def load_advent_dat(data):
import os
from data import parse
datapath = os.path.join(os.path.dirname(__file__), 'advent.dat')
with open(datapath, 'r') as datafile:
parse(data, datafile)
def __init__(self,gender):
self.description = {
'1': data.parse('<personality.trait.neutral>'),
'2': data.parse('<personality.trait.positive>'),
'3': data.parse('<personality.trait.positive> and <personality.trait.neutral>'),
'4': data.parse('<personality.trait.positive> and <personality.trait.neutral> but <personality.trait.negitive>'),
'5': data.parse('<personality.trait.positive> but <personality.trait.negitive>')
}[str(random.randint(1,5))];
def test_parse_GOLD_2_yields_list_of_Gold():
golds = data.parse("GOLD_2.csv", data.gold)
assert len(golds) > 0
for gold in golds:
assert isinstance(gold, data.Gold)
for column in gold:
assert_is_type_or_None(gold.date, datetime.date)
assert_is_type_or_None(gold.usd, float)
assert_is_type_or_None(gold.gbp, float)
assert_is_type_or_None(gold.eur, float)
def test_parse():
not_a_file = mock.Mock()
csv = """Date, Prime, Prime, Prime, Prime, Nothing, Nothing, FourPointSeven
2399-07-12,7,1597,13,229, ,,4.7
"""
stream = io.StringIO(csv)
with mock.patch("data.open_data_file", return_value=stream) as open_data_file:
data_set = data.parse("A StringIO", lambda args: args)
open_data_file.assert_called_with("A StringIO")
assert data_set == ["2399-07-12,7,1597,13,229, ,,4.7"]
def test_parse_NASDAQ_AAPL_yields_list_of_Aapl():
aapls = data.parse("NASDAQ_AAPL.csv", data.aapl)
assert len(aapls) > 0
for aapl in aapls:
assert isinstance(aapl, data.Aapl)
for column in aapl:
assert_is_type_or_None(aapl.date, datetime.date)
assert_is_type_or_None(aapl.open, float)
assert_is_type_or_None(aapl.high, float)
assert_is_type_or_None(aapl.low, float)
assert_is_type_or_None(aapl.close, float)
assert_is_type_or_None(aapl.volume, float)
def test_parse_MTGOXUSD_yields_list_of_Bitcoin():
bitcoins = data.parse("MTGOXUSD.csv", data.bitcoin)
assert len(bitcoins) > 0
for bitcoin in bitcoins:
assert isinstance(bitcoin, data.Bitcoin)
for column in bitcoin:
assert_is_type_or_None(bitcoin.date, datetime.date)
assert_is_type_or_None(bitcoin.open, float)
assert_is_type_or_None(bitcoin.high, float)
assert_is_type_or_None(bitcoin.low, float)
assert_is_type_or_None(bitcoin.close, float)
assert_is_type_or_None(bitcoin.volume_btc, float)
assert_is_type_or_None(bitcoin.volume_usd, float)
assert_is_type_or_None(bitcoin.weighted_price, float)
def insert_data():
filepath = select_file("osm")
data.parse(filepath)
"""
Sequential Child-Combination Tree-LSTM Network for PolEval 2017 evaluation campaign
Implementation inspired by "Efficient recursive (tree-structured) neural networks in TensorFlow" available at https://github.com/erickrf/treernn
"""
import sys
import numpy as np
import tensorflow as tf
import data
CHILDREN_NB = 5
word2idx, train_data, test_data = data.parse()
embed_size = 16
label_size = 3
max_epochs = 3
lr = 0.01
with tf.variable_scope('embed'):
embeddings = tf.get_variable('embeddings', [len(word2idx), embed_size])
with tf.variable_scope('lstm'):
W_i = tf.get_variable('W_i', [2 * embed_size, embed_size])
W_f = tf.get_variable('W_f', [2 * embed_size, embed_size])
W_o = tf.get_variable('W_o', [2 * embed_size, embed_size])
W_g = tf.get_variable('W_g', [2 * embed_size, embed_size])
c = tf.get_variable('c', [embed_size])
outbase = args.outbase or os.path.basename(os.path.splitext(args.input.rstrip('/'))[0])
def cast_if_number(s):
try:
return float(s) if '.' in s else int(s)
except ValueError:
return s
p_kwargs = {}
if args.parser_kwarg:
for item in args.parser_kwarg:
kw, val = item.split('=')
p_kwargs[kw] = cast_if_number(val)
log.debug(p_kwargs)
w_kwargs = {}
if args.writer_kwarg:
for item in args.writer_kwarg:
kw, val = item.split('=')
w_kwargs[kw] = cast_if_number(val)
log.debug(w_kwargs)
ds = data.parse(args.input, load_data=True, ignore_json=args.ignore_json, filetype=args.parser, **p_kwargs)
if not ds:
raise DataError('%s could not be parsed' % args.input)
if ds.data is None:
raise DataError('%s has no data' % args.input)
data.write(ds, ds.data, outbase, filetype=args.writer, **w_kwargs)
"""
Day 2 - Part 1 & 2
"""
import data
def compute(row):
for i in row.content:
for j in row.content:
if (i != j) & (i % j == 0):
return i / j
with open("inputs/day2", "r") as f:
read_data = f.read()
f.close()
d = data.parse(read_data)
res1 = 0
res2 = 0
for row in d.rows:
res1 += (row.max - row.min)
res2 += compute(row)
print "Part1 : " + str(res1) + ", Part2 : " + str(res2)
