Sergey Bartunov sbos

## Н.О.Ж. - Одноклассник
Снова учитель что-то
Пытается нам объяснять
Мне случать его неохота
Мне хочется погулять

Я после звонка наплевала
На следующий урок
Сломала окно и сьежала
Слушать тюменский панк-рок

## adagram_export.jl
function write_word2vec(path::AbstractString, vm::VectorModel, dict::Dictionary)
  fout = open(path, "w")
  sense_prob = zeros(T(vm))
  write(fout, "$(V(vm)) $(T(vm)) $(M(vm))\n")
  for v in 1:V(vm)
    write(fout, "$(dict.id2word[v])\n")
    expected_pi!(sense_prob, vm, v)
    for k in 1:T(vm)
      if sense_prob[k] < 1e-3 continue end
      write(fout, "$k $(sense_prob[k]) ")

## baseline.ipynb

      
              1 file
            
          
              0 forks
            
          
              0 comments
            
          
              0 stars
            
          
                sbos
                / baseline.ipynb
            
            
              Created
              December 15, 2015 04:03
            
          
      View baseline.ipynb
  
    
      Sorry, something went wrong. Reload?
      Sorry, we cannot display this file.
      Sorry, this file is invalid so it cannot be displayed.
      
          Viewer requires iframe.
      
    
## debug_th.py
import theano as th
import theano.tensor as T
import lasagne
import numpy as np
from theano.tensor.shared_randomstreams import RandomStreams
from lasagne.nonlinearities import tanh
import matplotlib.pyplot as plt
import sys

# change to True and see what happens

## random_bug.py
import cgt
import cgt.nn as nn
import numpy as np

batch = 5000

mean = 0.
for i in xrange(batch):
    mean += (cgt.randn() - mean) / (i+1)

## cgt_memory.py
import cgt
import cgt.nn as nn
import numpy as np
from scipy.stats import norm

def gaussian_density(x, mu, sigma):
    return cgt.exp(-cgt.square(x - mu) / 2 / cgt.square(sigma)) \
        / cgt.sqrt(2 * np.pi) / sigma

var_mu = nn.parameter(np.array(0.5))

## cgt_simplification_bug.py
import cgt
import cgt.nn as nn
import numpy as np
from scipy.stats import norm

def gaussian_density(x, mu, sigma):
    return cgt.exp(-cgt.square(x - mu) / 2 / cgt.square(sigma)) \
        / cgt.sqrt(2 * np.pi) / sigma

var_mu = nn.parameter(np.array(0.5))

## SVM.py
import numpy as np
from cvxopt import matrix, spmatrix
from cvxopt.solvers import qp
from cvxopt import solvers

class LinearSVM():
	def __init__(self, C):
		self.C = C

	def fit(self, data, labels):

## gist:018955227319ead18529
-.13015382	.033825178	-.02594826	-.288166	.04162671	-.06678655	-.22145432	.1749719	.030906504	-.12878266	.08305055	.11668899	-.06319859	-.0777372	.040560413	-.0305699	.17826276	-.10443129	-.00669569	.014138044	-.064837754	-.24928348	.060342744	-.08906554	.080405414	.0045637907	.007919261	-.14719364	.07927967	.011611855	-.0018610754	.08388769	.09033888	.0517999	.020472864	.19127382	.022346925	.06461436	.035737775	.049969383	.006792533	.11398394	-.03975431	-.112482205	-.03677935	.023018813	.044070702	.09208278	-.021259554	.13439591	-.048604295	.058010057	-.051273342	.03115164	.100593045	-.12984459	.10762949	.18540311	-.009627994	.0026174872	-.05229366	-.23992816	-.076821424	.12318988	-.1506341	-.13093792	.0035753904	-.09583123	.067648284	-.09169266	-.09788873	-.1272856	-.1690156	.017811107	-.027340638	.07608332	.10959153	-.027422326	.08745462	-.18536375	.013429028	-.05132068	.04882859	.23418726	-.026919987	.034001417	-.010213713	.035071775	-.038683213	-.092883974	-.056313466	.111097194	-.074547	-.035446037	-.00

## HMM.jl
module HMM

using Distributions

import Distributions.rand
import Distributions.fit

immutable HiddenMarkovModel{TP, K}
    theta::Vector{TP}
    A::Matrix{Float64}
	Снова учитель что-то
	Пытается нам объяснять
	Мне случать его неохота
	Мне хочется погулять

	Я после звонка наплевала
	На следующий урок
	Сломала окно и сьежала
	Слушать тюменский панк-рок
	function write_word2vec(path::AbstractString, vm::VectorModel, dict::Dictionary)
	fout = open(path, "w")
	sense_prob = zeros(T(vm))
	write(fout, "$(V(vm)) $(T(vm)) $(M(vm))\n")
	for v in 1:V(vm)
	write(fout, "$(dict.id2word[v])\n")
	expected_pi!(sense_prob, vm, v)
	for k in 1:T(vm)
	if sense_prob[k] < 1e-3 continue end
	write(fout, "$k $(sense_prob[k]) ")
	import theano as th
	import theano.tensor as T
	import lasagne
	import numpy as np
	from theano.tensor.shared_randomstreams import RandomStreams
	from lasagne.nonlinearities import tanh
	import matplotlib.pyplot as plt
	import sys

	# change to True and see what happens
	import cgt
	import cgt.nn as nn
	import numpy as np

	batch = 5000

	mean = 0.
	for i in xrange(batch):
	mean += (cgt.randn() - mean) / (i+1)
	import numpy as np
	from cvxopt import matrix, spmatrix
	from cvxopt.solvers import qp
	from cvxopt import solvers

	class LinearSVM():
	def __init__(self, C):
	self.C = C

	def fit(self, data, labels):
	module HMM

	using Distributions

	import Distributions.rand
	import Distributions.fit

	immutable HiddenMarkovModel{TP, K}
	theta::Vector{TP}
	A::Matrix{Float64}