(request, start_date, end_date):
bDoValidSyllogismsOnly = True
return dump3_syllogism_based_super(request, bDoValidSyllogismsOnly, start_date, end_date)
def dump3_syllogism_based_valid_and_invalid(request):
bDoValidSyllogismsOnly = False
start_date = None
end_date = None
return dump3_syllogism_based_super(request, bDoValidSyllogismsOnly, start_date, end_date)
def dump3_syllogism_based_valid_and_invalid_dates(request, start_date, end_date):
bDoValidSyllogismsOnly = False
return dump3_syllogism_based_super(request, bDoValidSyllogismsOnly, start_date, end_date)
def dump3_syllogism_based_super(request, bDoValidSyllogismsOnly, start_date, end_date):
if not request.user.is_authenticated:
return HttpResponseForbidden()
else:
import csv
answer_list = Answer3.objects.all().order_by("id")
result = {} # date --> { syllogism_name --> { mydict } }
mytz = GMT1()
if start_date == None and end_date == None:
bDoDate = True
else:
bDoDate = False
for answer3 in answer_list:
usernumber = answer3.usernumber
# Convert aware datetime into our own timezone
mydate = answer3.date
date = (u"%s" % mydate)[0:10]
time = (u"%s" % mydate)[11:19]
if bDoDate:
real_date = date
else:
bDoDate = False
if date >= start_date and date <= end_date:
pass
else:
continue
real_date = "0000-00-00"
timelapse = answer3.timelapse
figure = answer3.figure
quantor_1 = answer3.quantor_1
quantor_2 = answer3.quantor_2
quantor_3 = answer3.quantor_3
syllogism_name = getSyllogismName(figure, quantor_1, quantor_2, quantor_3)
correctness = answer3.correctness
conclusion_truth_value = answer3.conclusion_truth_value
student_answer = answer3.student_answer
propkind = answer3.propkind
if student_answer == "valid":
if correctness:
syllogism_validity = "valid"
else:
syllogism_validity = "invalid"
else:
if correctness:
syllogism_validity = "invalid"
else:
syllogism_validity = "valid"
if bDoValidSyllogismsOnly:
if syllogism_validity == "valid":
bDoIt = True
else:
bDoIt = False
else:
bDoIt = True
if bDoIt:
result.setdefault(real_date, {})
result[real_date].setdefault(syllogism_name, {
"question_count" : 0,
"correct_count" : 0,
"validity" : syllogism_validity,
})
result[real_date][syllogism_name]["question_count"] += 1
if correctness:
result[real_date][syllogism_name]["correct_count"] += 1
# Create the HttpResponse object with the appropriate CSV header.
response = HttpResponse("", content_type='text/csv')
response['Content-Disposition'] = 'attachment; filename=logsyllog-data3-syllogism_based.csv'
writer = csv.writer(response)
if bDoDate:
first_row = ['date', 'syl_name', 'validity', 'question_count', 'correct_count']
writer.writerow(first_row)
else:
first_row = ['date_range', 'syl_name', 'validity', 'question_count', 'correct_count']
writer.writerow(first_row)
for date in sorted(result):
for syllogism_name in sorted(result[date]):
question_count = result[date][syllogism_name]["question_count"]
correct_count = result[date][syllogism_name]["correct_count"]
validity = result[date][syllogism_name]["validity"]
next_row = []
if bDoDate:
next_row.append(date)
else:
if start_date == end_date:
next_row.append("%s" % start_date)
else:
next_row.append("%s-%s" % (start_date, end_date))
next_row.append(syllogism_name)
next_row.append(validity)
next_row.append('%d' % question_count)
next_row.append('%d' % correct_count)
writer.writerow(next_row)
del next_row
return response
def dump4(request):
bAggregateSessions = False
return dump4_super(request, bAggregateSessions)
def dump4_aggregatesess(request):
bAggregateSessions = True
return dump4_super(request, bAggregateSessions)
def dump4_super(request, bAggregateSessions):
if not request.user.is_authenticated:
return HttpResponseForbidden()
else:
import csv
answer_list = Answer4.objects.all().order_by("id")
result = {} # date --> { order --> { mydict } }
order_dict = {} #usernumber --> order-int
order_int = 0
mytz = GMT1()
for answer4 in answer_list:
usernumber = answer4.usernumber
# Convert aware datetime into our own timezone
mydate = answer4.date
date = (u"%s" % mydate)[0:10]
time = (u"%s" % mydate)[11:19]
timelapse = answer4.timelapse
rule_clicks = answer4.rule_clicks
figure = answer4.figure
quantor_1 = answer4.quantor_1
quantor_2 = answer4.quantor_2
quantor_3 = answer4.quantor_3
term_S = term2string[answer4.term_S]
term_M = term2string[answer4.term_M]
term_P = term2string[answer4.term_P]
correctness = answer4.correctness
conclusion_truth_value = answer4.conclusion_truth_value
student_answer = answer4.student_answer
propkind = answer4.propkind
if student_answer == "valid":
if correctness:
syllogism_validity = "valid"
else:
syllogism_validity = "invalid"
else:
if correctness:
syllogism_validity = "invalid"
else:
syllogism_validity = "valid"
if usernumber in order_dict:
order = order_dict[usernumber]
else:
order = order_int
order_int += 1
order_dict[usernumber] = order
result.setdefault(date, {})
result[date].setdefault(order, [])
result[date][order].append({
"time" : time,
"usernumber" : usernumber,
"timelapse" : timelapse,
"rule_clicks" : rule_clicks,
"figure" : figure,
"quantor_1" : quantor_1,
"quantor_2" : quantor_2,
"quantor_3" : quantor_3,
"term_S" : term_S,
"term_M" : term_M,
"term_P" : term_P,
"correctness" : correctness,
"syllogism_validity" : syllogism_validity,
"conclusion_truth_value" : conclusion_truth_value,
"student_answer" : student_answer,
"propkind" : propkind,
})
# Create the HttpResponse object with the appropriate CSV header.
response = HttpResponse("", content_type='text/csv')
response['Content-Disposition'] = 'attachment; filename=logsyllog-data4.csv'
writer = csv.writer(response)
if not bAggregateSessions:
first_row = ['date', 'time', 'usernumber', 'figure', 'quantor_1', 'quantor_2', 'quantor_3', 'term_S', 'term_M', 'term_P', 'conclusion_truth_value', 'syllogism_validity', 'answer_is_correct', 'answer', 'propkind', 'timelapse (s)', 'totaltime (s)', 'lastcorrectinarow', 'answercount', 'correctanswercount', 'rule_clicks']
else:
first_row = ['date', 'usernumber', 'totaltime (s)', 'lastcorrectinarow', 'answercount', 'correctanswercount', 'total_rule_clicks']
writer.writerow(first_row)
for date in sorted(result):
for order in sorted(result[date]):
usernumber = 0
last_correct_in_a_row = 0
totaltime = 0
answercount = 0
correctanswercount = 0
total_rule_clicks = 0
for mydict in result[date][order]:
answercount += 1
next_row = []
next_row.append(date)
next_row.append('%s' % mydict['time'])
next_row.append('%d' % mydict['usernumber'])
usernumber = mydict['usernumber']
next_row.append('%d' % mydict['figure'])
next_row.append('%s' % mydict['quantor_1'])
next_row.append('%s' % mydict['quantor_2'])
next_row.append('%s' % mydict['quantor_3'])
next_row.append('%s' % mydict['term_S'])
next_row.append('%s' % mydict['term_M'])
next_row.append('%s' % mydict['term_P'])
next_row.append('%s' % mydict['conclusion_truth_value'])
next_row.append('%s' % mydict['syllogism_validity'])
next_row.append('%s' % mydict['correctness'])
next_row.append('%s' % mydict['student_answer'])
next_row.append('%s' % mydict['propkind'])
next_row.append('%.3f' % ((1.0 * mydict['timelapse']) / 1000.0))
next_row.append('') # totaltime, only valid for the aggregate rows
next_row.append('%d' % last_correct_in_a_row)
next_row.append('%d' % answercount)
next_row.append('%d' % correctanswercount)
next_row.append('%d' % mydict['rule_clicks'])
totaltime += mydict['timelapse']
if not bAggregateSessions:
writer.writerow(next_row)
del next_row
if mydict["correctness"]:
last_correct_in_a_row += 1
correctanswercount += 1
else:
last_correct_in_a_row = 0
total_rule_clicks += mydict['rule_clicks']
if not bAggregateSessions:
next_row = [date, '', '%d' % usernumber, '', '', '', '', '', '', '', '', '', '', '', '', '', '%.3f' % ((1.0 * totaltime) / 1000.0), '%d' % last_correct_in_a_row, '%d' % answercount, '%d' % correctanswercount, '%d' % total_rule_clicks]
else:
next_row = [date, '%d' % usernumber, '%.3f' % ((1.0 * totaltime) / 1000.0), '%d' % last_correct_in_a_row, '%d' % answercount, '%d' % correctanswercount, '%d' % total_rule_clicks]
writer.writerow(next_row)
del next_row
return response
def dump5(request):
bAggregateSessions = False
return dump5_super(request, bAggregateSessions)
def dump5_aggregatesess(request):
bAggregateSessions = True
return dump5_super(request, bAggregateSessions)
def dump5_super(request, bAggregateSessions):
if not request.user.is_authenticated:
return HttpResponseForbidden()
else:
import csv
answer_list = Answer5.objects.all().order_by("id")
result = {} # date --> { order --> { mydict } }
order_dict = {} #usernumber --> order-int
order_int = 0
mytz = GMT1()
for answer5 in answer_list:
usernumber = answer5.usernumber
# Convert aware datetime into our own timezone
mydate = answer5.date
date = (u"%s" % mydate)[0:10]
time = (u"%s" % mydate)[11:19]
timelapse = answer5.timelapse
figure = answer5.figure
quantor_1 = answer5.quantor_1
quantor_2 = answer5.quantor_2
quantor_3 = answer5.quantor_3
term_S = artificialterm2string[answer5.term_S]
term_M = artificialterm2string[answer5.term_M]
term_P = artificialterm2string[answer5.term_P]
correctness = answer5.correctness
conclusion_truth_value = answer5.conclusion_truth_value
student_answer = answer5.student_answer
propkind = answer5.propkind
if student_answer == "valid":
if correctness:
syllogism_validity = "valid"
else:
syllogism_validity = "invalid"
else:
if correctness:
syllogism_validity = "invalid"
else:
syllogism_validity = "valid"
if usernumber in order_dict:
order = order_dict[usernumber]
else:
order = order_int
order_int += 1
order_dict[usernumber] = order
result.setdefault(date, {})
result[date].setdefault(order, [])
result[date][order].append({
"time" : time,
"usernumber" : usernumber,
"timelapse" : timelapse,
"figure" : figure,
"quantor_1" : quantor_1,
"quantor_2" : quantor_2,
"quantor_3" : quantor_3,
"term_S" : term_S,
"term_M" : term_M,
"term_P" : term_P,
"correctness" : correctness,
"syllogism_validity" : syllogism_validity,
"conclusion_truth_value" : conclusion_truth_value,
"student_answer" : student_answer,
"propkind" : propkind,
})
# Create the HttpResponse object with the appropriate CSV header.
response = HttpResponse("", content_type='text/csv')
response['Content-Disposition'] = 'attachment; filename=logsyllog-data5.csv'
writer = csv.writer(response)
if not bAggregateSessions:
first_row = ['date', 'time', 'usernumber', 'figure', 'quantor_1', 'quantor_2', 'quantor_3', 'term_S', 'term_M', 'term_P', 'conclusion_truth_value', 'syllogism_validity', 'answer_is_correct', 'answer', 'propkind', 'timelapse (s)', 'totaltime (s)', 'lastcorrectinarow', 'answercount', 'correctanswercount']
else:
first_row = ['date', 'usernumber', 'totaltime (s)', 'lastcorrectinarow', 'answercount', 'correctanswercount']
writer.writerow(first_row)
for date in sorted(result):
for order in sorted(result[date]):
usernumber = 0
last_correct_in_a_row = 0
totaltime = 0
answercount = 0
correctanswercount = 0
for mydict in result[date][order]:
answercount += 1
next_row = []
next_row.append(date)
next_row.append('%s' % mydict['time'])
next_row.append('%d' % mydict['usernumber'])
usernumber = mydict['usernumber']
next_row.append('%d' % mydict['figure'])
next_row.append('%s' % mydict['quantor_1'])
next_row.append('%s' % mydict['quantor_2'])
next_row.append('%s' % mydict['quantor_3'])
next_row.append('%s' % mydict['term_S'])
next_row.append('%s' % mydict['term_M'])
next_row.append('%s' % mydict['term_P'])
next_row.append('%s' % mydict['conclusion_truth_value'])
next_row.append('%s' % mydict['syllogism_validity'])
next_row.append('%s' % mydict['correctness'])
next_row.append('%s' % mydict['student_answer'])
next_row.append('%s' % mydict['propkind'])
next_row.append('%.3f' % ((1.0 * mydict['timelapse']) / 1000.0))
totaltime += mydict['timelapse']
if not bAggregateSessions:
writer.writerow(next_row)
del next_row
if mydict["correctness"]:
last_correct_in_a_row += 1
correctanswercount += 1
else:
last_correct_in_a_row = 0
if not bAggregateSessions:
next_row = [date, '', '%d' % usernumber, '', '', '', '', '', '', '', '', '', '', '', '', '', '%.3f' % ((1.0 * totaltime) / 1000.0), '%d' % last_correct_in_a_row, '%d' % answercount, '%d' % correctanswercount]
else:
next_row = [date, '%d' % usernumber, '%.3f' % ((1.0 * totaltime) / 1000.0), '%d' % last_correct_in_a_row, '%d' % answercount, '%d' % correctanswercount]
writer.writerow(next_row)
del next_row
return response
|
993,880 | abb7a373786a45177f303baf91095ee219ced55e |
import numpy as np
from collections import namedtuple
class ObjectInsertion:
"""
This class will handle every object manipulation inside the background image area
"""
def __init__(self, max_iou=0.4):
self._max_iou = max_iou
self._Centroid = namedtuple('Centroid', 'x, y, w, h')
def _iou(self, obj_a, obj_b):
"""
Calculate Itersection Over Union between 2 boxes
"""
# compute the area of both the prediction and ground-truth
# rectangles
box_a_area = (obj_a[2] - obj_a[0] + 1) * (obj_a[3] - obj_a[1] + 1)
box_b_area = (obj_b[2] - obj_b[0] + 1) * (obj_b[3] - obj_b[1] + 1)
# determine the (x, y)-coordinates of the intersection rectangle
x_a = max(obj_a[0], obj_b[0])
y_a = max(obj_a[1], obj_b[1])
x_b = min(obj_a[2], obj_b[2])
y_b = min(obj_a[3], obj_b[3])
# compute the area of intersection rectangle
inter_area = max(0, x_b - x_a + 1) * max(0, y_b - y_a + 1)
# compute the intersection over union by taking the intersection
# area and dividing it by the sum of prediction + ground-truth
# areas - the interesection area
iou = inter_area / float(box_a_area + box_b_area - inter_area)
# return the intersection over union value
return iou
@staticmethod
def _get_random_point_within_boundaries(position_range, boundary):
try:
return np.random.randint(boundary, position_range - boundary)
except ValueError:
return boundary
@staticmethod
def xywd_2_xyxy(obj):
x1 = obj.x - obj.w//2
x2 = obj.x + obj.w//2
y1 = obj.y - obj.h//2
y2 = obj.y + obj.h//2
return [x1, y1, x2, y2]
def _get_new_object_position(self, bgi_shape, img_shape):
background_height, background_width = bgi_shape[:2]
# Define x/y boundaries given the object size
x_boundary = (img_shape[1] // 2) + 10
y_boundary = (img_shape[0] // 2) + 10
x_candidate = self._get_random_point_within_boundaries(background_width, x_boundary)
y_candidate = self._get_random_point_within_boundaries(background_height, y_boundary)
return self._Centroid(x_candidate, y_candidate, img_shape[1], img_shape[0])
def __call__(self, bgi, img, centroid_list, *args, **kwargs):
while True:
img_centroid = self._get_new_object_position(bgi.shape, img.shape)
img_coordinates = self.xywd_2_xyxy(img_centroid)
is_feasible = True
for centroid in centroid_list:
if self._iou(img_coordinates, self.xywd_2_xyxy(centroid)) >= 0.4:
is_feasible = False
break
# is_feasible = not any([self._iou(img_coordinates, self._xywd_2_xyxy(centroid)) > 0.4
# for centroid in centroid_list])
if is_feasible:
break
bgi[img_coordinates[1]:img_coordinates[1]+img.shape[0], img_coordinates[0]:img_coordinates[0]+img.shape[1], ...] = img
return img_centroid, bgi
|
993,881 | abb223ca47f1104f8345b9273ab2bb63099eef06 | from django.contrib import admin
from . import models
admin.site.register([
models.Blog,
models.Author,
models.Entry
]) |
993,882 | aaf47db225475a3ab58932327d28d48e116f5863 | from setuptools import setup
setup(
name='array_to_struct',
version='1.0',
description='',
author='',
author_email='',
url='',
packages=['transp_solution'],
include_package_data=True,
zip_safe=False
) |
993,883 | a8aa4938391df1604ba7337789e7225314c86534 | class Solution(object):
def isIsomorphic(self, s, t):
"""
:type s: str
:type t: str
:rtype: bool
"""
if len(t) != len(s):
return False
d_s = [-1 for _ in xrange(256)]
d_t = [-1 for _ in xrange(256)]
s_index = range(len(s))
t_index = range(len(t))
for i in xrange(len(s)):
if d_s[ord(s[i])] == -1:
d_s[ord(s[i])] = i
else:
s_index[i] = d_s[ord(s[i])]
if d_t[ord(t[i])] == -1:
d_t[ord(t[i])] = i
else:
t_index[i] = d_t[ord(t[i])]
for i in xrange(len(s)):
if s_index[i] != t_index[i]:
return False
print s_index
print t_index
return True
s = Solution()
print s.isIsomorphic('ab', 'aa')
|
993,884 | e4bba0d0acb89cd2b6f84b190711487674794203 | import math
def next_position(m,i,j,d):
if not m[j-1][i-1] == 1:
m[j-1][i-1]=5
else:
#print("Error: Position Error")
return -1
#print("{} {} {}".format(i,j,d))
if d == 1:
j += 1
elif d == 2:
i += 1
elif d == 3:
j -= 1
else:
i -= 1
if j>10 or i>10 or i<1 or j<1 or m[j-1][i-1] == 1:
#print("Error: Out of Range or Obstacle")
return -1
else:
m[j-1][i-1]=5
#print("{} {} {}".format(i,j,d))
return i,j,d
def all_posi(m,i,j):
#print("{} {}".format(i,j))
answer = []
for x in range(4):
posi_n = next_position(m,i,j,x+1)
if not posi_n == -1:
answer.append(posi_n)
return answer
def find_distance(u,v):
x = u[0] - v[0]
y = u[1] - v[1]
dis = math.pow(x,2) + math.pow(y,2)
return math.sqrt(dis)
#m = [[0,0,0,0,0,0,0,0,1,1],
# [0,0,0,0,0,1,0,0,0,0],
# [1,0,1,0,0,0,1,0,0,0],
# [1,0,0,1,0,0,0,0,0,1],
# [0,1,1,1,1,0,0,0,0,1],
# [0,1,0,0,0,0,0,0,0,1],
# [0,0,0,0,0,0,0,1,0,0],
# [1,0,0,0,0,0,0,0,1,0],
# [1,0,0,1,0,0,0,1,0,1],
# [1,1,0,0,0,0,0,0,0,0]]
m = [[0,0,0,0,0,0,0,0,1,1],
[0,0,0,0,0,1,0,0,0,0],
[1,0,1,0,0,0,1,0,0,0],
[1,0,0,1,0,0,0,0,0,1],
[0,1,1,1,1,0,0,0,0,1],
[0,1,0,0,0,0,1,0,0,1],
[0,0,0,0,0,0,0,0,0,0],
[1,0,0,0,0,0,0,1,1,0],
[1,0,0,1,0,0,0,1,0,1],
[1,1,0,0,0,0,0,0,0,0]]
#answer = all_posi(m,3,10)
#for row in m:
# print(row)
#print(answer)
#for posi in answer:
# print(find_distance([3,4],posi))
#print(find_distance([1,1],[3,3]))
#print(find_distance([1,1],[1,3]))
path = []
current_post = [3,10,3]
target_post = [3,4]
path.append(current_post)
for _ in range(6):
closets_dis = float('inf')
answer = all_posi(m,current_post[0],current_post[1])
print(answer)
for posi in answer:
current_dis = find_distance(target_post,posi)
if current_dis<closets_dis:
current_post = posi
closets_dis = current_dis
path.append(current_post)
for row in m:
print(row)
print(path) |
993,885 | c2cf748f8f5e568f5cf432cc8eed72e93574f07f | import numpy as np, pandas as pd, os
# import my functions
import vwap
# read file. Only allows ONE product in each file!
# Make sure the titles are
# ['ProductionYear', 'VolIGWh', 'Price', 'TradeDate', 'CY', 'TradeYear', 'TradeMon', 'VolTimesPrice', 'DummyDate']
# only these are used: ProductionYear, VolIGWh, Price, TradeYear, TradeMon, VolTimesPrice
# make sure the titles' name
# from pandas import DataFrame
# may need to change file name if processing other products
trades = pd.read_csv("Nordic.hydro.trades.csv")
# test, true val = 8.66, 1540
# print(vwap.myvwap(2015,4,2015,trades), vwap.myvol(2015,4,2015,trades),)
# give months
monthsRaw = pd.date_range(start='20110401', periods=100, freq='1M')
months = []
# convert timestamp to a string
for item in monthsRaw:
months.append(item.strftime('%Y-%m-%d'))
# prod year series
ProdYearSer = np.linspace(2010, 2030, 21, dtype=int)
# tables for results
vwapTabel = pd.DataFrame(index=months, columns=ProdYearSer)
volTabel = pd.DataFrame(index=months, columns=ProdYearSer)
for thisMonth in months:
for thisProdYr in ProdYearSer:
# def myvwap(TY, TM, PY, data):
# print(int(thisMonth[0:4]), int(thisMonth[5:7]), thisProdYr)
# compute VWAP, round to 2 decimals
thisVwap = vwap.myvwap(int(thisMonth[0:4]), int(thisMonth[5:7]), thisProdYr, trades)
vwapTabel.ix[thisMonth, thisProdYr] = round(thisVwap, 2)
# compute volume in GWh, round to 2 decimalsh
thisVol = vwap.myvol(int(thisMonth[0:4]), int(thisMonth[5:7]), thisProdYr, trades)
volTabel.ix[thisMonth, thisProdYr] = round(thisVol, 2)
print(vwapTabel)
print(volTabel)
print('Calculation completed!')
print('\n')
# give file name by product name
myFileName = input("Product name (e.g. Nordic hydro) = ")
print('\n')
# write to CSV
vwapTabel.to_csv('VWAP_' + myFileName + '.csv')
volTabel.to_csv('Volume_in_GWh_' + myFileName + '.csv')
# open directory
print('CSV files are ready. Check the folder!')
print('\n')
os.startfile(os.getcwd())
# plot heat map
# NOT working yet. Attempt to plot "Months" and get error!
# import seaborn as sns
# import matplotlib.pyplot as plt
#
#
# print(subVWAP)
# myHeat = sns.heatmap(subVWAP, annot=True, linewidths=.1)
#
# myHeat.xticks(np.arange(7) + 0.5, ProdYearSer)
# myHeat.yticks(np.arange(5) + 0.5, months)
#
# myHeat.set_title('VWAP of ' + myFileName)
# myHeat.set_xlabel('Production Year')
# myHeat.set_ylabel('')
#
# plt.show() |
993,886 | 50b367470b0ada99b801fd46216bca8fee772454 | # This file was automatically generated by SWIG (http://www.swig.org).
# Version 3.0.12
#
# Do not make changes to this file unless you know what you are doing--modify
# the SWIG interface file instead.
from sys import version_info as _swig_python_version_info
if _swig_python_version_info >= (2, 7, 0):
def swig_import_helper():
import importlib
pkg = __name__.rpartition('.')[0]
mname = '.'.join((pkg, '_PyMeshUtils')).lstrip('.')
try:
return importlib.import_module(mname)
except ImportError:
return importlib.import_module('_PyMeshUtils')
_PyMeshUtils = swig_import_helper()
del swig_import_helper
elif _swig_python_version_info >= (2, 6, 0):
def swig_import_helper():
from os.path import dirname
import imp
fp = None
try:
fp, pathname, description = imp.find_module('_PyMeshUtils', [dirname(__file__)])
except ImportError:
import _PyMeshUtils
return _PyMeshUtils
try:
_mod = imp.load_module('_PyMeshUtils', fp, pathname, description)
finally:
if fp is not None:
fp.close()
return _mod
_PyMeshUtils = swig_import_helper()
del swig_import_helper
else:
import _PyMeshUtils
del _swig_python_version_info
try:
_swig_property = property
except NameError:
pass # Python < 2.2 doesn't have 'property'.
try:
import builtins as __builtin__
except ImportError:
import __builtin__
def _swig_setattr_nondynamic(self, class_type, name, value, static=1):
if (name == "thisown"):
return self.this.own(value)
if (name == "this"):
if type(value).__name__ == 'SwigPyObject':
self.__dict__[name] = value
return
method = class_type.__swig_setmethods__.get(name, None)
if method:
return method(self, value)
if (not static):
if _newclass:
object.__setattr__(self, name, value)
else:
self.__dict__[name] = value
else:
raise AttributeError("You cannot add attributes to %s" % self)
def _swig_setattr(self, class_type, name, value):
return _swig_setattr_nondynamic(self, class_type, name, value, 0)
def _swig_getattr(self, class_type, name):
if (name == "thisown"):
return self.this.own()
method = class_type.__swig_getmethods__.get(name, None)
if method:
return method(self)
raise AttributeError("'%s' object has no attribute '%s'" % (class_type.__name__, name))
def _swig_repr(self):
try:
strthis = "proxy of " + self.this.__repr__()
except __builtin__.Exception:
strthis = ""
return "<%s.%s; %s >" % (self.__class__.__module__, self.__class__.__name__, strthis,)
try:
_object = object
_newclass = 1
except __builtin__.Exception:
class _object:
pass
_newclass = 0
class ZSparseMatrix(_object):
__swig_setmethods__ = {}
__setattr__ = lambda self, name, value: _swig_setattr(self, ZSparseMatrix, name, value)
__swig_getmethods__ = {}
__getattr__ = lambda self, name: _swig_getattr(self, ZSparseMatrix, name)
__repr__ = _swig_repr
def __init__(self, *args):
this = _PyMeshUtils.new_ZSparseMatrix(*args)
try:
self.this.append(this)
except __builtin__.Exception:
self.this = this
__swig_destroy__ = _PyMeshUtils.delete_ZSparseMatrix
__del__ = lambda self: None
def num_rows(self):
return _PyMeshUtils.ZSparseMatrix_num_rows(self)
def num_cols(self):
return _PyMeshUtils.ZSparseMatrix_num_cols(self)
def get_inner_size(self):
return _PyMeshUtils.ZSparseMatrix_get_inner_size(self)
def get_outer_size(self):
return _PyMeshUtils.ZSparseMatrix_get_outer_size(self)
def get_value_size(self):
return _PyMeshUtils.ZSparseMatrix_get_value_size(self)
def get_inner_indices(self, np_idx_array):
return _PyMeshUtils.ZSparseMatrix_get_inner_indices(self, np_idx_array)
def get_outer_indices(self, np_idx_array):
return _PyMeshUtils.ZSparseMatrix_get_outer_indices(self, np_idx_array)
def get_values(self, np_value_array):
return _PyMeshUtils.ZSparseMatrix_get_values(self, np_value_array)
def import_raw_csc(self, num_rows, num_cols, inner_idx_array, outer_idx_array, value_array):
return _PyMeshUtils.ZSparseMatrix_import_raw_csc(self, num_rows, num_cols, inner_idx_array, outer_idx_array, value_array)
def import_raw_coo(self, num_rows, num_cols, row_indices, col_indices, value_array):
return _PyMeshUtils.ZSparseMatrix_import_raw_coo(self, num_rows, num_cols, row_indices, col_indices, value_array)
ZSparseMatrix_swigregister = _PyMeshUtils.ZSparseMatrix_swigregister
ZSparseMatrix_swigregister(ZSparseMatrix)
SHARED_PTR_DISOWN = _PyMeshUtils.SHARED_PTR_DISOWN
class SwigPyIterator(_object):
__swig_setmethods__ = {}
__setattr__ = lambda self, name, value: _swig_setattr(self, SwigPyIterator, name, value)
__swig_getmethods__ = {}
__getattr__ = lambda self, name: _swig_getattr(self, SwigPyIterator, name)
def __init__(self, *args, **kwargs):
raise AttributeError("No constructor defined - class is abstract")
__repr__ = _swig_repr
__swig_destroy__ = _PyMeshUtils.delete_SwigPyIterator
__del__ = lambda self: None
def value(self):
return _PyMeshUtils.SwigPyIterator_value(self)
def incr(self, n=1):
return _PyMeshUtils.SwigPyIterator_incr(self, n)
def decr(self, n=1):
return _PyMeshUtils.SwigPyIterator_decr(self, n)
def distance(self, x):
return _PyMeshUtils.SwigPyIterator_distance(self, x)
def equal(self, x):
return _PyMeshUtils.SwigPyIterator_equal(self, x)
def copy(self):
return _PyMeshUtils.SwigPyIterator_copy(self)
def next(self):
return _PyMeshUtils.SwigPyIterator_next(self)
def __next__(self):
return _PyMeshUtils.SwigPyIterator___next__(self)
def previous(self):
return _PyMeshUtils.SwigPyIterator_previous(self)
def advance(self, n):
return _PyMeshUtils.SwigPyIterator_advance(self, n)
def __eq__(self, x):
return _PyMeshUtils.SwigPyIterator___eq__(self, x)
def __ne__(self, x):
return _PyMeshUtils.SwigPyIterator___ne__(self, x)
def __iadd__(self, n):
return _PyMeshUtils.SwigPyIterator___iadd__(self, n)
def __isub__(self, n):
return _PyMeshUtils.SwigPyIterator___isub__(self, n)
def __add__(self, n):
return _PyMeshUtils.SwigPyIterator___add__(self, n)
def __sub__(self, *args):
return _PyMeshUtils.SwigPyIterator___sub__(self, *args)
def __iter__(self):
return self
SwigPyIterator_swigregister = _PyMeshUtils.SwigPyIterator_swigregister
SwigPyIterator_swigregister(SwigPyIterator)
class vectori(_object):
__swig_setmethods__ = {}
__setattr__ = lambda self, name, value: _swig_setattr(self, vectori, name, value)
__swig_getmethods__ = {}
__getattr__ = lambda self, name: _swig_getattr(self, vectori, name)
__repr__ = _swig_repr
def iterator(self):
return _PyMeshUtils.vectori_iterator(self)
def __iter__(self):
return self.iterator()
def __nonzero__(self):
return _PyMeshUtils.vectori___nonzero__(self)
def __bool__(self):
return _PyMeshUtils.vectori___bool__(self)
def __len__(self):
return _PyMeshUtils.vectori___len__(self)
def __getslice__(self, i, j):
return _PyMeshUtils.vectori___getslice__(self, i, j)
def __setslice__(self, *args):
return _PyMeshUtils.vectori___setslice__(self, *args)
def __delslice__(self, i, j):
return _PyMeshUtils.vectori___delslice__(self, i, j)
def __delitem__(self, *args):
return _PyMeshUtils.vectori___delitem__(self, *args)
def __getitem__(self, *args):
return _PyMeshUtils.vectori___getitem__(self, *args)
def __setitem__(self, *args):
return _PyMeshUtils.vectori___setitem__(self, *args)
def pop(self):
return _PyMeshUtils.vectori_pop(self)
def append(self, x):
return _PyMeshUtils.vectori_append(self, x)
def empty(self):
return _PyMeshUtils.vectori_empty(self)
def size(self):
return _PyMeshUtils.vectori_size(self)
def swap(self, v):
return _PyMeshUtils.vectori_swap(self, v)
def begin(self):
return _PyMeshUtils.vectori_begin(self)
def end(self):
return _PyMeshUtils.vectori_end(self)
def rbegin(self):
return _PyMeshUtils.vectori_rbegin(self)
def rend(self):
return _PyMeshUtils.vectori_rend(self)
def clear(self):
return _PyMeshUtils.vectori_clear(self)
def get_allocator(self):
return _PyMeshUtils.vectori_get_allocator(self)
def pop_back(self):
return _PyMeshUtils.vectori_pop_back(self)
def erase(self, *args):
return _PyMeshUtils.vectori_erase(self, *args)
def __init__(self, *args):
this = _PyMeshUtils.new_vectori(*args)
try:
self.this.append(this)
except __builtin__.Exception:
self.this = this
def push_back(self, x):
return _PyMeshUtils.vectori_push_back(self, x)
def front(self):
return _PyMeshUtils.vectori_front(self)
def back(self):
return _PyMeshUtils.vectori_back(self)
def assign(self, n, x):
return _PyMeshUtils.vectori_assign(self, n, x)
def resize(self, *args):
return _PyMeshUtils.vectori_resize(self, *args)
def insert(self, *args):
return _PyMeshUtils.vectori_insert(self, *args)
def reserve(self, n):
return _PyMeshUtils.vectori_reserve(self, n)
def capacity(self):
return _PyMeshUtils.vectori_capacity(self)
__swig_destroy__ = _PyMeshUtils.delete_vectori
__del__ = lambda self: None
vectori_swigregister = _PyMeshUtils.vectori_swigregister
vectori_swigregister(vectori)
class vectord(_object):
__swig_setmethods__ = {}
__setattr__ = lambda self, name, value: _swig_setattr(self, vectord, name, value)
__swig_getmethods__ = {}
__getattr__ = lambda self, name: _swig_getattr(self, vectord, name)
__repr__ = _swig_repr
def iterator(self):
return _PyMeshUtils.vectord_iterator(self)
def __iter__(self):
return self.iterator()
def __nonzero__(self):
return _PyMeshUtils.vectord___nonzero__(self)
def __bool__(self):
return _PyMeshUtils.vectord___bool__(self)
def __len__(self):
return _PyMeshUtils.vectord___len__(self)
def __getslice__(self, i, j):
return _PyMeshUtils.vectord___getslice__(self, i, j)
def __setslice__(self, *args):
return _PyMeshUtils.vectord___setslice__(self, *args)
def __delslice__(self, i, j):
return _PyMeshUtils.vectord___delslice__(self, i, j)
def __delitem__(self, *args):
return _PyMeshUtils.vectord___delitem__(self, *args)
def __getitem__(self, *args):
return _PyMeshUtils.vectord___getitem__(self, *args)
def __setitem__(self, *args):
return _PyMeshUtils.vectord___setitem__(self, *args)
def pop(self):
return _PyMeshUtils.vectord_pop(self)
def append(self, x):
return _PyMeshUtils.vectord_append(self, x)
def empty(self):
return _PyMeshUtils.vectord_empty(self)
def size(self):
return _PyMeshUtils.vectord_size(self)
def swap(self, v):
return _PyMeshUtils.vectord_swap(self, v)
def begin(self):
return _PyMeshUtils.vectord_begin(self)
def end(self):
return _PyMeshUtils.vectord_end(self)
def rbegin(self):
return _PyMeshUtils.vectord_rbegin(self)
def rend(self):
return _PyMeshUtils.vectord_rend(self)
def clear(self):
return _PyMeshUtils.vectord_clear(self)
def get_allocator(self):
return _PyMeshUtils.vectord_get_allocator(self)
def pop_back(self):
return _PyMeshUtils.vectord_pop_back(self)
def erase(self, *args):
return _PyMeshUtils.vectord_erase(self, *args)
def __init__(self, *args):
this = _PyMeshUtils.new_vectord(*args)
try:
self.this.append(this)
except __builtin__.Exception:
self.this = this
def push_back(self, x):
return _PyMeshUtils.vectord_push_back(self, x)
def front(self):
return _PyMeshUtils.vectord_front(self)
def back(self):
return _PyMeshUtils.vectord_back(self)
def assign(self, n, x):
return _PyMeshUtils.vectord_assign(self, n, x)
def resize(self, *args):
return _PyMeshUtils.vectord_resize(self, *args)
def insert(self, *args):
return _PyMeshUtils.vectord_insert(self, *args)
def reserve(self, n):
return _PyMeshUtils.vectord_reserve(self, n)
def capacity(self):
return _PyMeshUtils.vectord_capacity(self)
__swig_destroy__ = _PyMeshUtils.delete_vectord
__del__ = lambda self: None
vectord_swigregister = _PyMeshUtils.vectord_swigregister
vectord_swigregister(vectord)
class vectors(_object):
__swig_setmethods__ = {}
__setattr__ = lambda self, name, value: _swig_setattr(self, vectors, name, value)
__swig_getmethods__ = {}
__getattr__ = lambda self, name: _swig_getattr(self, vectors, name)
__repr__ = _swig_repr
def iterator(self):
return _PyMeshUtils.vectors_iterator(self)
def __iter__(self):
return self.iterator()
def __nonzero__(self):
return _PyMeshUtils.vectors___nonzero__(self)
def __bool__(self):
return _PyMeshUtils.vectors___bool__(self)
def __len__(self):
return _PyMeshUtils.vectors___len__(self)
def __getslice__(self, i, j):
return _PyMeshUtils.vectors___getslice__(self, i, j)
def __setslice__(self, *args):
return _PyMeshUtils.vectors___setslice__(self, *args)
def __delslice__(self, i, j):
return _PyMeshUtils.vectors___delslice__(self, i, j)
def __delitem__(self, *args):
return _PyMeshUtils.vectors___delitem__(self, *args)
def __getitem__(self, *args):
return _PyMeshUtils.vectors___getitem__(self, *args)
def __setitem__(self, *args):
return _PyMeshUtils.vectors___setitem__(self, *args)
def pop(self):
return _PyMeshUtils.vectors_pop(self)
def append(self, x):
return _PyMeshUtils.vectors_append(self, x)
def empty(self):
return _PyMeshUtils.vectors_empty(self)
def size(self):
return _PyMeshUtils.vectors_size(self)
def swap(self, v):
return _PyMeshUtils.vectors_swap(self, v)
def begin(self):
return _PyMeshUtils.vectors_begin(self)
def end(self):
return _PyMeshUtils.vectors_end(self)
def rbegin(self):
return _PyMeshUtils.vectors_rbegin(self)
def rend(self):
return _PyMeshUtils.vectors_rend(self)
def clear(self):
return _PyMeshUtils.vectors_clear(self)
def get_allocator(self):
return _PyMeshUtils.vectors_get_allocator(self)
def pop_back(self):
return _PyMeshUtils.vectors_pop_back(self)
def erase(self, *args):
return _PyMeshUtils.vectors_erase(self, *args)
def __init__(self, *args):
this = _PyMeshUtils.new_vectors(*args)
try:
self.this.append(this)
except __builtin__.Exception:
self.this = this
def push_back(self, x):
return _PyMeshUtils.vectors_push_back(self, x)
def front(self):
return _PyMeshUtils.vectors_front(self)
def back(self):
return _PyMeshUtils.vectors_back(self)
def assign(self, n, x):
return _PyMeshUtils.vectors_assign(self, n, x)
def resize(self, *args):
return _PyMeshUtils.vectors_resize(self, *args)
def insert(self, *args):
return _PyMeshUtils.vectors_insert(self, *args)
def reserve(self, n):
return _PyMeshUtils.vectors_reserve(self, n)
def capacity(self):
return _PyMeshUtils.vectors_capacity(self)
__swig_destroy__ = _PyMeshUtils.delete_vectors
__del__ = lambda self: None
vectors_swigregister = _PyMeshUtils.vectors_swigregister
vectors_swigregister(vectors)
class Mesh(_object):
__swig_setmethods__ = {}
__setattr__ = lambda self, name, value: _swig_setattr(self, Mesh, name, value)
__swig_getmethods__ = {}
__getattr__ = lambda self, name: _swig_getattr(self, Mesh, name)
def __init__(self, *args, **kwargs):
raise AttributeError("No constructor defined")
__repr__ = _swig_repr
__swig_destroy__ = _PyMeshUtils.delete_Mesh
__del__ = lambda self: None
def get_dim(self):
return _PyMeshUtils.Mesh_get_dim(self)
def get_num_vertices(self):
return _PyMeshUtils.Mesh_get_num_vertices(self)
def get_num_faces(self):
return _PyMeshUtils.Mesh_get_num_faces(self)
def get_num_voxels(self):
return _PyMeshUtils.Mesh_get_num_voxels(self)
def get_vertex(self, *args):
return _PyMeshUtils.Mesh_get_vertex(self, *args)
def get_face(self, *args):
return _PyMeshUtils.Mesh_get_face(self, *args)
def get_voxel(self, *args):
return _PyMeshUtils.Mesh_get_voxel(self, *args)
def get_vertices(self, *args):
return _PyMeshUtils.Mesh_get_vertices(self, *args)
def get_faces(self, *args):
return _PyMeshUtils.Mesh_get_faces(self, *args)
def get_voxels(self, *args):
return _PyMeshUtils.Mesh_get_voxels(self, *args)
def get_vertex_per_face(self):
return _PyMeshUtils.Mesh_get_vertex_per_face(self)
def get_vertex_per_voxel(self):
return _PyMeshUtils.Mesh_get_vertex_per_voxel(self)
def enable_connectivity(self):
return _PyMeshUtils.Mesh_enable_connectivity(self)
def enable_vertex_connectivity(self):
return _PyMeshUtils.Mesh_enable_vertex_connectivity(self)
def enable_face_connectivity(self):
return _PyMeshUtils.Mesh_enable_face_connectivity(self)
def enable_voxel_connectivity(self):
return _PyMeshUtils.Mesh_enable_voxel_connectivity(self)
def get_vertex_adjacent_vertices(self, vi):
return _PyMeshUtils.Mesh_get_vertex_adjacent_vertices(self, vi)
def get_vertex_adjacent_faces(self, vi):
return _PyMeshUtils.Mesh_get_vertex_adjacent_faces(self, vi)
def get_vertex_adjacent_voxels(self, vi):
return _PyMeshUtils.Mesh_get_vertex_adjacent_voxels(self, vi)
def get_face_adjacent_faces(self, fi):
return _PyMeshUtils.Mesh_get_face_adjacent_faces(self, fi)
def get_face_adjacent_voxels(self, fi):
return _PyMeshUtils.Mesh_get_face_adjacent_voxels(self, fi)
def get_voxel_adjacent_faces(self, Vi):
return _PyMeshUtils.Mesh_get_voxel_adjacent_faces(self, Vi)
def get_voxel_adjacent_voxels(self, Vi):
return _PyMeshUtils.Mesh_get_voxel_adjacent_voxels(self, Vi)
def has_attribute(self, attr_name):
return _PyMeshUtils.Mesh_has_attribute(self, attr_name)
def add_attribute(self, attr_name):
return _PyMeshUtils.Mesh_add_attribute(self, attr_name)
def remove_attribute(self, attr_name):
return _PyMeshUtils.Mesh_remove_attribute(self, attr_name)
def get_attribute(self, *args):
return _PyMeshUtils.Mesh_get_attribute(self, *args)
def set_attribute(self, attr_name, attr_value):
return _PyMeshUtils.Mesh_set_attribute(self, attr_name, attr_value)
def get_attribute_names(self):
return _PyMeshUtils.Mesh_get_attribute_names(self)
Mesh_swigregister = _PyMeshUtils.Mesh_swigregister
Mesh_swigregister(Mesh)
def convert_to_vertex_attribute(mesh, attribute):
return _PyMeshUtils.convert_to_vertex_attribute(mesh, attribute)
convert_to_vertex_attribute = _PyMeshUtils.convert_to_vertex_attribute
def convert_to_vertex_attribute_from_name(mesh, attribute_name):
return _PyMeshUtils.convert_to_vertex_attribute_from_name(mesh, attribute_name)
convert_to_vertex_attribute_from_name = _PyMeshUtils.convert_to_vertex_attribute_from_name
def convert_to_face_attribute(mesh, attribute):
return _PyMeshUtils.convert_to_face_attribute(mesh, attribute)
convert_to_face_attribute = _PyMeshUtils.convert_to_face_attribute
def convert_to_face_attribute_from_name(mesh, attribute_name):
return _PyMeshUtils.convert_to_face_attribute_from_name(mesh, attribute_name)
convert_to_face_attribute_from_name = _PyMeshUtils.convert_to_face_attribute_from_name
def convert_to_voxel_attribute(mesh, attribute):
return _PyMeshUtils.convert_to_voxel_attribute(mesh, attribute)
convert_to_voxel_attribute = _PyMeshUtils.convert_to_voxel_attribute
def convert_to_voxel_attribute_from_name(mesh, attribute_name):
return _PyMeshUtils.convert_to_voxel_attribute_from_name(mesh, attribute_name)
convert_to_voxel_attribute_from_name = _PyMeshUtils.convert_to_voxel_attribute_from_name
class TriangleMetric(_object):
__swig_setmethods__ = {}
__setattr__ = lambda self, name, value: _swig_setattr(self, TriangleMetric, name, value)
__swig_getmethods__ = {}
__getattr__ = lambda self, name: _swig_getattr(self, TriangleMetric, name)
__repr__ = _swig_repr
def __init__(self):
this = _PyMeshUtils.new_TriangleMetric()
try:
self.this.append(this)
except __builtin__.Exception:
self.this = this
def update(self, p1, p2, p3, p4, p5, p6):
return _PyMeshUtils.TriangleMetric_update(self, p1, p2, p3, p4, p5, p6)
def getClosestPts(self, p1, p2):
return _PyMeshUtils.TriangleMetric_getClosestPts(self, p1, p2)
def getDistance(self):
return _PyMeshUtils.TriangleMetric_getDistance(self)
if _newclass:
test = staticmethod(_PyMeshUtils.TriangleMetric_test)
else:
test = _PyMeshUtils.TriangleMetric_test
if _newclass:
setUseLinCanny = staticmethod(_PyMeshUtils.TriangleMetric_setUseLinCanny)
else:
setUseLinCanny = _PyMeshUtils.TriangleMetric_setUseLinCanny
__swig_destroy__ = _PyMeshUtils.delete_TriangleMetric
__del__ = lambda self: None
TriangleMetric_swigregister = _PyMeshUtils.TriangleMetric_swigregister
TriangleMetric_swigregister(TriangleMetric)
def TriangleMetric_test():
return _PyMeshUtils.TriangleMetric_test()
TriangleMetric_test = _PyMeshUtils.TriangleMetric_test
def TriangleMetric_setUseLinCanny(v):
return _PyMeshUtils.TriangleMetric_setUseLinCanny(v)
TriangleMetric_setUseLinCanny = _PyMeshUtils.TriangleMetric_setUseLinCanny
class ObtuseTriangleRemoval(_object):
__swig_setmethods__ = {}
__setattr__ = lambda self, name, value: _swig_setattr(self, ObtuseTriangleRemoval, name, value)
__swig_getmethods__ = {}
__getattr__ = lambda self, name: _swig_getattr(self, ObtuseTriangleRemoval, name)
__repr__ = _swig_repr
def __init__(self, vertices, faces):
this = _PyMeshUtils.new_ObtuseTriangleRemoval(vertices, faces)
try:
self.this.append(this)
except __builtin__.Exception:
self.this = this
def run(self, max_angle_allowed, max_iterations=1):
return _PyMeshUtils.ObtuseTriangleRemoval_run(self, max_angle_allowed, max_iterations)
def get_vertices(self):
return _PyMeshUtils.ObtuseTriangleRemoval_get_vertices(self)
def get_faces(self):
return _PyMeshUtils.ObtuseTriangleRemoval_get_faces(self)
__swig_destroy__ = _PyMeshUtils.delete_ObtuseTriangleRemoval
__del__ = lambda self: None
ObtuseTriangleRemoval_swigregister = _PyMeshUtils.ObtuseTriangleRemoval_swigregister
ObtuseTriangleRemoval_swigregister(ObtuseTriangleRemoval)
class ShortEdgeRemoval(_object):
__swig_setmethods__ = {}
__setattr__ = lambda self, name, value: _swig_setattr(self, ShortEdgeRemoval, name, value)
__swig_getmethods__ = {}
__getattr__ = lambda self, name: _swig_getattr(self, ShortEdgeRemoval, name)
__repr__ = _swig_repr
def __init__(self, vertices, faces):
this = _PyMeshUtils.new_ShortEdgeRemoval(vertices, faces)
try:
self.this.append(this)
except __builtin__.Exception:
self.this = this
def set_importance(self, importance):
return _PyMeshUtils.ShortEdgeRemoval_set_importance(self, importance)
def run(self, threshold):
return _PyMeshUtils.ShortEdgeRemoval_run(self, threshold)
def get_vertices(self):
return _PyMeshUtils.ShortEdgeRemoval_get_vertices(self)
def get_faces(self):
return _PyMeshUtils.ShortEdgeRemoval_get_faces(self)
def get_face_indices(self):
return _PyMeshUtils.ShortEdgeRemoval_get_face_indices(self)
__swig_destroy__ = _PyMeshUtils.delete_ShortEdgeRemoval
__del__ = lambda self: None
ShortEdgeRemoval_swigregister = _PyMeshUtils.ShortEdgeRemoval_swigregister
ShortEdgeRemoval_swigregister(ShortEdgeRemoval)
def extract_exterior_faces(voxels):
return _PyMeshUtils.extract_exterior_faces(voxels)
extract_exterior_faces = _PyMeshUtils.extract_exterior_faces
class MeshSeparator(_object):
__swig_setmethods__ = {}
__setattr__ = lambda self, name, value: _swig_setattr(self, MeshSeparator, name, value)
__swig_getmethods__ = {}
__getattr__ = lambda self, name: _swig_getattr(self, MeshSeparator, name)
__repr__ = _swig_repr
def __init__(self, elements):
this = _PyMeshUtils.new_MeshSeparator(elements)
try:
self.this.append(this)
except __builtin__.Exception:
self.this = this
VERTEX = _PyMeshUtils.MeshSeparator_VERTEX
FACE = _PyMeshUtils.MeshSeparator_FACE
VOXEL = _PyMeshUtils.MeshSeparator_VOXEL
def set_connectivity_type(self, connectivity):
return _PyMeshUtils.MeshSeparator_set_connectivity_type(self, connectivity)
def separate(self):
return _PyMeshUtils.MeshSeparator_separate(self)
def get_component(self, i):
return _PyMeshUtils.MeshSeparator_get_component(self, i)
def get_sources(self, i):
return _PyMeshUtils.MeshSeparator_get_sources(self, i)
def clear(self):
return _PyMeshUtils.MeshSeparator_clear(self)
__swig_destroy__ = _PyMeshUtils.delete_MeshSeparator
__del__ = lambda self: None
MeshSeparator_swigregister = _PyMeshUtils.MeshSeparator_swigregister
MeshSeparator_swigregister(MeshSeparator)
class Boundary(_object):
__swig_setmethods__ = {}
__setattr__ = lambda self, name, value: _swig_setattr(self, Boundary, name, value)
__swig_getmethods__ = {}
__getattr__ = lambda self, name: _swig_getattr(self, Boundary, name)
def __init__(self, *args, **kwargs):
raise AttributeError("No constructor defined - class is abstract")
__repr__ = _swig_repr
if _newclass:
extract_surface_boundary = staticmethod(_PyMeshUtils.Boundary_extract_surface_boundary)
else:
extract_surface_boundary = _PyMeshUtils.Boundary_extract_surface_boundary
if _newclass:
extract_surface_boundary_raw = staticmethod(_PyMeshUtils.Boundary_extract_surface_boundary_raw)
else:
extract_surface_boundary_raw = _PyMeshUtils.Boundary_extract_surface_boundary_raw
if _newclass:
extract_volume_boundary = staticmethod(_PyMeshUtils.Boundary_extract_volume_boundary)
else:
extract_volume_boundary = _PyMeshUtils.Boundary_extract_volume_boundary
if _newclass:
extract_volume_boundary_raw = staticmethod(_PyMeshUtils.Boundary_extract_volume_boundary_raw)
else:
extract_volume_boundary_raw = _PyMeshUtils.Boundary_extract_volume_boundary_raw
def get_num_boundaries(self):
return _PyMeshUtils.Boundary_get_num_boundaries(self)
def get_boundaries(self):
return _PyMeshUtils.Boundary_get_boundaries(self)
def get_boundary(self, bi):
return _PyMeshUtils.Boundary_get_boundary(self, bi)
def get_boundary_element(self, bi):
return _PyMeshUtils.Boundary_get_boundary_element(self, bi)
def get_num_boundary_nodes(self):
return _PyMeshUtils.Boundary_get_num_boundary_nodes(self)
def get_boundary_nodes(self):
return _PyMeshUtils.Boundary_get_boundary_nodes(self)
__swig_destroy__ = _PyMeshUtils.delete_Boundary
__del__ = lambda self: None
Boundary_swigregister = _PyMeshUtils.Boundary_swigregister
Boundary_swigregister(Boundary)
def Boundary_extract_surface_boundary(mesh):
return _PyMeshUtils.Boundary_extract_surface_boundary(mesh)
Boundary_extract_surface_boundary = _PyMeshUtils.Boundary_extract_surface_boundary
def Boundary_extract_surface_boundary_raw(vertices, faces):
return _PyMeshUtils.Boundary_extract_surface_boundary_raw(vertices, faces)
Boundary_extract_surface_boundary_raw = _PyMeshUtils.Boundary_extract_surface_boundary_raw
def Boundary_extract_volume_boundary(mesh):
return _PyMeshUtils.Boundary_extract_volume_boundary(mesh)
Boundary_extract_volume_boundary = _PyMeshUtils.Boundary_extract_volume_boundary
def Boundary_extract_volume_boundary_raw(vertices, voxels):
return _PyMeshUtils.Boundary_extract_volume_boundary_raw(vertices, voxels)
Boundary_extract_volume_boundary_raw = _PyMeshUtils.Boundary_extract_volume_boundary_raw
class PointLocator(_object):
__swig_setmethods__ = {}
__setattr__ = lambda self, name, value: _swig_setattr(self, PointLocator, name, value)
__swig_getmethods__ = {}
__getattr__ = lambda self, name: _swig_getattr(self, PointLocator, name)
__repr__ = _swig_repr
def __init__(self, mesh):
this = _PyMeshUtils.new_PointLocator(mesh)
try:
self.this.append(this)
except __builtin__.Exception:
self.this = this
def locate(self, points):
return _PyMeshUtils.PointLocator_locate(self, points)
def get_enclosing_voxels(self):
return _PyMeshUtils.PointLocator_get_enclosing_voxels(self)
def get_barycentric_coords(self):
return _PyMeshUtils.PointLocator_get_barycentric_coords(self)
def clear(self):
return _PyMeshUtils.PointLocator_clear(self)
__swig_destroy__ = _PyMeshUtils.delete_PointLocator
__del__ = lambda self: None
PointLocator_swigregister = _PyMeshUtils.PointLocator_swigregister
PointLocator_swigregister(PointLocator)
class Subdivision(_object):
__swig_setmethods__ = {}
__setattr__ = lambda self, name, value: _swig_setattr(self, Subdivision, name, value)
__swig_getmethods__ = {}
__getattr__ = lambda self, name: _swig_getattr(self, Subdivision, name)
def __init__(self, *args, **kwargs):
raise AttributeError("No constructor defined - class is abstract")
__repr__ = _swig_repr
if _newclass:
create = staticmethod(_PyMeshUtils.Subdivision_create)
else:
create = _PyMeshUtils.Subdivision_create
__swig_destroy__ = _PyMeshUtils.delete_Subdivision
__del__ = lambda self: None
def subdivide(self, vertices, faces, num_iterations):
return _PyMeshUtils.Subdivision_subdivide(self, vertices, faces, num_iterations)
def get_subdivision_matrices(self):
return _PyMeshUtils.Subdivision_get_subdivision_matrices(self)
def get_vertices(self):
return _PyMeshUtils.Subdivision_get_vertices(self)
def get_faces(self):
return _PyMeshUtils.Subdivision_get_faces(self)
def get_face_indices(self):
return _PyMeshUtils.Subdivision_get_face_indices(self)
def get_num_vertices(self):
return _PyMeshUtils.Subdivision_get_num_vertices(self)
def get_num_faces(self):
return _PyMeshUtils.Subdivision_get_num_faces(self)
Subdivision_swigregister = _PyMeshUtils.Subdivision_swigregister
Subdivision_swigregister(Subdivision)
def Subdivision_create(type):
return _PyMeshUtils.Subdivision_create(type)
Subdivision_create = _PyMeshUtils.Subdivision_create
class DuplicatedVertexRemoval(_object):
__swig_setmethods__ = {}
__setattr__ = lambda self, name, value: _swig_setattr(self, DuplicatedVertexRemoval, name, value)
__swig_getmethods__ = {}
__getattr__ = lambda self, name: _swig_getattr(self, DuplicatedVertexRemoval, name)
__repr__ = _swig_repr
def __init__(self, vertices, faces):
this = _PyMeshUtils.new_DuplicatedVertexRemoval(vertices, faces)
try:
self.this.append(this)
except __builtin__.Exception:
self.this = this
def run(self, tol):
return _PyMeshUtils.DuplicatedVertexRemoval_run(self, tol)
def get_vertices(self):
return _PyMeshUtils.DuplicatedVertexRemoval_get_vertices(self)
def get_faces(self):
return _PyMeshUtils.DuplicatedVertexRemoval_get_faces(self)
def set_importance_level(self, level):
return _PyMeshUtils.DuplicatedVertexRemoval_set_importance_level(self, level)
def get_index_map(self):
return _PyMeshUtils.DuplicatedVertexRemoval_get_index_map(self)
__swig_destroy__ = _PyMeshUtils.delete_DuplicatedVertexRemoval
__del__ = lambda self: None
DuplicatedVertexRemoval_swigregister = _PyMeshUtils.DuplicatedVertexRemoval_swigregister
DuplicatedVertexRemoval_swigregister(DuplicatedVertexRemoval)
class IsolatedVertexRemoval(_object):
__swig_setmethods__ = {}
__setattr__ = lambda self, name, value: _swig_setattr(self, IsolatedVertexRemoval, name, value)
__swig_getmethods__ = {}
__getattr__ = lambda self, name: _swig_getattr(self, IsolatedVertexRemoval, name)
__repr__ = _swig_repr
def __init__(self, vertices, faces):
this = _PyMeshUtils.new_IsolatedVertexRemoval(vertices, faces)
try:
self.this.append(this)
except __builtin__.Exception:
self.this = this
def run(self):
return _PyMeshUtils.IsolatedVertexRemoval_run(self)
def get_vertices(self):
return _PyMeshUtils.IsolatedVertexRemoval_get_vertices(self)
def get_faces(self):
return _PyMeshUtils.IsolatedVertexRemoval_get_faces(self)
def get_ori_vertex_indices(self):
return _PyMeshUtils.IsolatedVertexRemoval_get_ori_vertex_indices(self)
__swig_destroy__ = _PyMeshUtils.delete_IsolatedVertexRemoval
__del__ = lambda self: None
IsolatedVertexRemoval_swigregister = _PyMeshUtils.IsolatedVertexRemoval_swigregister
IsolatedVertexRemoval_swigregister(IsolatedVertexRemoval)
class LongEdgeRemoval(_object):
__swig_setmethods__ = {}
__setattr__ = lambda self, name, value: _swig_setattr(self, LongEdgeRemoval, name, value)
__swig_getmethods__ = {}
__getattr__ = lambda self, name: _swig_getattr(self, LongEdgeRemoval, name)
__repr__ = _swig_repr
def __init__(self, vertices, faces):
this = _PyMeshUtils.new_LongEdgeRemoval(vertices, faces)
try:
self.this.append(this)
except __builtin__.Exception:
self.this = this
def run(self, max_length, recursive=True):
return _PyMeshUtils.LongEdgeRemoval_run(self, max_length, recursive)
def get_vertices(self):
return _PyMeshUtils.LongEdgeRemoval_get_vertices(self)
def get_faces(self):
return _PyMeshUtils.LongEdgeRemoval_get_faces(self)
__swig_destroy__ = _PyMeshUtils.delete_LongEdgeRemoval
__del__ = lambda self: None
LongEdgeRemoval_swigregister = _PyMeshUtils.LongEdgeRemoval_swigregister
LongEdgeRemoval_swigregister(LongEdgeRemoval)
class FinFaceRemoval(_object):
__swig_setmethods__ = {}
__setattr__ = lambda self, name, value: _swig_setattr(self, FinFaceRemoval, name, value)
__swig_getmethods__ = {}
__getattr__ = lambda self, name: _swig_getattr(self, FinFaceRemoval, name)
__repr__ = _swig_repr
def __init__(self, vertices, faces):
this = _PyMeshUtils.new_FinFaceRemoval(vertices, faces)
try:
self.this.append(this)
except __builtin__.Exception:
self.this = this
def set_fins_only(self):
return _PyMeshUtils.FinFaceRemoval_set_fins_only(self)
def run(self):
return _PyMeshUtils.FinFaceRemoval_run(self)
def get_vertices(self):
return _PyMeshUtils.FinFaceRemoval_get_vertices(self)
def get_faces(self):
return _PyMeshUtils.FinFaceRemoval_get_faces(self)
def get_face_indices(self):
return _PyMeshUtils.FinFaceRemoval_get_face_indices(self)
__swig_destroy__ = _PyMeshUtils.delete_FinFaceRemoval
__del__ = lambda self: None
FinFaceRemoval_swigregister = _PyMeshUtils.FinFaceRemoval_swigregister
FinFaceRemoval_swigregister(FinFaceRemoval)
class MeshChecker(_object):
__swig_setmethods__ = {}
__setattr__ = lambda self, name, value: _swig_setattr(self, MeshChecker, name, value)
__swig_getmethods__ = {}
__getattr__ = lambda self, name: _swig_getattr(self, MeshChecker, name)
__repr__ = _swig_repr
def __init__(self, vertices, faces, voxels):
this = _PyMeshUtils.new_MeshChecker(vertices, faces, voxels)
try:
self.this.append(this)
except __builtin__.Exception:
self.this = this
def is_vertex_manifold(self):
return _PyMeshUtils.MeshChecker_is_vertex_manifold(self)
def is_edge_manifold(self):
return _PyMeshUtils.MeshChecker_is_edge_manifold(self)
def is_closed(self):
return _PyMeshUtils.MeshChecker_is_closed(self)
def has_edge_with_odd_adj_faces(self):
return _PyMeshUtils.MeshChecker_has_edge_with_odd_adj_faces(self)
def is_oriented(self):
return _PyMeshUtils.MeshChecker_is_oriented(self)
def has_complex_boundary(self):
return _PyMeshUtils.MeshChecker_has_complex_boundary(self)
def get_num_boundary_edges(self):
return _PyMeshUtils.MeshChecker_get_num_boundary_edges(self)
def get_num_boundary_loops(self):
return _PyMeshUtils.MeshChecker_get_num_boundary_loops(self)
def get_boundary_edges(self):
return _PyMeshUtils.MeshChecker_get_boundary_edges(self)
def get_boundary_loops(self):
return _PyMeshUtils.MeshChecker_get_boundary_loops(self)
def get_genus(self):
return _PyMeshUtils.MeshChecker_get_genus(self)
def get_euler_characteristic(self):
return _PyMeshUtils.MeshChecker_get_euler_characteristic(self)
def get_num_connected_components(self):
return _PyMeshUtils.MeshChecker_get_num_connected_components(self)
def get_num_connected_surface_components(self):
return _PyMeshUtils.MeshChecker_get_num_connected_surface_components(self)
def get_num_connected_volume_components(self):
return _PyMeshUtils.MeshChecker_get_num_connected_volume_components(self)
def get_num_isolated_vertices(self):
return _PyMeshUtils.MeshChecker_get_num_isolated_vertices(self)
def get_num_duplicated_faces(self):
return _PyMeshUtils.MeshChecker_get_num_duplicated_faces(self)
def compute_signed_volume_from_surface(self):
return _PyMeshUtils.MeshChecker_compute_signed_volume_from_surface(self)
__swig_destroy__ = _PyMeshUtils.delete_MeshChecker
__del__ = lambda self: None
MeshChecker_swigregister = _PyMeshUtils.MeshChecker_swigregister
MeshChecker_swigregister(MeshChecker)
class DegeneratedTriangleRemoval(_object):
__swig_setmethods__ = {}
__setattr__ = lambda self, name, value: _swig_setattr(self, DegeneratedTriangleRemoval, name, value)
__swig_getmethods__ = {}
__getattr__ = lambda self, name: _swig_getattr(self, DegeneratedTriangleRemoval, name)
__repr__ = _swig_repr
def __init__(self, vertices, faces):
this = _PyMeshUtils.new_DegeneratedTriangleRemoval(vertices, faces)
try:
self.this.append(this)
except __builtin__.Exception:
self.this = this
def run(self, num_iteraitons=5):
return _PyMeshUtils.DegeneratedTriangleRemoval_run(self, num_iteraitons)
def get_vertices(self):
return _PyMeshUtils.DegeneratedTriangleRemoval_get_vertices(self)
def get_faces(self):
return _PyMeshUtils.DegeneratedTriangleRemoval_get_faces(self)
def get_ori_face_indices(self):
return _PyMeshUtils.DegeneratedTriangleRemoval_get_ori_face_indices(self)
__swig_destroy__ = _PyMeshUtils.delete_DegeneratedTriangleRemoval
__del__ = lambda self: None
DegeneratedTriangleRemoval_swigregister = _PyMeshUtils.DegeneratedTriangleRemoval_swigregister
DegeneratedTriangleRemoval_swigregister(DegeneratedTriangleRemoval)
class vector_size_t(_object):
__swig_setmethods__ = {}
__setattr__ = lambda self, name, value: _swig_setattr(self, vector_size_t, name, value)
__swig_getmethods__ = {}
__getattr__ = lambda self, name: _swig_getattr(self, vector_size_t, name)
__repr__ = _swig_repr
def iterator(self):
return _PyMeshUtils.vector_size_t_iterator(self)
def __iter__(self):
return self.iterator()
def __nonzero__(self):
return _PyMeshUtils.vector_size_t___nonzero__(self)
def __bool__(self):
return _PyMeshUtils.vector_size_t___bool__(self)
def __len__(self):
return _PyMeshUtils.vector_size_t___len__(self)
def __getslice__(self, i, j):
return _PyMeshUtils.vector_size_t___getslice__(self, i, j)
def __setslice__(self, *args):
return _PyMeshUtils.vector_size_t___setslice__(self, *args)
def __delslice__(self, i, j):
return _PyMeshUtils.vector_size_t___delslice__(self, i, j)
def __delitem__(self, *args):
return _PyMeshUtils.vector_size_t___delitem__(self, *args)
def __getitem__(self, *args):
return _PyMeshUtils.vector_size_t___getitem__(self, *args)
def __setitem__(self, *args):
return _PyMeshUtils.vector_size_t___setitem__(self, *args)
def pop(self):
return _PyMeshUtils.vector_size_t_pop(self)
def append(self, x):
return _PyMeshUtils.vector_size_t_append(self, x)
def empty(self):
return _PyMeshUtils.vector_size_t_empty(self)
def size(self):
return _PyMeshUtils.vector_size_t_size(self)
def swap(self, v):
return _PyMeshUtils.vector_size_t_swap(self, v)
def begin(self):
return _PyMeshUtils.vector_size_t_begin(self)
def end(self):
return _PyMeshUtils.vector_size_t_end(self)
def rbegin(self):
return _PyMeshUtils.vector_size_t_rbegin(self)
def rend(self):
return _PyMeshUtils.vector_size_t_rend(self)
def clear(self):
return _PyMeshUtils.vector_size_t_clear(self)
def get_allocator(self):
return _PyMeshUtils.vector_size_t_get_allocator(self)
def pop_back(self):
return _PyMeshUtils.vector_size_t_pop_back(self)
def erase(self, *args):
return _PyMeshUtils.vector_size_t_erase(self, *args)
def __init__(self, *args):
this = _PyMeshUtils.new_vector_size_t(*args)
try:
self.this.append(this)
except __builtin__.Exception:
self.this = this
def push_back(self, x):
return _PyMeshUtils.vector_size_t_push_back(self, x)
def front(self):
return _PyMeshUtils.vector_size_t_front(self)
def back(self):
return _PyMeshUtils.vector_size_t_back(self)
def assign(self, n, x):
return _PyMeshUtils.vector_size_t_assign(self, n, x)
def resize(self, *args):
return _PyMeshUtils.vector_size_t_resize(self, *args)
def insert(self, *args):
return _PyMeshUtils.vector_size_t_insert(self, *args)
def reserve(self, n):
return _PyMeshUtils.vector_size_t_reserve(self, n)
def capacity(self):
return _PyMeshUtils.vector_size_t_capacity(self)
__swig_destroy__ = _PyMeshUtils.delete_vector_size_t
__del__ = lambda self: None
vector_size_t_swigregister = _PyMeshUtils.vector_size_t_swigregister
vector_size_t_swigregister(vector_size_t)
def is_colinear_2D(v0, v1, v2):
return _PyMeshUtils.is_colinear_2D(v0, v1, v2)
is_colinear_2D = _PyMeshUtils.is_colinear_2D
def is_colinear_3D(v0, v1, v2):
return _PyMeshUtils.is_colinear_3D(v0, v1, v2)
is_colinear_3D = _PyMeshUtils.is_colinear_3D
def get_degenerated_faces(vertices, faces):
return _PyMeshUtils.get_degenerated_faces(vertices, faces)
get_degenerated_faces = _PyMeshUtils.get_degenerated_faces
# This file is compatible with both classic and new-style classes.
|
993,887 | aaf09570f7efa4fe18c8860e106eb008e90c551f | import pytest
from solution import validate
data_ok = [
('Timmy', 'password', 'Successfully Logged in!'),
('Alice', 'alice', 'Successfully Logged in!'),
]
@pytest.mark.parametrize(
"username, password, result", data_ok
)
def test_should_succesfully_login(username, password, result):
assert validate(username, password) == result
data_pwd_wrong = [
('Timmy', 'h4x0r', 'Wrong username or password!'),
]
@pytest.mark.parametrize(
"username, password, result", data_pwd_wrong
)
def test_should_detect_wrong_password(username, password, result):
assert validate(username, password) == result
data_injected_code = [
('Timmy', 'password"||""=="', 'Wrong username or password!'),
('Admin', 'gs5bw"||1==1//', 'Wrong username or password!'),
]
@pytest.mark.parametrize(
"username, password, result", data_injected_code
)
def test_should_fail_to_login_due_to_injected_code(username, password, result):
assert validate(username, password) == result
|
993,888 | 3d25575916bf92ec19cfeaf4833255f8a7b4f5eb |
leagues = {}
base_url = 'http://www.football-data.co.uk/mmz4281/'
seasons = ['1819', '1718', '1617', '1516', '1415', '1314']
for league in leagues:
if leagues[league]['hasSeasons']:
for season in seasons:
url = '{0}{1}/{2}.csv'.format(base_url, season,
leagues[league]['slug'])
else:
url = leagues[league]['url']
|
993,889 | 83cc94ab18287e595b7504692f384704f880ef46 | #!/usr/bin/python
# Route Time
# Displays leave and return time for city routes
# Parts needed LCD Pi Plate, 4 7 segment displays
# ! Imports
import Adafruit_CharLCD as LCD
from Adafruit_LED_Backpack import SevenSegment
import Adafruit_Trellis
# Trellis setup
# Momentary mode LED lights only when pressed
# Latching mode LED lights on/off when pressed
MOMENTARY = 0
LATCHING = 1
# Will need to set this mode by what data is being entered
MODE = MOMENTARY
matrix0 = Adafruit_Trellis.Adafruit_Trellis()
trellis = Adafruit_Trellis.Adafruit_TrellisSet(matrix0)
# Number of Trellis Boards
NUMTRELLIS = 1
numKeys = NUMTRELLIS * 1
I2C_BUS = 1
trellis.begin((0x75, I2C_BUS))
# Initialize 7-Segment Matrix Displays
display1 = SevenSegment.SevenSegment(address=0x71)
display2 = SevenSegment.SevenSegment(address=0x72)
display3 = SevenSegment.SevenSegment(address=0x73)
display4 = SevenSegment.SevenSegment(address=0x74)
display1.begin()
display2.begin()
display3.begin()
display4.begin()
colon = True
# Initialize the LCD using the pins
lcd = LCD.Adafruit_CharLCDPlate()
# Custom characters used on display for
# Directional buttons Up(1), Down(3), Left(4), Right(2)
# As well as Select(5)
# In order as shown
# ! Custom Characters
lcd.create_char(1, [0,0,4,14,31,0,0,0])
lcd.create_char(2, [0,8,12,14,12,8,0,0])
lcd.create_char(3, [0,0,31,14,4,0,0,0])
lcd.create_char(4, [0,2,6,14,6,2,0,0])
lcd.create_char(5, [0,0,10,0,17,14,0,0])
# Clear the LCD screen, set the color to blue and set the initial message
lcd.clear()
lcd.set_color(0.0, 0.0, 1.0)
lcd.message('Leave/Return\n\x01\x03 C1 & \x02\x04 C2 \x05 Toggle')
# Wait for button press to set times
# Then enter text via USB or wireless keyboard
# LCD Plate directional buttons for Leave/Return times
# Select button currently unused
# Green color for Leave times, Red for Return while clear LCD before entry
# And have LCD display cursor and blink menacingly
# ! Main Loop
while True:
# Set Select to swap from Momentary to Latching
if lcd.is_pressed(LCD.SELECT):
lcd.clear()
lcd.blink(True)
lcd.show_cursor(True)
lcd.set_color(1.0, 1.0, 0.0)
MODE = LATCHING
lcd.message('Set to %s') % MODE
else:
MODE = MOMENTARY
if lcd.is_pressed(LCD.UP):
lcd.clear()
lcd.blink(True)
lcd.show_cursor(True)
lcd.set_color(0.0, 1.0, 0.0)
lcd.message('Enter C1 Start Time\n')
display1.clear()
display1.set_colon(True)
c1up = float(raw_input())
lcd.message(c1up)
display1.print_float(c1up, 0)
display1.write_display()
if lcd.is_pressed(LCD.RIGHT):
lcd.clear()
lcd.blink(True)
lcd.show_cursor(True)
lcd.set_color(0.0, 1.0, 0.0)
c2right = float(raw_input('C2 Start Time\n'))
display3.clear()
display3.print_float(c2right,0)
display3.set_colon(True)
display3.write_display()
lcd.message('c2right')
if lcd.is_pressed(LCD.DOWN):
lcd.clear()
lcd.blink(True)
lcd.show_cursor(True)
lcd.set_color(1.0, 0.0, 0.0)
c1down = float(raw_input('C1 Return Time\n'))
display2.clear()
display2.print_float(c1down,0)
display2.set_colon(True)
display2.write_display()
lcd.message('c1down')
if lcd.is_pressed(LCD.LEFT):
lcd.clear()
lcd.blink(True)
lcd.show_cursor(True)
lcd.set_color(1.0, 0.0, 0.0)
c2left = float(raw_input('C2 Return Time\n'))
display4.clear()
display4.print_float(c2left,0)
display4.set_colon(True)
display4.write_display()
lcd.message('c2left')
time.sleep(0.03)
if MODE == MOMENTARY:
# If a button was just pressed or released...
if trellis.readSwitches():
# go through every button
for i in range(numKeys):
# if it was pressed, turn it on
if trellis.justPressed(i):
print 'v{0}'.format(i)
trellis.setLED(i)
# if it was released, turn it off
if trellis.justReleased(i):
print '^{0}'.format(i)
trellis.clrLED(i)
# tell the trellis to set the LEDs we requested
trellis.writeDisplay()
if MODE == LATCHING:
# If a button was just pressed or released...
if trellis.readSwitches():
# go through every button
for i in range(numKeys):
# if it was pressed...
if trellis.justPressed(i):
print 'v{0}'.format(i)
# Alternate the LED
if trellis.isLED(i):
trellis.clrLED(i)
else:
trellis.setLED(i)
# tell the trellis to set the LEDs we requested
trellis.writeDisplay()
|
993,890 | 71c24a2a51c521728caea6e508757544a71fc8ab |
input = "$0859: .BYTE $41,$20,$27,$4C,$49,$54,$54,$4C\n$0861: .BYTE $45,$20,$53,$4F,$4D,$45,$54,$48\n$0869: .BYTE $49,$4E,$47,$27,$20,$46,$52,$4F\n$0871: .BYTE $4D,$20,$49,$52,$49,$44,$49,$53\n$0879: .BYTE $2D,$41,$4C,$50,$48,$41\n"
bytes = [b for l in input.split('\n') for b in l[33:65].strip().split(',') if b != ""]
chars = [chr(int("0x" + b[-2:], 16)) for b in bytes]
n = 8
lines = [bytes[i:i + n] for i in range(0, len(bytes), n)]
comments = [chars[i:i + n] for i in range(0, len(chars), n)]
print(lines)
print(comments)
for i, l in enumerate(lines):
print((" " * 8) + "; " + " ".join(comments[i]))
print((" " * 8) + ".BYTE " + (",".join(l)))
print((" " * 8) + ".TEXT \"" + "".join(chars) + "\"")
|
993,891 | 3fb93b782f6afc9049f9f2a5b90b251f7611f25b | import csv
import requests
import json
import re
import logging
from bs4 import BeautifulSoup
from six.moves.html_parser import HTMLParser
from functools import reduce
from datetime import datetime
from dateutil.parser import parse
import pandas as pd
now = datetime.now().isoformat().replace(':', '').split('.')[0]
academic_affiliation_file = "academicAffiliationMap.csv"
csvfile = open(academic_affiliation_file, 'r')
academicMap = [{k: v for k, v in row.items()} for row in csv.DictReader(
csvfile, delimiter='|', skipinitialspace=True)]
csvfile.close()
# Original Query
#query = '{"filter":{"document_type":{"$in":["bookreview","book_review","editorial","leg_brief","leg_pub","memo","nepc_review","pol_brief","re_brief"]}}}&page_size=0'
# Error Query
query = '{"filter":{"document_type":{"$in":["","bookreview","book_review","editorial","leg_brief","leg_pub","memo","nepc_review","pol_brief","re_brief"]},"samvera_url":{"$exists":false}}}&page_size=0'
api_url = 'https://libapps.colorado.edu/api/catalog/data/catalog/cuscholar-final-2019-12-20.json?query=' + query
# api_url = 'https://libapps.colorado.edu/api/catalog/data/catalog/cuscholar-final.json?query={"filter":{"document_type":"presentation"}}&page_size=0'
# confernece
# api_url = 'https://libapps.colorado.edu/api/catalog/data/catalog/cuscholar-final.json?query={"filter":{"document_type":"conference"}}&page_size=0'
headers = {'Content-Type': 'application/json'}
csv_divider = "|~|"
# 'description','date_created',
csv_headers = ['title', 'date created', 'resource type', 'creator', 'contributor', 'keyword', 'license', 'rights statement', 'publisher',
'subject', 'language', 'identifier', 'location', 'related_url', 'bibliographic_citation', 'source', 'abstract', 'academic_affiliation',
'editor', 'in_series', 'additional_information', 'alt_title', 'date_available', 'date_issued',
'doi', 'file_extent', 'file_format', 'embargo_reason', 'peerreviewed', 'replaces', 'language', 'admin_set_id', 'visibility', 'files']
# 'conference_location', 'conference_name',
defaults = {'language': 'http://id.loc.gov/vocabulary/iso639-2/eng',
'rights statement': 'http://rightsstatements.org/vocab/InC/1.0/',
'admin_set_id': 'admin_set/default',
'visibility': 'open',
}
# Test admin set
# 'admin_set_id':'qb98mf449',
# Production
# <option data-sharing="true" value="j098zb08p">Reports</option>
# <option data-sharing="true" value="dn39x152w">Book</option>
# <option data-sharing="true" value="pr76f341v">Book Chapter</option>
# <option data-sharing="true" value="c534fn941">Conference Proceedings</option>
# <option data-sharing="true" value="0p0966899">Presentation</option>
# <option data-sharing="true" value="j6731378s">Dataset</option>
# <option data-sharing="true" value="m326m172d">Undergraduate Honors Thesis</option>
# <option data-sharing="true" value="nv935286k">Article</option>
# <option data-sharing="true" value="k643b116n">Graduate Thesis Or Dissertation</option>
# <option data-sharing="true" value="admin_set/default">Default Admin Set</option>
class Error(Exception):
"""Base class for other exceptions"""
pass
class UnableToDownload(Error):
"""Raised when unable to download"""
pass
def deep_get(_dict, keys, default=None):
"""
Deep get on python dictionary. Key is in dot notation.
Returns value if found. Default returned if not found.
Default can be set to another deep_get function.
"""
keys = keys.split('.')
def _reducer(d, key):
if isinstance(d, dict):
return d.get(key, default)
return default
return reduce(_reducer, keys, _dict)
def clean_format_name(names, divider="|~|"):
names = re.sub(', and ', ',', names, re.IGNORECASE)
names = re.sub(' and ', ',', names, re.IGNORECASE)
newName = []
for name in names.split(','):
tmp = name.split(" ")
try:
tmp.remove('')
except:
pass
if tmp:
newName.append("{0}, {1}".format(tmp[-1], " ".join(tmp[:-1])))
return divider.join(newName)
def super_sub_script_replace(text, start_tag, stop_tag):
starts = [m.start() for m in re.finditer(start_tag, text)]
stops = [m.start() for m in re.finditer(stop_tag, text)]
SUBnumeric = str.maketrans("0123456789", "₀₁₂₃₄₅₆₇₈₉")
SUBalhpa = str.maketrans("abcdeijoqruvwxyzAEIJORUVX-+",
"ₐᵦ𝒸𝒹ₑᵢⱼₒᵩᵣᵤᵥ𝓌ₓᵧ𝓏ₐₑᵢⱼₒᵣᵤᵥₓ₋₊")
SUPnumeric = str.maketrans("0123456789", "⁰¹²³⁴⁵⁶⁷⁸⁹")
SUPalhpa = str.maketrans("abcdefghijklmnopqrstuvwxyzABCDEFGHIJKLMNOPQRSTUVWXYZ-+",
"ᵃᵇᶜᵈᵉᶠᵍʰᶦʲᵏˡᵐⁿᵒᵖᑫʳˢᵗᵘᵛʷˣʸᶻᴬᴮᶜᴰᴱᶠᴳᴴᴵᴶᴷᴸᴹᴺᴼᴾQᴿˢᵀᵁⱽᵂˣʸᶻ⁻⁺")
# print(starts, stops)
# print(tag)
search_replace = []
for idx, itm in enumerate(starts, start=0):
# print(starts, stops)
if itm:
soup = BeautifulSoup(text[itm:stops[idx]+6], 'html.parser')
tmp = soup.get_text()
if start_tag == '<sub>':
tmp = tmp.translate(SUBnumeric)
tmp = tmp.translate(SUBalhpa)
search_replace.append((text[itm:stops[idx]+6], tmp))
else:
tmp = tmp.translate(SUPnumeric)
tmp = tmp.translate(SUPalhpa)
search_replace.append((text[itm:stops[idx]+6], tmp))
for sr in search_replace:
text = text.replace(sr[0], sr[1])
# print(search_replace)
return text
def clean_abstract_text(html):
# Super and Sub scripts insertion
text = super_sub_script_replace(html, '<sub>', '</sub>')
text = super_sub_script_replace(text, '<sup>', '</sup>')
# Translate special characters
h = HTMLParser()
text = h.unescape(text)
# Remove all tags remaining
soup = BeautifulSoup(text, 'html.parser')
text = soup.get_text()
return text
def getFiles(itm):
# query = '{"filter":{"front_end_url":"' + \
# itm['front_end_url'] + '","context_key":"' + itm['context_key'] + '"}}'
#url = 'https://libapps.colorado.edu/api/catalog/data/catalog/cuscholar-final-final.json?query='
#req = requests.get(url + query)
#data = req.json()
# if data['count'] == 1:
# itm = data['results'][0]
# # print("update")
files = []
main_file = deep_get(itm, 'data_files.s3.processed.key', default=None)
if main_file.strip():
files.append(
deep_get(itm, 'data_files.s3.processed.key', default=None))
elif main_file.strip() == "":
raise UnableToDownload
else:
raise
alt_file = deep_get(
itm, 'data_files.s3.processed.additional_files', default=[])
alt_file = list(set(alt_file))
return csv_divider.join(files + alt_file)
def articleAcademicAffiliation(itm):
if itm['department'].strip():
val = itm['department']
else:
val = academicAffiliation(itm)
return val
def academicAffiliation(itm):
# csvfile = open(academic_affiliation_file,'r')
# amap = [{k: v for k, v in row.items()} for row in csv.DictReader(csvfile,delimiter='|', skipinitialspace=True)]
academicMap
match = itm['front_end_url'].split('/')[-2]
for aa in academicMap:
if aa['bepress'] == match:
return aa['samvera']
return 'Other'
def graduationYear(itm):
year = itm["publication_date"].split('-')[0]
if not year:
return '9999'
return year
def replaces(itm):
return "{0}|{1}".format(itm['context_key'], itm['front_end_url'])
def pubDateFormat(itm):
try:
mydate = parse(itm["publication_date"])
value = mydate.strftime("%Y-%m-%d")
except:
value = ""
return value
def identifierFormat(itm):
value = ''
if itm['pubmedid'].strip():
value = "PubMed ID: {0}".format(itm['pubmedid'])
if itm['identifier'].strip():
if value:
value = "{0}; {1}".foramt(value, itm['identifier'])
else:
value = itm['identifier']
if itm['external_article_id'].strip():
if value:
value = "{0}; External Arcticle ID: {1}".format(
value, itm['external_article_id'])
else:
value = "External Arcticle ID: {0}".format(
itm['external_article_id'])
return value
def setFileExtent(itm):
value = ''
if itm['fpage'].strip() and itm['lpage'].strip():
value = "{0}-{1}".format(itm['fpage'].strip(), itm['lpage'].strip())
elif itm['fpage'].strip() or itm['lpage'].strip():
value = "{0}{1}".format(
itm['fpage'].strip(), itm['lpage'].strip()).strip()
return value
def eventDate(itm):
value = ''
if itm['conference_dates'].strip():
value = "Event Date: {0}".format(itm['conference_dates'])
return value
def additonal_information(itm):
if itm['source_publication'].strip() and itm['comments'].strip():
value = "{0} - {1}".format(clean_abstract_text(
itm['comments']), itm['source_publication'])
elif itm['comments'].strip():
value = "{0}".format(clean_abstract_text(itm['comments']))
elif itm['source_publication'].strip():
value = "{0}".format(itm['source_publication'])
else:
value = ''
return value
def setResourceType(itm):
if itm['document_type'] == "technical_report":
value = "Technical Report"
elif itm['document_type'] == "working_paper":
value = "Working Paper"
else:
value = 'Other'
return value
def transform(itm):
data_row = dict.fromkeys(csv_headers, '')
data_row.update(defaults)
data_row['title'] = itm['title']
data_row['creator'] = clean_format_name(
itm['authors'], divider=csv_divider)
data_row['abstract'] = clean_abstract_text(itm['abstract'])
# data_row['date created'] = itm['publication_date'].split(' ')[0]
# data_row['date_created'] = itm['publication_date'].split(' ')[0]
data_row['subject'] = csv_divider.join(itm['keywords'])
data_row['keyword'] = csv_divider.join(itm['keywords'])
if not data_row['keyword']:
data_row['keyword'] = 'Keyword'
data_row['academic_affiliation'] = academicAffiliation(itm)
data_row['license'] = itm['distribution_license']
data_row['publisher'] = itm["publisher"]
data_row['identifier'] = identifierFormat(itm)
data_row['related url'] = itm['url']
# data_row['date_available'] = itm["publication_date"].split(' ')[0]
# itm["publication_date"].split('-')[0]
data_row['date_issued'] = pubDateFormat(itm)
data_row['doi'] = itm['doi']
data_row['peerreviewed'] = itm['peer_reviewed']
data_row['replaces'] = replaces(itm)
# Article
# data_row['has_journal'] = itm['source_publication']
# data_row['has_number'] = itm['issnum']
# data_row['has_volume'] = itm['volnum']
# data_row['issn'] = itm['issn']
# data_row['isbn'] = itm['isbn']
data_row['editor'] = itm['editor']
data_row['bibliographic_citation'] = itm['custom_citation']
# data_row['conference_location'] = itm['conference_city']
# data_row['conference_name'] = itm['conference_name']
data_row['additional_information'] = additonal_information(itm)
data_row['file_extent'] = setFileExtent(itm)
data_row['resource type'] = setResourceType(itm)
try:
#data_row['files'] = 'ableToDownload.pdf'
data_row['files'] = getFiles(itm)
except UnableToDownload:
data_row['files'] = 'unableToDownload.pdf'
return data_row
def writeCsvFile(csv_data, error_data, count, wtype="Default"):
now = datetime.now().isoformat().replace(':', '').split('.')[0]
keys = csv_data[0].keys()
filename = 'csv_output/{0}_{1}_dataload_{2}.csv'.format(
now, wtype.replace(' ', '_').lower(), count)
errorfilename = 'csv_output/{0}_{1}_error_{2}.json'.format(
now, wtype.replace(' ', '_').lower(), count)
with open(filename, 'w') as output_file:
dict_writer = csv.DictWriter(output_file, keys)
dict_writer.writeheader()
dict_writer.writerows(csv_data)
if error_data:
with open(errorfilename, 'w') as output_file:
output_file.write(json.dumps(error_data, indent=4))
# Dedup
df = pd.read_csv(filename, sep=",")
df.drop_duplicates(subset=None, inplace=True)
df.to_csv(filename, index=False)
def loadItems(work_type="graduate_thesis_or_dissertations"):
# login()
req = requests.get(api_url)
data = req.json()
csv_data = []
error_data = []
count = 0
for itm in data['results']:
try:
# data = transform(itm)
csv_data.append(transform(itm))
# print(itm)
# print("List Count: ", len(csv_data))
except Exception as e:
print(e)
logging.error('Error at %s', 'division', exc_info=e)
error_data.append(itm)
count += 1
if (count % 50 == 0 and count != 0):
writeCsvFile(csv_data, error_data, count)
csv_data = []
error_data = []
# Write remaining csv if data
if csv_data or error_data:
writeCsvFile(csv_data, error_data, count)
print("count: ", count)
if __name__ == "__main__":
loadItems()
|
993,892 | d2d518c13036e655f74ea5b12921996d99722a1e | NUM_SEGMENTS = 5
def push_on_stack(segment):
line = [
"@%s" % segment,
"D=M",
"@SP",
"A=M",
"M=D",
"@SP",
"M=M+1",
]
return line
def restore_call(segment, iframe):
line = [
"@%s" % iframe,
"D=A",
"@R13",
"A=M-D",
"D=M",
"@%s" % segment,
"M=D"
]
return line
def def_function(args, ifunc):
i = ifunc()
line = [
"(%s)" % args[0],
"@%s" % args[1],
"D=A",
"(LOOP.ADD_LOCALS.%s)" % i,
"@NO_LOCALS.%s" % i,
"D;JEQ",
"@SP",
"A=M",
"M=0",
"@SP",
"M=M+1",
"D=D-1",
"@LOOP.ADD_LOCALS.%s" % i,
"D;JNE",
"(NO_LOCALS.%s)" % i
]
return line
def call_function(args, icall):
i = icall()
line = [
"@RET_ADDRESS.%s" % i,
"D=A",
"@SP",
"A=M",
"M=D",
"@SP",
"M=M+1",
*push_on_stack("LCL"),
*push_on_stack("ARG"),
*push_on_stack("THIS"),
*push_on_stack("THAT"),
"@SP",
"D=M",
"@%s" % args[1],
"D=D-A",
"@%s" % NUM_SEGMENTS,
"D=D-A",
"@ARG",
"M=D",
"@SP",
"D=M",
"@LCL",
"M=D",
"@%s" % args[0],
"0;JMP",
"(RET_ADDRESS.%s)" % i,
]
return line
def return_():
line = [
"@LCL",
"D=M",
"@R13",
"M=D",
"@5",
"D=A",
"@R13",
"A=M-D",
"D=M",
"@R14",
"M=D",
"@SP",
"AM=M-1",
"D=M",
"@ARG",
"A=M",
"M=D",
"@ARG",
"D=M+1",
"@SP",
"M=D",
*restore_call("THAT",1),
*restore_call("THIS",2),
*restore_call("ARG",3),
*restore_call("LCL",4),
"@R14",
"A=M",
"0;JMP"
]
return line
def init_asm(icall):
line = [
"@256",
'D=A',
'@SP',
'M=D'
]
line += call_function(["Sys.init", "0"], icall)
return line
|
993,893 | 2015866dc1b380f2d4456ba71a768be3834afe2e | # Samuel Fandiño #
num1=input('Introduzca el primer número ')
num1=int(num1)
num2=input('Introduzca el segundo número ')
num2=int(num2)
while num1 >= num2:
num2=input('Introduzca el segundo número nuevamente ')
num2=int(num2)
print('Los números introducidos fueron: '+str(num1)+' '+str(num2))
|
993,894 | 2bdbe984ae4b52d6b2758429f28a98b9fdabb34a | from django.conf.urls import url
from django.views.generic import RedirectView
from oscar.apps.catalogue.app import (
BaseCatalogueApplication as DefaultCatApp,
ReviewsApplication as DefaultReviewsApp
)
from shop.catalogue.views import product_or_category
class BaseCatalogueApplication(DefaultCatApp):
def get_urls(self):
urlpatterns = [
url(
r'^catalogue/$',
self.catalogue_view.as_view(),
name='index'
),
url(
r'^category/(?P<category_slug>[\w-]+(/[\w-]+)*)_(?P<pk>\d+)/$',
RedirectView.as_view(
permanent=True,
pattern_name='product_or_category'
),
name='category'
),
url(
r'^catalogue/ranges/(?P<slug>[\w-]+)/$',
self.range_view.as_view(),
name='range'
),
url(
r'^(?P<slug>[\w-]+(/[\w-]+)*)(?:/(?P<query>(?:\w+:[^\s\/]+)*))?(?:/p=(?P<page>(\d+)))?/$',
product_or_category,
name='product_or_category'
),
]
return self.post_process_urls(urlpatterns)
class ReviewsApplication(DefaultReviewsApp):
def get_urls(self):
urlpatterns = [
url(r'^(?P<product_slug>[\w-]*)_(?P<product_pk>\d+)/reviews/',
self.reviews_app.urls)
]
urlpatterns += super(ReviewsApplication, self).get_urls()
return self.post_process_urls(urlpatterns)
class CatalogueApplication(ReviewsApplication, BaseCatalogueApplication):
name = 'catalogue'
application = CatalogueApplication()
|
993,895 | 88e90ae388b10478681576c34cb255ff1afae0e2 | #!/usr/bin/python
# -*- coding: UTF-8 -*-
#
# This file is part of Womoon.
# Copyright: Kamila Chyla, 2006
#
# Womoon is free software; you can redistribute it and/or modify
# it under the terms of the GNU General Public License as published by
# the Free Software Foundation; either version 2 of the License, or
# (at your option) any later version.
#
# Womoon is distributed in the hope that it will be useful,
# but WITHOUT ANY WARRANTY; without even the implied warranty of
# MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the
# GNU General Public License for more details.
#
# You should have received a copy of the GNU General Public License
# along with Womoon; if not, write to the Free Software
# Foundation, Inc., 51 Franklin St, Fifth Floor, Boston, MA 02110-1301 USA
import chartmodel
from chartmodel import Notifier, ChartModel
from chartmodelio import ChartModelLoader, ChartModelSaver
class ChartProject(object):
"""
Stores information about diagram's filename
and its state loaded to one window.
"""
loader = ChartModelLoader()
saver = ChartModelSaver()
SAVED = "saved"
LOADED = "loaded"
CHANGED = "changed"
CREATED = "created"
EVENT_DESC = {
SAVED : "Diagram został zapisany",
LOADED : "Diagram został odczytany z pliku.",
CHANGED : "Diagram został zmieniony",
CREATED : "Utworzono nowy diagram",
}
def __init__(self):
self.on_project_changed_cb = []
self.chart = None
self.fname = None
self.is_modified = False
def is_untouched(self):
"""
Defines condition for loading diagram to the current
window (only if both no project was loaded before
and empty project was not yet modified)
"""
return self.fname is None and not self.is_modified
def project_changed(self, day_idx):
"""
Fires notification about ChartProject change.
This is called in two cases:
a) model field was set on self or
b) model has been modified
"""
self.is_modified = True
self.fire_project_changed(ChartProject.CHANGED)
def on_model_changed(self, attr, oldv, newv):
if self.chart is not None:
self.chart.add_on_change_callback(self.project_changed)
chart = Notifier("chart", on_model_changed)
def save(self, fname):
if fname is None:
return
ChartProject.saver.save(self.chart, fname)
self.is_modified = False
self.fire_project_changed(ChartProject.SAVED)
def load(self, fname):
if fname is None:
return
self.fname = fname
self.chart = ChartProject.loader.load(self.fname)
self.is_modified = False
self.fire_project_changed(ChartProject.LOADED)
def create(self, start_date):
self.is_modified = False
self.chart = ChartModel()
self.chart.generate_days(start_date)
self.fname = None
self.fire_project_changed(ChartProject.CREATED)
def add_on_project_changed_cb(self, cb):
self.on_project_changed_cb.append(cb)
def remove_on_project_changed_cb(self, cb):
if cb in self.on_project_changed_cb:
self.on_project_changed_cb.remove(cb)
def fire_project_changed(self, change_flag):
for cb in self.on_project_changed_cb:
cb(self, change_flag)
|
993,896 | 56e29abbd66bff7eb2902f06c953e27d4033bc5e | from sqlalchemy import Table, Column
class Victories(object):
def __init__(self,uid,won,lost):
self.uid=uid
self.won=won
self.lost=lost |
993,897 | f90bccc2ffec169328d9642b5b0eab036cd444a1 | class Guest:
def __init__(self, name, celebration, status, wallet, favourite_song):
self.name = name
self.celebration = celebration
self.status = status
self.wallet = wallet
self.favourite_song = favourite_song
def pay_fee(self, room):
if self.status == "VIP":
self.wallet -= room.vip_fee
room.till += room.vip_fee
else:
self.wallet -= room.regular_fee
room.till += room.regular_fee |
993,898 | f439918b6995f36a7a6fb1a0285b9f744fd5905c | from sklearn.preprocessing import StandardScaler
from sklearn.model_selection import GridSearchCV
from imblearn.over_sampling import SMOTE
from sklearn.model_selection import KFold
from sklearn.model_selection import train_test_split
from sklearn.svm import SVC, LinearSVC
from sklearn.metrics import confusion_matrix
from sklearn.neural_network import MLPClassifier
from matplotlib import pyplot as plt
from scipy.fftpack import fft, ifft
from scipy import signal
from mne.filter import filter_data
import scipy.signal as scisig
import numpy as np
import pandas as pd
import datetime
import pickle
import glob
import csv
import mne
def LoadEEGData(filename, EEGdevice):
""" This function converts a single .easy file (from NIC2) to an easy-to-use dataframe.
Uses both the .easy file and .info file (containing metadata)
---- Input ----
filename: string containing the .easy filepath
---- Output ----
df: dataframe containing all the EEG, accelerometer, and event marker data
fs: sampling rate for the EEG data (Hz)
fs_accel: sampling rate for the accelerometer data (Hz)
"""
if EEGdevice == 7:
x = 1
elif EEGdevice == 8:
# Read in the .easy file
df = pd.read_csv(filename, delimiter='\t', header=None)
# Get metadata from the .info file
fname = filename[:-5] + '.info'
with open(fname) as f:
content = f.readlines()
content = [x.strip() for x in content]
# Get the channel names
channel_info = [x for x in content if 'Channel ' in x]
channel_names = []
for ch in range(len(channel_info)):
channel_names.append(channel_info[ch].split(': ')[1])
channel_names.append('X')
channel_names.append('Y')
channel_names.append('Z')
channel_names.append('STI 014')
channel_names.append('DateTime')
# Get sampling rates
sampling_rates = [x for x in content if 'sampling rate: ' in x]
fs_all = []
for freq in range(len(sampling_rates)):
tmp = sampling_rates[freq].split(': ')[1].split(' ')[0]
if tmp in ['N/A']:
print('Skipping N/A')
else:
fs_all.append(float(sampling_rates[freq].split(': ')[1].split(' ')[0]))
# Store sampling rates
fs = fs_all[0]
fs_accel = fs_all[1]
# Assign the column names
df.columns = channel_names
# Return dataframe and sampling rates
return df, fs, fs_accel
def LoadBehavioralData(filename_behavioral):
"""
This function loads behavioral data for the motor screening task and formats it to use in this script
"""
behavioralData = pd.read_csv(filename_behavioral, ',')
behavioralData = behavioralData.transpose()
behavioralHeader = behavioralData.iloc[0]
behavioralData = behavioralData.iloc[2:]
behavioralData.columns = behavioralHeader
return behavioralData
def SyncTriggerPulses(EEGdata, EEGdevice, fs, behavioralData):
"""
This function returns the indices for events of interest
"""
if EEGdevice == 7:
print('Put code here')
elif EEGdevice == 8:
# Store where the values in trigger are equal to 8 (the audio trigger input channel number)
index_trigger = np.where(EEGdata['STI 014']!=0)
index_trigger = index_trigger[0]
# Check number of trials
num_of_trials = behavioralData.shape[0]
if num_of_trials > len(index_trigger):
num_of_trials = num_of_trials - 1
num_trials_removed = 1
else:
num_trials_removed = 0
trialLength = int(behavioralData['trialLength'][1])
# Get trial timing
t_trial_start = list()
t_trial_end = list()
# Creating lists of all trigger start and end locations
for i in range(0,num_of_trials):
t_trial_start.append(index_trigger[i])
t_trial_end.append(index_trigger[i] + int(trialLength*fs))
# Save rest period epochs as well as trials for comparison
t_rest_start = list()
t_rest_end = list()
for i in range(num_of_trials-1):
t_rest_start.append(t_trial_end[i])
t_rest_end.append(t_trial_start[i+1])
return num_of_trials, t_trial_start, t_trial_end, t_rest_start, t_rest_end
def EpochData(EEGdata, t_trial_start, t_trial_end):
"""
This function epochs the data
"""
if EEGdevice == 7:
channels = EEGdata.columns[1:8]
elif EEGdevice == 8:
channels = EEGdata.columns[0:8]
epochs = []
epochs_norm = []
for trial in range(0,len(t_trial_start)):
t_start = t_trial_start[trial]
t_end = t_trial_end[trial]
# Baseline
if trial == 0:
tb_start = t_trial_start[trial] - np.round(1.5*fs)
tb_end = t_trial_start[trial]
else:
tb_start = t_trial_end[trial-1]
tb_end = t_trial_start[trial]
baseline = EEGdata.loc[tb_start:tb_end][channels]
# Store epoch
tmp = (EEGdata.loc[t_start:t_end][channels] - np.mean(baseline,0))/np.std(baseline,0)
epochs_norm.append(tmp)
epochs.append(EEGdata.loc[t_start:t_end][channels])
return epochs, epochs_norm
def CutEpochs(epochs, fs, trial_type):
epochs_cut = []
trial_type_cut = list()
num_of_epochs = len(epochs)
full_epoch_length = np.shape(epochs[0])[0]
cut_epoch_length = int(1.750*fs)
sliding_window_starts = np.floor(np.linspace(1*fs, full_epoch_length - cut_epoch_length, 10))
# For each epoch
for epochOfInt in range(0,num_of_epochs):
# Reset the index within each epoch temporarily
reset_index = epochs[epochOfInt].reset_index(drop=True)
# Sliding window of 750 ms
for new_start in sliding_window_starts:
tmp = reset_index.loc[new_start:new_start+cut_epoch_length]
epochs_cut.append(tmp)
trial_type_cut.append(trial_type[epochOfInt])
num_of_trials_cut = len(trial_type_cut)
return epochs_cut, trial_type_cut, num_of_trials_cut
def OrganizeTrials(behavioralData):
"""
Organizes trials
"""
# Create lists for each trial type
trialL = list()
trialR = list()
i = 0
for letter in behavioralData['trialType']:
if letter == 'L':
trialL.append(i)
elif letter == 'R':
trialR.append(i)
i += 1
# Create a single list that includes which trial is which (L = 0, R = 1)
trial_type = list()
i = 0
for letter in behavioralData['trialType']:
if letter == 'L':
trial_type.append(0)
elif letter == 'R':
trial_type.append(1)
i += 1
print('np.shape(trial_type): ' + str(np.shape(trial_type)))
return trial_type, trialL, trialR
def ExtractFeatures(epochs, num_of_trials, channelsToUse, ds_factor):
"""
Extract signal features of interest
"""
# Get the summed delta power for each trial
alpha_power = dict.fromkeys(channelsToUse)
beta_power = dict.fromkeys(channelsToUse)
ds_f = ds_factor # downsampling factor
for chanOfInt in channelsToUse:
tmp_alpha = list()
tmp_beta = list()
for trial in range(0, num_of_trials):
f, Pxx_den = signal.welch(signal.decimate(epochs[trial][chanOfInt],ds_f), fs/ds_f, scaling='spectrum')
alpha_idx = np.where(np.logical_and(np.round(f) >= 8, np.round(f) <= 12))
tmp_alpha.append(np.sum(Pxx_den[alpha_idx]))
beta_idx = np.where(np.logical_and(np.round(f) >= 13, np.round(f) <= 30))
tmp_beta.append(np.sum(Pxx_den[beta_idx]))
alpha_power[chanOfInt] = tmp_alpha
beta_power[chanOfInt] = tmp_beta
return alpha_power, beta_power
def TrainDecoder(X, y):
"""
Trains the decoder on ALL the data (does not split into test and train because this is all train)
"""
# preprocess dataset, split into training and test part
args = np.arange(len(X))
np.random.shuffle(args)
X = [X[i] for i in args]
y = [y[i] for i in args]
X_not_scaled = X
X = StandardScaler().fit_transform(X)
# Resample to account for imbalance
method = SMOTE(kind='regular')
X_balanced, y_balanced = method.fit_sample(X, y)
# Determine model parameters
activations = ['relu','tanh']
alphas = np.logspace(-6, 3, 10)
solvers = ['lbfgs','sgd']
hyper_params = {"activation":activations, "alpha":alphas, "solver":solvers}
grid = GridSearchCV(MLPClassifier(learning_rate='constant', random_state=1), param_grid=hyper_params, cv=KFold(n_splits=5), verbose=True)
grid.fit(X_balanced, y_balanced)
# Fit the model
clf = grid.best_estimator_
clf.fit(X_balanced,y_balanced)
"""
# Determine model parameters
activations = ['relu','tanh']
alphas = np.logspace(-6, 3, 10)
solvers = ['lbfgs','sgd']
hyper_params = {"activation":activations, "alpha":alphas, "solver":solvers}
grid = GridSearchCV(MLPClassifier(learning_rate='constant', random_state=1), param_grid=hyper_params, cv=KFold(n_splits=5), verbose=True)
grid.fit(X, y)
# Fit the model
clf = grid.best_estimator_
clf.fit(X,y)
"""
return clf, X, X_not_scaled, y
def SaveDecoderAndData(clf, X, X_not_scaled, y, subjID):
"""
Save the decoder and the data it was trained/tested on
"""
time_to_save = datetime.datetime.now().isoformat()
time_to_save = time_to_save.replace('T','-')
time_to_save = time_to_save.replace(':','-')
model = clf
model_file = 'Models/' + subjID + '_MI_classifier_' + time_to_save[:19] + '.sav'
pickle.dump(model, open(model_file, 'wb'))
filepath_export_data = 'Models/' + subjID + '_data_for_MI_classifier_' + time_to_save[:19] + '.npz'
np.savez_compressed(filepath_export_data, subjID=subjID, X=X, X_not_scaled=X_not_scaled, y=y)
if __name__ == "__main__":
print()
print('-------------------------------------------------------------------')
print('---If you want to use automatic file selection, ---')
print('---move your EEG data files (.easy, .info) for motor screening ---')
print('---into the SaveData folder in this directory! ---')
print('-------------------------------------------------------------------')
print()
subjID = input('Enter subject ID: ')
EEGdevice = int(input('Enter EEG device (7 for DSI-7, 8 for Enobio): ')) # 7 for DSI-7, 8 for Enobio
if EEGdevice == 7:
print('DSI-7 selected')
elif EEGdevice == 8:
print('Enobio selected')
else:
raise ValueError('Invalid EEG device number')
findPaths = input('Manually enter in file paths (y/n)?: ')
if findPaths in ['y', 'Y', 'yes', 'Yes', 'YES']:
# Let user manually define file paths
filename_eeg = input('Enter EEG data file path: ')
filename_behavioral = input('Enter behavioral data file path: ')
elif findPaths in ['n', 'N', 'no', 'No', 'NO']:
# Automatically find the most recent files
eeg_files = glob.glob('SaveData/*' + subjID + '_Motor*.easy')
filename_eeg = eeg_files[-1] # load the most recent eeg file
print(filename_eeg)
behav_files = glob.glob('SaveData/' + subjID + '_Motor_Screening_*.csv')
filename_behavioral = behav_files[-1] # load the most recent behavioral file
print(filename_behavioral)
# Load EEG data
EEGdata, fs, fs_accel = LoadEEGData(filename_eeg, EEGdevice)
# Load behavioral data
behavioralData = LoadBehavioralData(filename_behavioral)
# Sync up trigger pulses
num_of_trials, t_trial_start, t_trial_end, t_rest_start, t_rest_end = SyncTriggerPulses(EEGdata, EEGdevice, fs, behavioralData)
# Filter the data
if EEGdevice == 7:
channels = EEGdata.columns[1:8]
elif EEGdevice == 8:
channels = EEGdata.columns[0:8]
EEGdata_filt = EEGdata.copy()
eeg_data = EEGdata[channels].values * 1.0 # multiply by 1.0 to convert int to float
filtered = filter_data(eeg_data.T, sfreq=fs, l_freq=1, h_freq=40)
EEGdata_filt[channels] = filtered.T
# Epoch the data
epochs, epochs_norm = EpochData(EEGdata_filt, t_trial_start, t_trial_end)
# Organize trial types
trial_type, trialL, trialR = OrganizeTrials(behavioralData)
# Cut epochs
epochs_cut, trial_type_cut, num_of_trials_cut = CutEpochs(epochs_norm, fs, trial_type)
# Get signal features
alpha_power, beta_power = ExtractFeatures(epochs_cut, num_of_trials_cut, ['C3','C4'], 1)
motor_features = [alpha_power['C3'], beta_power['C3'], alpha_power['C4'], beta_power['C4']]
motor_features = np.transpose(motor_features)
# Train model
print(np.shape(motor_features))
print(np.shape(trial_type_cut))
print(num_of_trials)
clf, X, X_not_scaled, y = TrainDecoder(motor_features, trial_type_cut)
# Save decoder and data it was trained/tested on
SaveDecoderAndData(clf, X, X_not_scaled, y, subjID) |
993,899 | c287b5f834c313060f4ad2004de99619a66b123a | def small_words(words):
for word in words:
if len(word) <= 3:
yield word
print [w for w in small_words(["The", "quick", "brown", "fox"])]
|
Subsets and Splits
No community queries yet
The top public SQL queries from the community will appear here once available.