In [1]:
import sys
sys.path.append("../../")
from elman import Elman
from rnn_disf_detection.load.load import load_data_from_array, get_tag_data_from_corpus_file, load_word_rep
from rnn_disf_detection.utils.tools import contextwinbackwards
In [2]:
nclasses = 43
vocsize = 9068
possize = 125
s = { 'emb_dimension' : 50, 'nhidden' : 50, 'acoustic' : 350, 'window' : 3, 'update_embeddings' : True }
In [3]:
rnn = Elman(ne = vocsize,
de = s['emb_dimension'],
nh = s['nhidden'],
na = 0,
nc = nclasses,
cs = s['window'],
npos = possize,
update_embeddings=s['update_embeddings'])
/Library/Python/2.7/site-packages/theano/scan_module/scan_perform_ext.py:133: RuntimeWarning: numpy.ndarray size changed, may indicate binary incompatibility
from scan_perform.scan_perform import *
In [4]:
print (possize * 3) + (s['emb_dimension']* 3)
525
In [5]:
test_file = "../data/audio/switchboard/4646A.npy"
In [6]:
import numpy as np
In [7]:
data = np.load(test_file)
In [8]:
_, acoustic_data, lex_data, pos_data, indices, labels = load_data_from_array(data, s['acoustic'])
In [9]:
from theano import tensor as T
import theano
y = T.concatenate((rnn.emb[rnn.idxs].reshape((rnn.idxs.shape[0], s['emb_dimension']*3)),\
rnn.pos[rnn.pos_idxs].reshape((rnn.pos_idxs.shape[0],possize*3))), 1)
def print_y():
return y
printy = theano.function( inputs = [rnn.idxs, rnn.pos_idxs],
outputs = y )
In [16]:
word_rep = load_word_rep("../data/tag_representations/swbd_word_rep.csv", dimension=None, word_rep_type="one_hot")
#print word_rep
idx2word = dict((k,v) for v,k in word_rep.iteritems())
acoustic = True
lr = 0.005
current_index = 0
_, acoustic_data, lex_data, pos_data, indices, labels = load_data_from_array(data, s['acoustic'])
nw = acoustic_data.shape[1] # number of examples in this dialogue
loss= 0
if acoustic:
test = 0
#load in the data to shared vars, can use 'set value too'
#self.lexical_data = self.shared_dataset(lex_data) #loads dummy data set as a shared variable
#self.pos_data = self.shared_dataset(pos_data)
#self.acoustic_data = data = self.shared_dataset(acoustic_data,dtype=theano.config.floatX)
#self.labels = self.shared_dataset(labels)
for start,stop in indices:
current_index+=1
test+=1 #TODO for testing
if test > 50: break #TODO for testing
print "llllll"
for row in lex_data[start:stop+1]:
print "r"
for c in row:
print c
print idx2word.get(c)
print 'words'
print lex_data[start:stop+1,:].shape
print lex_data[start:stop+1].shape
newwords = np.swapaxes(lex_data[start:stop+1,:],0,1)
print newwords.shape
print newwords
print 'pos'
print pos_data[start:stop+1,:].shape
print pos_data[start:stop+1].shape
newpos = np.swapaxes(pos_data[start:stop+1,:],0,1)
print newpos.shape
print newpos
#print 'acoust'
#print acoustic_data[:,start:stop+1].shape
#newacoust = np.swapaxes(acoustic_data[:,start:stop+1],0,1)
#print newacoust.shape
print 'label', labels[stop]
print printy(lex_data[start:stop+1],pos_data[start:stop+1])[:,525-possize:]
raw_input()
#print np.asarray(acoustic_data[:,start:stop+1],dtype='float32')
#x = rnn.train(lex_data[start:stop+1,:],pos_data[start:stop+1,:],labels[stop],lr)
#loss+=x
##rnn.normalize()
#print '[learning] >> %2.2f%%'%((stop+1)*100./nw),'of file {} / {}'.format(i+1,len(dialogues)),\
#'completed in %.2f (sec) <<\r'%(time.time()-tic),
sys.stdout.flush()
print "current train_loss (may include reg)is", loss/float(current_index)
llllll
r
-1
None
-1
None
2548
do
words
(1, 3)
(1, 3)
(3, 1)
[[ -1]
[ -1]
[2548]]
pos
(1, 3)
(1, 3)
(3, 1)
[[-1]
[-1]
[65]]
label 6.0
[[ 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0.
0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0.
0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0.
0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 1. 0. 0. 0. 0. 0. 0.
0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0.
0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0.
0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0.]]
llllll
r
-1
None
-1
None
2548
do
r
-1
None
2548
do
3391
you
words
(2, 3)
(2, 3)
(3, 2)
[[ -1 -1]
[ -1 2548]
[2548 3391]]
pos
(2, 3)
(2, 3)
(3, 2)
[[-1 -1]
[-1 65]
[65 16]]
label 4.0
[[ 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0.
0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0.
0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0.
0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 1. 0. 0. 0. 0. 0. 0.
0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0.
0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0.
0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0.]
[ 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 1. 0.
0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0.
0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0.
0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0.
0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0.
0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0.
0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0.]]
llllll
r
-1
None
-1
None
2548
do
r
-1
None
2548
do
3391
you
r
2548
do
3391
you
8958
know
words
(3, 3)
(3, 3)
(3, 3)
[[ -1 -1 2548]
[ -1 2548 3391]
[2548 3391 8958]]
pos
(3, 3)
(3, 3)
(3, 3)
[[ -1 -1 65]
[ -1 65 16]
[ 65 16 113]]
label 4.0
[[ 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0.
0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0.
0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0.
0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 1. 0. 0. 0. 0. 0. 0.
0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0.
0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0.
0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0.]
[ 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 1. 0.
0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0.
0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0.
0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0.
0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0.
0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0.
0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0.]
[ 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0.
0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0.
0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0.
0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0.
0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0.
0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0.
0. 0. 0. 0. 0. 1. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0.]]
llllll
r
-1
None
-1
None
2548
do
r
-1
None
2548
do
3391
you
r
2548
do
3391
you
8958
know
r
3391
you
8958
know
5318
anyone
words
(4, 3)
(4, 3)
(3, 4)
[[ -1 -1 2548 3391]
[ -1 2548 3391 8958]
[2548 3391 8958 5318]]
pos
(4, 3)
(4, 3)
(3, 4)
[[ -1 -1 65 16]
[ -1 65 16 113]
[ 65 16 113 68]]
label 4.0
[[ 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0.
0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0.
0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0.
0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 1. 0. 0. 0. 0. 0. 0.
0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0.
0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0.
0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0.]
[ 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 1. 0.
0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0.
0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0.
0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0.
0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0.
0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0.
0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0.]
[ 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0.
0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0.
0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0.
0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0.
0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0.
0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0.
0. 0. 0. 0. 0. 1. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0.]
[ 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0.
0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0.
0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0.
0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 1. 0. 0. 0.
0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0.
0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0.
0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0.]]
llllll
r
-1
None
-1
None
2548
do
r
-1
None
2548
do
3391
you
r
2548
do
3391
you
8958
know
r
3391
you
8958
know
5318
anyone
r
8958
know
5318
anyone
3064
that
words
(5, 3)
(5, 3)
(3, 5)
[[ -1 -1 2548 3391 8958]
[ -1 2548 3391 8958 5318]
[2548 3391 8958 5318 3064]]
pos
(5, 3)
(5, 3)
(3, 5)
[[ -1 -1 65 16 113]
[ -1 65 16 113 68]
[ 65 16 113 68 3]]
label 4.0
[[ 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0.
0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0.
0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0.
0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 1. 0. 0. 0. 0. 0. 0.
0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0.
0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0.
0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0.]
[ 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 1. 0.
0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0.
0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0.
0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0.
0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0.
0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0.
0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0.]
[ 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0.
0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0.
0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0.
0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0.
0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0.
0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0.
0. 0. 0. 0. 0. 1. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0.]
[ 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0.
0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0.
0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0.
0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 1. 0. 0. 0.
0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0.
0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0.
0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0.]
[ 0. 0. 0. 1. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0.
0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0.
0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0.
0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0.
0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0.
0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0.
0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0.]]
llllll
r
-1
None
-1
None
2548
do
r
-1
None
2548
do
3391
you
r
2548
do
3391
you
8958
know
r
3391
you
8958
know
5318
anyone
r
8958
know
5318
anyone
3064
that
r
5318
anyone
3064
that
1647
uh
words
(6, 3)
(6, 3)
(3, 6)
[[ -1 -1 2548 3391 8958 5318]
[ -1 2548 3391 8958 5318 3064]
[2548 3391 8958 5318 3064 1647]]
pos
(6, 3)
(6, 3)
(3, 6)
[[ -1 -1 65 16 113 68]
[ -1 65 16 113 68 3]
[ 65 16 113 68 3 57]]
label 0.0
[[ 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0.
0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0.
0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0.
0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 1. 0. 0. 0. 0. 0. 0.
0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0.
0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0.
0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0.]
[ 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 1. 0.
0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0.
0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0.
0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0.
0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0.
0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0.
0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0.]
[ 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0.
0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0.
0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0.
0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0.
0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0.
0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0.
0. 0. 0. 0. 0. 1. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0.]
[ 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0.
0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0.
0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0.
0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 1. 0. 0. 0.
0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0.
0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0.
0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0.]
[ 0. 0. 0. 1. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0.
0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0.
0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0.
0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0.
0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0.
0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0.
0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0.]
[ 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0.
0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0.
0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0.
0. 0. 0. 1. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0.
0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0.
0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0.
0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0.]]
llllll
r
-1
None
-1
None
2548
do
r
-1
None
2548
do
3391
you
r
2548
do
3391
you
8958
know
r
3391
you
8958
know
5318
anyone
r
8958
know
5318
anyone
3064
that
r
5318
anyone
3064
that
1647
uh
r
3064
that
1647
uh
4903
is
words
(7, 3)
(7, 3)
(3, 7)
[[ -1 -1 2548 3391 8958 5318 3064]
[ -1 2548 3391 8958 5318 3064 1647]
[2548 3391 8958 5318 3064 1647 4903]]
pos
(7, 3)
(7, 3)
(3, 7)
[[ -1 -1 65 16 113 68 3]
[ -1 65 16 113 68 3 57]
[ 65 16 113 68 3 57 67]]
label 4.0
[[ 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0.
0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0.
0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0.
0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 1. 0. 0. 0. 0. 0. 0.
0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0.
0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0.
0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0.]
[ 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 1. 0.
0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0.
0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0.
0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0.
0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0.
0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0.
0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0.]
[ 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0.
0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0.
0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0.
0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0.
0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0.
0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0.
0. 0. 0. 0. 0. 1. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0.]
[ 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0.
0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0.
0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0.
0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 1. 0. 0. 0.
0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0.
0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0.
0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0.]
[ 0. 0. 0. 1. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0.
0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0.
0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0.
0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0.
0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0.
0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0.
0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0.]
[ 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0.
0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0.
0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0.
0. 0. 0. 1. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0.
0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0.
0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0.
0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0.]
[ 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0.
0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0.
0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0.
0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 1. 0. 0. 0. 0.
0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0.
0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0.
0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0.]]
llllll
r
-1
None
-1
None
2548
do
r
-1
None
2548
do
3391
you
r
2548
do
3391
you
8958
know
r
3391
you
8958
know
5318
anyone
r
8958
know
5318
anyone
3064
that
r
5318
anyone
3064
that
1647
uh
r
3064
that
1647
uh
4903
is
r
1647
uh
4903
is
4903
is
words
(8, 3)
(8, 3)
(3, 8)
[[ -1 -1 2548 3391 8958 5318 3064 1647]
[ -1 2548 3391 8958 5318 3064 1647 4903]
[2548 3391 8958 5318 3064 1647 4903 4903]]
pos
(8, 3)
(8, 3)
(3, 8)
[[ -1 -1 65 16 113 68 3 57]
[ -1 65 16 113 68 3 57 67]
[ 65 16 113 68 3 57 67 67]]
label 10.0
[[ 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0.
0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0.
0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0.
0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 1. 0. 0. 0. 0. 0. 0.
0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0.
0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0.
0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0.]
[ 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 1. 0.
0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0.
0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0.
0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0.
0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0.
0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0.
0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0.]
[ 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0.
0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0.
0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0.
0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0.
0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0.
0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0.
0. 0. 0. 0. 0. 1. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0.]
[ 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0.
0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0.
0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0.
0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 1. 0. 0. 0.
0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0.
0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0.
0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0.]
[ 0. 0. 0. 1. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0.
0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0.
0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0.
0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0.
0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0.
0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0.
0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0.]
[ 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0.
0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0.
0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0.
0. 0. 0. 1. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0.
0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0.
0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0.
0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0.]
[ 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0.
0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0.
0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0.
0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 1. 0. 0. 0. 0.
0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0.
0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0.
0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0.]
[ 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0.
0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0.
0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0.
0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 1. 0. 0. 0. 0.
0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0.
0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0.
0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0.]]
llllll
r
-1
None
-1
None
2548
do
r
-1
None
2548
do
3391
you
r
2548
do
3391
you
8958
know
r
3391
you
8958
know
5318
anyone
r
8958
know
5318
anyone
3064
that
r
5318
anyone
3064
that
1647
uh
r
3064
that
1647
uh
4903
is
r
1647
uh
4903
is
4903
is
r
4903
is
4903
is
4909
in
words
(9, 3)
(9, 3)
(3, 9)
[[ -1 -1 2548 3391 8958 5318 3064 1647 4903]
[ -1 2548 3391 8958 5318 3064 1647 4903 4903]
[2548 3391 8958 5318 3064 1647 4903 4903 4909]]
pos
(9, 3)
(9, 3)
(3, 9)
[[ -1 -1 65 16 113 68 3 57 67]
[ -1 65 16 113 68 3 57 67 67]
[ 65 16 113 68 3 57 67 67 119]]
label 4.0
[[ 0. 0. 0. ..., 0. 0. 0.]
[ 0. 0. 0. ..., 0. 0. 0.]
[ 0. 0. 0. ..., 0. 0. 0.]
...,
[ 0. 0. 0. ..., 0. 0. 0.]
[ 0. 0. 0. ..., 0. 0. 0.]
[ 0. 0. 0. ..., 0. 0. 0.]]
llllll
r
-1
None
2548
do
3391
you
r
2548
do
3391
you
8958
know
r
3391
you
8958
know
5318
anyone
r
8958
know
5318
anyone
3064
that
r
5318
anyone
3064
that
1647
uh
r
3064
that
1647
uh
4903
is
r
1647
uh
4903
is
4903
is
r
4903
is
4903
is
4909
in
r
4903
is
4909
in
3518
a
words
(9, 3)
(9, 3)
(3, 9)
[[ -1 2548 3391 8958 5318 3064 1647 4903 4903]
[2548 3391 8958 5318 3064 1647 4903 4903 4909]
[3391 8958 5318 3064 1647 4903 4903 4909 3518]]
pos
(9, 3)
(9, 3)
(3, 9)
[[ -1 65 16 113 68 3 57 67 67]
[ 65 16 113 68 3 57 67 67 119]
[ 16 113 68 3 57 67 67 119 100]]
label 4.0
[[ 0. 0. 0. ..., 0. 0. 0.]
[ 0. 0. 0. ..., 0. 0. 0.]
[ 0. 0. 0. ..., 0. 0. 0.]
...,
[ 0. 0. 0. ..., 0. 0. 0.]
[ 0. 0. 0. ..., 0. 0. 0.]
[ 0. 0. 0. ..., 0. 0. 0.]]
llllll
r
2548
do
3391
you
8958
know
r
3391
you
8958
know
5318
anyone
r
8958
know
5318
anyone
3064
that
r
5318
anyone
3064
that
1647
uh
r
3064
that
1647
uh
4903
is
r
1647
uh
4903
is
4903
is
r
4903
is
4903
is
4909
in
r
4903
is
4909
in
3518
a
r
4909
in
3518
a
7484
nursing
words
(9, 3)
(9, 3)
(3, 9)
[[2548 3391 8958 5318 3064 1647 4903 4903 4909]
[3391 8958 5318 3064 1647 4903 4903 4909 3518]
[8958 5318 3064 1647 4903 4903 4909 3518 7484]]
pos
(9, 3)
(9, 3)
(3, 9)
[[ 65 16 113 68 3 57 67 67 119]
[ 16 113 68 3 57 67 67 119 100]
[113 68 3 57 67 67 119 100 68]]
label 4.0
[[ 0. 0. 0. ..., 0. 0. 0.]
[ 0. 0. 0. ..., 0. 0. 0.]
[ 0. 0. 0. ..., 0. 0. 0.]
...,
[ 0. 0. 0. ..., 0. 0. 0.]
[ 0. 0. 0. ..., 0. 0. 0.]
[ 0. 0. 0. ..., 0. 0. 0.]]
llllll
r
3391
you
8958
know
5318
anyone
r
8958
know
5318
anyone
3064
that
r
5318
anyone
3064
that
1647
uh
r
3064
that
1647
uh
4903
is
r
1647
uh
4903
is
4903
is
r
4903
is
4903
is
4909
in
r
4903
is
4909
in
3518
a
r
4909
in
3518
a
7484
nursing
r
3518
a
7484
nursing
4411
home
words
(9, 3)
(9, 3)
(3, 9)
[[3391 8958 5318 3064 1647 4903 4903 4909 3518]
[8958 5318 3064 1647 4903 4903 4909 3518 7484]
[5318 3064 1647 4903 4903 4909 3518 7484 4411]]
pos
(9, 3)
(9, 3)
(3, 9)
[[ 16 113 68 3 57 67 67 119 100]
[113 68 3 57 67 67 119 100 68]
[ 68 3 57 67 67 119 100 68 68]]
label 4.0
[[ 0. 0. 0. ..., 0. 0. 0.]
[ 0. 0. 0. ..., 0. 0. 0.]
[ 0. 0. 0. ..., 0. 0. 0.]
...,
[ 0. 0. 0. ..., 0. 0. 0.]
[ 0. 0. 0. ..., 0. 0. 0.]
[ 0. 0. 0. ..., 0. 0. 0.]]
llllll
r
8958
know
5318
anyone
3064
that
r
5318
anyone
3064
that
1647
uh
r
3064
that
1647
uh
4903
is
r
1647
uh
4903
is
4903
is
r
4903
is
4903
is
4909
in
r
4903
is
4909
in
3518
a
r
4909
in
3518
a
7484
nursing
r
3518
a
7484
nursing
4411
home
r
7484
nursing
4411
home
7881
or
words
(9, 3)
(9, 3)
(3, 9)
[[8958 5318 3064 1647 4903 4903 4909 3518 7484]
[5318 3064 1647 4903 4903 4909 3518 7484 4411]
[3064 1647 4903 4903 4909 3518 7484 4411 7881]]
pos
(9, 3)
(9, 3)
(3, 9)
[[113 68 3 57 67 67 119 100 68]
[ 68 3 57 67 67 119 100 68 68]
[ 3 57 67 67 119 100 68 68 78]]
label 4.0
[[ 0. 0. 0. ..., 0. 0. 0.]
[ 0. 0. 0. ..., 0. 0. 0.]
[ 0. 0. 0. ..., 0. 0. 0.]
...,
[ 0. 0. 0. ..., 0. 0. 0.]
[ 0. 0. 0. ..., 0. 0. 0.]
[ 0. 0. 0. ..., 0. 0. 0.]]
llllll
r
5318
anyone
3064
that
1647
uh
r
3064
that
1647
uh
4903
is
r
1647
uh
4903
is
4903
is
r
4903
is
4903
is
4909
in
r
4903
is
4909
in
3518
a
r
4909
in
3518
a
7484
nursing
r
3518
a
7484
nursing
4411
home
r
7484
nursing
4411
home
7881
or
r
4411
home
7881
or
424
has
words
(9, 3)
(9, 3)
(3, 9)
[[5318 3064 1647 4903 4903 4909 3518 7484 4411]
[3064 1647 4903 4903 4909 3518 7484 4411 7881]
[1647 4903 4903 4909 3518 7484 4411 7881 424]]
pos
(9, 3)
(9, 3)
(3, 9)
[[ 68 3 57 67 67 119 100 68 68]
[ 3 57 67 67 119 100 68 68 78]
[ 57 67 67 119 100 68 68 78 67]]
label 4.0
[[ 0. 0. 0. ..., 0. 0. 0.]
[ 0. 0. 0. ..., 0. 0. 0.]
[ 0. 0. 0. ..., 0. 0. 0.]
...,
[ 0. 0. 0. ..., 0. 0. 0.]
[ 0. 0. 0. ..., 0. 0. 0.]
[ 0. 0. 0. ..., 0. 0. 0.]]
llllll
r
3064
that
1647
uh
4903
is
r
1647
uh
4903
is
4903
is
r
4903
is
4903
is
4909
in
r
4903
is
4909
in
3518
a
r
4909
in
3518
a
7484
nursing
r
3518
a
7484
nursing
4411
home
r
7484
nursing
4411
home
7881
or
r
4411
home
7881
or
424
has
r
7881
or
424
has
4561
ever
words
(9, 3)
(9, 3)
(3, 9)
[[3064 1647 4903 4903 4909 3518 7484 4411 7881]
[1647 4903 4903 4909 3518 7484 4411 7881 424]
[4903 4903 4909 3518 7484 4411 7881 424 4561]]
pos
(9, 3)
(9, 3)
(3, 9)
[[ 3 57 67 67 119 100 68 68 78]
[ 57 67 67 119 100 68 68 78 67]
[ 67 67 119 100 68 68 78 67 17]]
label 4.0
[[ 0. 0. 0. ..., 0. 0. 0.]
[ 0. 0. 0. ..., 0. 0. 0.]
[ 0. 0. 0. ..., 0. 0. 0.]
...,
[ 0. 0. 0. ..., 0. 0. 0.]
[ 0. 0. 0. ..., 0. 0. 0.]
[ 0. 0. 0. ..., 0. 0. 0.]]
llllll
r
1647
uh
4903
is
4903
is
r
4903
is
4903
is
4909
in
r
4903
is
4909
in
3518
a
r
4909
in
3518
a
7484
nursing
r
3518
a
7484
nursing
4411
home
r
7484
nursing
4411
home
7881
or
r
4411
home
7881
or
424
has
r
7881
or
424
has
4561
ever
r
424
has
4561
ever
7171
been
words
(9, 3)
(9, 3)
(3, 9)
[[1647 4903 4903 4909 3518 7484 4411 7881 424]
[4903 4903 4909 3518 7484 4411 7881 424 4561]
[4903 4909 3518 7484 4411 7881 424 4561 7171]]
pos
(9, 3)
(9, 3)
(3, 9)
[[ 57 67 67 119 100 68 68 78 67]
[ 67 67 119 100 68 68 78 67 17]
[ 67 119 100 68 68 78 67 17 64]]
label 4.0
[[ 0. 0. 0. ..., 0. 0. 0.]
[ 0. 0. 0. ..., 0. 0. 0.]
[ 0. 0. 0. ..., 0. 0. 0.]
...,
[ 0. 0. 0. ..., 0. 0. 0.]
[ 0. 0. 0. ..., 0. 0. 0.]
[ 0. 0. 0. ..., 0. 0. 0.]]
llllll
r
4903
is
4903
is
4909
in
r
4903
is
4909
in
3518
a
r
4909
in
3518
a
7484
nursing
r
3518
a
7484
nursing
4411
home
r
7484
nursing
4411
home
7881
or
r
4411
home
7881
or
424
has
r
7881
or
424
has
4561
ever
r
424
has
4561
ever
7171
been
r
4561
ever
7171
been
4909
in
words
(9, 3)
(9, 3)
(3, 9)
[[4903 4903 4909 3518 7484 4411 7881 424 4561]
[4903 4909 3518 7484 4411 7881 424 4561 7171]
[4909 3518 7484 4411 7881 424 4561 7171 4909]]
pos
(9, 3)
(9, 3)
(3, 9)
[[ 67 67 119 100 68 68 78 67 17]
[ 67 119 100 68 68 78 67 17 64]
[119 100 68 68 78 67 17 64 119]]
label 4.0
[[ 0. 0. 0. ..., 0. 0. 0.]
[ 0. 0. 0. ..., 0. 0. 0.]
[ 0. 0. 0. ..., 0. 0. 0.]
...,
[ 0. 0. 0. ..., 0. 0. 0.]
[ 0. 0. 0. ..., 0. 0. 0.]
[ 0. 0. 0. ..., 0. 0. 0.]]
llllll
r
4903
is
4909
in
3518
a
r
4909
in
3518
a
7484
nursing
r
3518
a
7484
nursing
4411
home
r
7484
nursing
4411
home
7881
or
r
4411
home
7881
or
424
has
r
7881
or
424
has
4561
ever
r
424
has
4561
ever
7171
been
r
4561
ever
7171
been
4909
in
r
7171
been
4909
in
713
$unc$one
words
(9, 3)
(9, 3)
(3, 9)
[[4903 4909 3518 7484 4411 7881 424 4561 7171]
[4909 3518 7484 4411 7881 424 4561 7171 4909]
[3518 7484 4411 7881 424 4561 7171 4909 713]]
pos
(9, 3)
(9, 3)
(3, 9)
[[ 67 119 100 68 68 78 67 17 64]
[119 100 68 68 78 67 17 64 119]
[100 68 68 78 67 17 64 119 82]]
label 5.0
[[ 0. 0. 0. ..., 0. 0. 0.]
[ 0. 0. 0. ..., 0. 0. 0.]
[ 0. 0. 0. ..., 0. 0. 0.]
...,
[ 0. 0. 0. ..., 0. 0. 0.]
[ 0. 0. 0. ..., 0. 0. 0.]
[ 0. 0. 0. ..., 0. 0. 0.]]
llllll
r
4909
in
3518
a
7484
nursing
r
3518
a
7484
nursing
4411
home
r
7484
nursing
4411
home
7881
or
r
4411
home
7881
or
424
has
r
7881
or
424
has
4561
ever
r
424
has
4561
ever
7171
been
r
4561
ever
7171
been
4909
in
r
7171
been
4909
in
713
$unc$one
r
4909
in
713
$unc$one
1332
uh-huh
words
(9, 3)
(9, 3)
(3, 9)
[[4909 3518 7484 4411 7881 424 4561 7171 4909]
[3518 7484 4411 7881 424 4561 7171 4909 713]
[7484 4411 7881 424 4561 7171 4909 713 1332]]
pos
(9, 3)
(9, 3)
(3, 9)
[[119 100 68 68 78 67 17 64 119]
[100 68 68 78 67 17 64 119 82]
[ 68 68 78 67 17 64 119 82 57]]
label 7.0
[[ 0. 0. 0. ..., 0. 0. 0.]
[ 0. 0. 0. ..., 0. 0. 0.]
[ 0. 0. 0. ..., 0. 0. 0.]
...,
[ 0. 0. 0. ..., 0. 0. 0.]
[ 0. 0. 0. ..., 0. 0. 0.]
[ 0. 0. 0. ..., 0. 0. 0.]]
llllll
r
3518
a
7484
nursing
4411
home
r
7484
nursing
4411
home
7881
or
r
4411
home
7881
or
424
has
r
7881
or
424
has
4561
ever
r
424
has
4561
ever
7171
been
r
4561
ever
7171
been
4909
in
r
7171
been
4909
in
713
$unc$one
r
4909
in
713
$unc$one
1332
uh-huh
r
713
$unc$one
1332
uh-huh
4362
well
words
(9, 3)
(9, 3)
(3, 9)
[[3518 7484 4411 7881 424 4561 7171 4909 713]
[7484 4411 7881 424 4561 7171 4909 713 1332]
[4411 7881 424 4561 7171 4909 713 1332 4362]]
pos
(9, 3)
(9, 3)
(3, 9)
[[100 68 68 78 67 17 64 119 82]
[ 68 68 78 67 17 64 119 82 57]
[ 68 78 67 17 64 119 82 57 57]]
label 2.0
[[ 0. 0. 0. ..., 0. 0. 0.]
[ 0. 0. 0. ..., 0. 0. 0.]
[ 0. 0. 0. ..., 0. 0. 0.]
...,
[ 0. 0. 0. ..., 0. 0. 0.]
[ 0. 0. 0. ..., 0. 0. 0.]
[ 0. 0. 0. ..., 0. 0. 0.]]
llllll
r
7484
nursing
4411
home
7881
or
r
4411
home
7881
or
424
has
r
7881
or
424
has
4561
ever
r
424
has
4561
ever
7171
been
r
4561
ever
7171
been
4909
in
r
7171
been
4909
in
713
$unc$one
r
4909
in
713
$unc$one
1332
uh-huh
r
713
$unc$one
1332
uh-huh
4362
well
r
1332
uh-huh
4362
well
4907
im
words
(9, 3)
(9, 3)
(3, 9)
[[7484 4411 7881 424 4561 7171 4909 713 1332]
[4411 7881 424 4561 7171 4909 713 1332 4362]
[7881 424 4561 7171 4909 713 1332 4362 4907]]
pos
(9, 3)
(9, 3)
(3, 9)
[[ 68 68 78 67 17 64 119 82 57]
[ 68 78 67 17 64 119 82 57 57]
[ 78 67 17 64 119 82 57 57 96]]
label 4.0
[[ 0. 0. 0. ..., 0. 0. 0.]
[ 0. 0. 0. ..., 0. 0. 0.]
[ 0. 0. 0. ..., 0. 0. 0.]
...,
[ 0. 0. 0. ..., 0. 0. 0.]
[ 0. 0. 0. ..., 0. 0. 0.]
[ 0. 0. 0. ..., 0. 0. 0.]]
llllll
r
4411
home
7881
or
424
has
r
7881
or
424
has
4561
ever
r
424
has
4561
ever
7171
been
r
4561
ever
7171
been
4909
in
r
7171
been
4909
in
713
$unc$one
r
4909
in
713
$unc$one
1332
uh-huh
r
713
$unc$one
1332
uh-huh
4362
well
r
1332
uh-huh
4362
well
4907
im
r
4362
well
4907
im
3364
trying
words
(9, 3)
(9, 3)
(3, 9)
[[4411 7881 424 4561 7171 4909 713 1332 4362]
[7881 424 4561 7171 4909 713 1332 4362 4907]
[ 424 4561 7171 4909 713 1332 4362 4907 3364]]
pos
(9, 3)
(9, 3)
(3, 9)
[[ 68 78 67 17 64 119 82 57 57]
[ 78 67 17 64 119 82 57 57 96]
[ 67 17 64 119 82 57 57 96 62]]
label 4.0
[[ 0. 0. 0. ..., 0. 0. 0.]
[ 0. 0. 0. ..., 0. 0. 0.]
[ 0. 0. 0. ..., 0. 0. 0.]
...,
[ 0. 0. 0. ..., 0. 0. 0.]
[ 0. 0. 0. ..., 0. 0. 0.]
[ 0. 0. 0. ..., 0. 0. 0.]]
llllll
r
7881
or
424
has
4561
ever
r
424
has
4561
ever
7171
been
r
4561
ever
7171
been
4909
in
r
7171
been
4909
in
713
$unc$one
r
4909
in
713
$unc$one
1332
uh-huh
r
713
$unc$one
1332
uh-huh
4362
well
r
1332
uh-huh
4362
well
4907
im
r
4362
well
4907
im
3364
trying
r
4907
im
3364
trying
1167
to
words
(9, 3)
(9, 3)
(3, 9)
[[7881 424 4561 7171 4909 713 1332 4362 4907]
[ 424 4561 7171 4909 713 1332 4362 4907 3364]
[4561 7171 4909 713 1332 4362 4907 3364 1167]]
pos
(9, 3)
(9, 3)
(3, 9)
[[ 78 67 17 64 119 82 57 57 96]
[ 67 17 64 119 82 57 57 96 62]
[ 17 64 119 82 57 57 96 62 107]]
label 4.0
[[ 0. 0. 0. ..., 0. 0. 0.]
[ 0. 0. 0. ..., 0. 0. 0.]
[ 0. 0. 0. ..., 0. 0. 0.]
...,
[ 0. 0. 0. ..., 0. 0. 0.]
[ 0. 0. 0. ..., 0. 0. 0.]
[ 0. 0. 0. ..., 0. 0. 0.]]
llllll
r
424
has
4561
ever
7171
been
r
4561
ever
7171
been
4909
in
r
7171
been
4909
in
713
$unc$one
r
4909
in
713
$unc$one
1332
uh-huh
r
713
$unc$one
1332
uh-huh
4362
well
r
1332
uh-huh
4362
well
4907
im
r
4362
well
4907
im
3364
trying
r
4907
im
3364
trying
1167
to
r
3364
trying
1167
to
5301
think
words
(9, 3)
(9, 3)
(3, 9)
[[ 424 4561 7171 4909 713 1332 4362 4907 3364]
[4561 7171 4909 713 1332 4362 4907 3364 1167]
[7171 4909 713 1332 4362 4907 3364 1167 5301]]
pos
(9, 3)
(9, 3)
(3, 9)
[[ 67 17 64 119 82 57 57 96 62]
[ 17 64 119 82 57 57 96 62 107]
[ 64 119 82 57 57 96 62 107 113]]
label 5.0
[[ 0. 0. 0. ..., 0. 0. 0.]
[ 0. 0. 0. ..., 0. 0. 0.]
[ 0. 0. 0. ..., 0. 0. 0.]
...,
[ 0. 0. 0. ..., 0. 0. 0.]
[ 0. 0. 0. ..., 0. 0. 0.]
[ 0. 0. 0. ..., 0. 0. 0.]]
llllll
r
4561
ever
7171
been
4909
in
r
7171
been
4909
in
713
$unc$one
r
4909
in
713
$unc$one
1332
uh-huh
r
713
$unc$one
1332
uh-huh
4362
well
r
1332
uh-huh
4362
well
4907
im
r
4362
well
4907
im
3364
trying
r
4907
im
3364
trying
1167
to
r
3364
trying
1167
to
5301
think
r
1167
to
5301
think
6929
my
words
(9, 3)
(9, 3)
(3, 9)
[[4561 7171 4909 713 1332 4362 4907 3364 1167]
[7171 4909 713 1332 4362 4907 3364 1167 5301]
[4909 713 1332 4362 4907 3364 1167 5301 6929]]
pos
(9, 3)
(9, 3)
(3, 9)
[[ 17 64 119 82 57 57 96 62 107]
[ 64 119 82 57 57 96 62 107 113]
[119 82 57 57 96 62 107 113 0]]
label 6.0
[[ 0. 0. 0. ..., 0. 0. 0.]
[ 0. 0. 0. ..., 0. 0. 0.]
[ 0. 0. 0. ..., 0. 0. 0.]
...,
[ 0. 0. 0. ..., 0. 0. 0.]
[ 0. 0. 0. ..., 0. 0. 0.]
[ 1. 0. 0. ..., 0. 0. 0.]]
llllll
r
7171
been
4909
in
713
$unc$one
r
4909
in
713
$unc$one
1332
uh-huh
r
713
$unc$one
1332
uh-huh
4362
well
r
1332
uh-huh
4362
well
4907
im
r
4362
well
4907
im
3364
trying
r
4907
im
3364
trying
1167
to
r
3364
trying
1167
to
5301
think
r
1167
to
5301
think
6929
my
r
5301
think
6929
my
1647
uh
words
(9, 3)
(9, 3)
(3, 9)
[[7171 4909 713 1332 4362 4907 3364 1167 5301]
[4909 713 1332 4362 4907 3364 1167 5301 6929]
[ 713 1332 4362 4907 3364 1167 5301 6929 1647]]
pos
(9, 3)
(9, 3)
(3, 9)
[[ 64 119 82 57 57 96 62 107 113]
[119 82 57 57 96 62 107 113 0]
[ 82 57 57 96 62 107 113 0 57]]
label 0.0
[[ 0. 0. 0. ..., 0. 0. 0.]
[ 0. 0. 0. ..., 0. 0. 0.]
[ 0. 0. 0. ..., 0. 0. 0.]
...,
[ 0. 0. 0. ..., 0. 0. 0.]
[ 1. 0. 0. ..., 0. 0. 0.]
[ 0. 0. 0. ..., 0. 0. 0.]]
llllll
r
4909
in
713
$unc$one
1332
uh-huh
r
713
$unc$one
1332
uh-huh
4362
well
r
1332
uh-huh
4362
well
4907
im
r
4362
well
4907
im
3364
trying
r
4907
im
3364
trying
1167
to
r
3364
trying
1167
to
5301
think
r
1167
to
5301
think
6929
my
r
5301
think
6929
my
1647
uh
r
6929
my
1647
uh
1647
uh
words
(9, 3)
(9, 3)
(3, 9)
[[4909 713 1332 4362 4907 3364 1167 5301 6929]
[ 713 1332 4362 4907 3364 1167 5301 6929 1647]
[1332 4362 4907 3364 1167 5301 6929 1647 1647]]
pos
(9, 3)
(9, 3)
(3, 9)
[[119 82 57 57 96 62 107 113 0]
[ 82 57 57 96 62 107 113 0 57]
[ 57 57 96 62 107 113 0 57 57]]
label 0.0
[[ 0. 0. 0. ..., 0. 0. 0.]
[ 0. 0. 0. ..., 0. 0. 0.]
[ 0. 0. 0. ..., 0. 0. 0.]
...,
[ 1. 0. 0. ..., 0. 0. 0.]
[ 0. 0. 0. ..., 0. 0. 0.]
[ 0. 0. 0. ..., 0. 0. 0.]]
llllll
r
713
$unc$one
1332
uh-huh
4362
well
r
1332
uh-huh
4362
well
4907
im
r
4362
well
4907
im
3364
trying
r
4907
im
3364
trying
1167
to
r
3364
trying
1167
to
5301
think
r
1167
to
5301
think
6929
my
r
5301
think
6929
my
1647
uh
r
6929
my
1647
uh
1647
uh
r
1647
uh
1647
uh
5551
wifes
words
(9, 3)
(9, 3)
(3, 9)
[[ 713 1332 4362 4907 3364 1167 5301 6929 1647]
[1332 4362 4907 3364 1167 5301 6929 1647 1647]
[4362 4907 3364 1167 5301 6929 1647 1647 5551]]
pos
(9, 3)
(9, 3)
(3, 9)
[[ 82 57 57 96 62 107 113 0 57]
[ 57 57 96 62 107 113 0 57 57]
[ 57 96 62 107 113 0 57 57 55]]
label 4.0
[[ 0. 0. 0. ..., 0. 0. 0.]
[ 0. 0. 0. ..., 0. 0. 0.]
[ 0. 0. 0. ..., 0. 0. 0.]
...,
[ 0. 0. 0. ..., 0. 0. 0.]
[ 0. 0. 0. ..., 0. 0. 0.]
[ 0. 0. 0. ..., 0. 0. 0.]]
llllll
r
1332
uh-huh
4362
well
4907
im
r
4362
well
4907
im
3364
trying
r
4907
im
3364
trying
1167
to
r
3364
trying
1167
to
5301
think
r
1167
to
5301
think
6929
my
r
5301
think
6929
my
1647
uh
r
6929
my
1647
uh
1647
uh
r
1647
uh
1647
uh
5551
wifes
r
1647
uh
5551
wifes
7131
grandmother
words
(9, 3)
(9, 3)
(3, 9)
[[1332 4362 4907 3364 1167 5301 6929 1647 1647]
[4362 4907 3364 1167 5301 6929 1647 1647 5551]
[4907 3364 1167 5301 6929 1647 1647 5551 7131]]
pos
(9, 3)
(9, 3)
(3, 9)
[[ 57 57 96 62 107 113 0 57 57]
[ 57 96 62 107 113 0 57 57 55]
[ 96 62 107 113 0 57 57 55 68]]
label 4.0
[[ 0. 0. 0. ..., 0. 0. 0.]
[ 0. 0. 0. ..., 0. 0. 0.]
[ 0. 0. 0. ..., 0. 0. 0.]
...,
[ 0. 0. 0. ..., 0. 0. 0.]
[ 0. 0. 0. ..., 0. 0. 0.]
[ 0. 0. 0. ..., 0. 0. 0.]]
llllll
r
4362
well
4907
im
3364
trying
r
4907
im
3364
trying
1167
to
r
3364
trying
1167
to
5301
think
r
1167
to
5301
think
6929
my
r
5301
think
6929
my
1647
uh
r
6929
my
1647
uh
1647
uh
r
1647
uh
1647
uh
5551
wifes
r
1647
uh
5551
wifes
7131
grandmother
r
5551
wifes
7131
grandmother
419
had
words
(9, 3)
(9, 3)
(3, 9)
[[4362 4907 3364 1167 5301 6929 1647 1647 5551]
[4907 3364 1167 5301 6929 1647 1647 5551 7131]
[3364 1167 5301 6929 1647 1647 5551 7131 419]]
pos
(9, 3)
(9, 3)
(3, 9)
[[ 57 96 62 107 113 0 57 57 55]
[ 96 62 107 113 0 57 57 55 68]
[ 62 107 113 0 57 57 55 68 63]]
label 4.0
[[ 0. 0. 0. ..., 0. 0. 0.]
[ 0. 0. 0. ..., 0. 0. 0.]
[ 0. 0. 0. ..., 0. 0. 0.]
...,
[ 0. 0. 0. ..., 0. 0. 0.]
[ 0. 0. 0. ..., 0. 0. 0.]
[ 0. 0. 0. ..., 0. 0. 0.]]
llllll
r
4907
im
3364
trying
1167
to
r
3364
trying
1167
to
5301
think
r
1167
to
5301
think
6929
my
r
5301
think
6929
my
1647
uh
r
6929
my
1647
uh
1647
uh
r
1647
uh
1647
uh
5551
wifes
r
1647
uh
5551
wifes
7131
grandmother
r
5551
wifes
7131
grandmother
419
had
r
7131
grandmother
419
had
681
alzheimers
words
(9, 3)
(9, 3)
(3, 9)
[[4907 3364 1167 5301 6929 1647 1647 5551 7131]
[3364 1167 5301 6929 1647 1647 5551 7131 419]
[1167 5301 6929 1647 1647 5551 7131 419 681]]
pos
(9, 3)
(9, 3)
(3, 9)
[[ 96 62 107 113 0 57 57 55 68]
[ 62 107 113 0 57 57 55 68 63]
[107 113 0 57 57 55 68 63 72]]
label 5.0
[[ 0. 0. 0. ..., 0. 0. 0.]
[ 0. 0. 0. ..., 0. 0. 0.]
[ 1. 0. 0. ..., 0. 0. 0.]
...,
[ 0. 0. 0. ..., 0. 0. 0.]
[ 0. 0. 0. ..., 0. 0. 0.]
[ 0. 0. 0. ..., 0. 0. 0.]]
llllll
r
3364
trying
1167
to
5301
think
r
1167
to
5301
think
6929
my
r
5301
think
6929
my
1647
uh
r
6929
my
1647
uh
1647
uh
r
1647
uh
1647
uh
5551
wifes
r
1647
uh
5551
wifes
7131
grandmother
r
5551
wifes
7131
grandmother
419
had
r
7131
grandmother
419
had
681
alzheimers
r
419
had
681
alzheimers
7613
and
words
(9, 3)
(9, 3)
(3, 9)
[[3364 1167 5301 6929 1647 1647 5551 7131 419]
[1167 5301 6929 1647 1647 5551 7131 419 681]
[5301 6929 1647 1647 5551 7131 419 681 7613]]
pos
(9, 3)
(9, 3)
(3, 9)
[[ 62 107 113 0 57 57 55 68 63]
[107 113 0 57 57 55 68 63 72]
[113 0 57 57 55 68 63 72 78]]
label 6.0
[[ 0. 0. 0. ..., 0. 0. 0.]
[ 1. 0. 0. ..., 0. 0. 0.]
[ 0. 0. 0. ..., 0. 0. 0.]
...,
[ 0. 0. 0. ..., 0. 0. 0.]
[ 0. 0. 0. ..., 0. 0. 0.]
[ 0. 0. 0. ..., 0. 0. 0.]]
llllll
r
1167
to
5301
think
6929
my
r
5301
think
6929
my
1647
uh
r
6929
my
1647
uh
1647
uh
r
1647
uh
1647
uh
5551
wifes
r
1647
uh
5551
wifes
7131
grandmother
r
5551
wifes
7131
grandmother
419
had
r
7131
grandmother
419
had
681
alzheimers
r
419
had
681
alzheimers
7613
and
r
681
alzheimers
7613
and
2423
they
words
(9, 3)
(9, 3)
(3, 9)
[[1167 5301 6929 1647 1647 5551 7131 419 681]
[5301 6929 1647 1647 5551 7131 419 681 7613]
[6929 1647 1647 5551 7131 419 681 7613 2423]]
pos
(9, 3)
(9, 3)
(3, 9)
[[107 113 0 57 57 55 68 63 72]
[113 0 57 57 55 68 63 72 78]
[ 0 57 57 55 68 63 72 78 16]]
label 4.0
[[ 1. 0. 0. ..., 0. 0. 0.]
[ 0. 0. 0. ..., 0. 0. 0.]
[ 0. 0. 0. ..., 0. 0. 0.]
...,
[ 0. 0. 0. ..., 0. 0. 0.]
[ 0. 0. 0. ..., 0. 0. 0.]
[ 0. 0. 0. ..., 0. 0. 0.]]
llllll
r
5301
think
6929
my
1647
uh
r
6929
my
1647
uh
1647
uh
r
1647
uh
1647
uh
5551
wifes
r
1647
uh
5551
wifes
7131
grandmother
r
5551
wifes
7131
grandmother
419
had
r
7131
grandmother
419
had
681
alzheimers
r
419
had
681
alzheimers
7613
and
r
681
alzheimers
7613
and
2423
they
r
7613
and
2423
they
5340
were
words
(9, 3)
(9, 3)
(3, 9)
[[5301 6929 1647 1647 5551 7131 419 681 7613]
[6929 1647 1647 5551 7131 419 681 7613 2423]
[1647 1647 5551 7131 419 681 7613 2423 5340]]
pos
(9, 3)
(9, 3)
(3, 9)
[[113 0 57 57 55 68 63 72 78]
[ 0 57 57 55 68 63 72 78 16]
[ 57 57 55 68 63 72 78 16 63]]
label 4.0
[[ 0. 0. 0. ..., 0. 0. 0.]
[ 0. 0. 0. ..., 0. 0. 0.]
[ 0. 0. 0. ..., 0. 0. 0.]
...,
[ 0. 0. 0. ..., 0. 0. 0.]
[ 0. 0. 0. ..., 0. 0. 0.]
[ 0. 0. 0. ..., 0. 0. 0.]]
llllll
r
6929
my
1647
uh
1647
uh
r
1647
uh
1647
uh
5551
wifes
r
1647
uh
5551
wifes
7131
grandmother
r
5551
wifes
7131
grandmother
419
had
r
7131
grandmother
419
had
681
alzheimers
r
419
had
681
alzheimers
7613
and
r
681
alzheimers
7613
and
2423
they
r
7613
and
2423
they
5340
were
r
2423
they
5340
were
872
going
words
(9, 3)
(9, 3)
(3, 9)
[[6929 1647 1647 5551 7131 419 681 7613 2423]
[1647 1647 5551 7131 419 681 7613 2423 5340]
[1647 5551 7131 419 681 7613 2423 5340 872]]
pos
(9, 3)
(9, 3)
(3, 9)
[[ 0 57 57 55 68 63 72 78 16]
[57 57 55 68 63 72 78 16 63]
[57 55 68 63 72 78 16 63 62]]
label 4.0
[[ 0. 0. 0. ..., 0. 0. 0.]
[ 0. 0. 0. ..., 0. 0. 0.]
[ 0. 0. 0. ..., 0. 0. 0.]
...,
[ 0. 0. 0. ..., 0. 0. 0.]
[ 0. 0. 0. ..., 0. 0. 0.]
[ 0. 0. 0. ..., 0. 0. 0.]]
llllll
r
1647
uh
1647
uh
5551
wifes
r
1647
uh
5551
wifes
7131
grandmother
r
5551
wifes
7131
grandmother
419
had
r
7131
grandmother
419
had
681
alzheimers
r
419
had
681
alzheimers
7613
and
r
681
alzheimers
7613
and
2423
they
r
7613
and
2423
they
5340
were
r
2423
they
5340
were
872
going
r
5340
were
872
going
1167
to
words
(9, 3)
(9, 3)
(3, 9)
[[1647 1647 5551 7131 419 681 7613 2423 5340]
[1647 5551 7131 419 681 7613 2423 5340 872]
[5551 7131 419 681 7613 2423 5340 872 1167]]
pos
(9, 3)
(9, 3)
(3, 9)
[[ 57 57 55 68 63 72 78 16 63]
[ 57 55 68 63 72 78 16 63 62]
[ 55 68 63 72 78 16 63 62 107]]
label 4.0
[[ 0. 0. 0. ..., 0. 0. 0.]
[ 0. 0. 0. ..., 0. 0. 0.]
[ 0. 0. 0. ..., 0. 0. 0.]
...,
[ 0. 0. 0. ..., 0. 0. 0.]
[ 0. 0. 0. ..., 0. 0. 0.]
[ 0. 0. 0. ..., 0. 0. 0.]]
llllll
r
1647
uh
5551
wifes
7131
grandmother
r
5551
wifes
7131
grandmother
419
had
r
7131
grandmother
419
had
681
alzheimers
r
419
had
681
alzheimers
7613
and
r
681
alzheimers
7613
and
2423
they
r
7613
and
2423
they
5340
were
r
2423
they
5340
were
872
going
r
5340
were
872
going
1167
to
r
872
going
1167
to
3469
put
words
(9, 3)
(9, 3)
(3, 9)
[[1647 5551 7131 419 681 7613 2423 5340 872]
[5551 7131 419 681 7613 2423 5340 872 1167]
[7131 419 681 7613 2423 5340 872 1167 3469]]
pos
(9, 3)
(9, 3)
(3, 9)
[[ 57 55 68 63 72 78 16 63 62]
[ 55 68 63 72 78 16 63 62 107]
[ 68 63 72 78 16 63 62 107 113]]
label 4.0
[[ 0. 0. 0. ..., 0. 0. 0.]
[ 0. 0. 0. ..., 0. 0. 0.]
[ 0. 0. 0. ..., 0. 0. 0.]
...,
[ 0. 0. 0. ..., 0. 0. 0.]
[ 0. 0. 0. ..., 0. 0. 0.]
[ 0. 0. 0. ..., 0. 0. 0.]]
llllll
r
5551
wifes
7131
grandmother
419
had
r
7131
grandmother
419
had
681
alzheimers
r
419
had
681
alzheimers
7613
and
r
681
alzheimers
7613
and
2423
they
r
7613
and
2423
they
5340
were
r
2423
they
5340
were
872
going
r
5340
were
872
going
1167
to
r
872
going
1167
to
3469
put
r
1167
to
3469
put
1081
her
words
(9, 3)
(9, 3)
(3, 9)
[[5551 7131 419 681 7613 2423 5340 872 1167]
[7131 419 681 7613 2423 5340 872 1167 3469]
[ 419 681 7613 2423 5340 872 1167 3469 1081]]
pos
(9, 3)
(9, 3)
(3, 9)
[[ 55 68 63 72 78 16 63 62 107]
[ 68 63 72 78 16 63 62 107 113]
[ 63 72 78 16 63 62 107 113 16]]
label 4.0
[[ 0. 0. 0. ..., 0. 0. 0.]
[ 0. 0. 0. ..., 0. 0. 0.]
[ 0. 0. 0. ..., 0. 0. 0.]
...,
[ 0. 0. 0. ..., 0. 0. 0.]
[ 0. 0. 0. ..., 0. 0. 0.]
[ 0. 0. 0. ..., 0. 0. 0.]]
llllll
r
7131
grandmother
419
had
681
alzheimers
r
419
had
681
alzheimers
7613
and
r
681
alzheimers
7613
and
2423
they
r
7613
and
2423
they
5340
were
r
2423
they
5340
were
872
going
r
5340
were
872
going
1167
to
r
872
going
1167
to
3469
put
r
1167
to
3469
put
1081
her
r
3469
put
1081
her
1614
into
words
(9, 3)
(9, 3)
(3, 9)
[[7131 419 681 7613 2423 5340 872 1167 3469]
[ 419 681 7613 2423 5340 872 1167 3469 1081]
[ 681 7613 2423 5340 872 1167 3469 1081 1614]]
pos
(9, 3)
(9, 3)
(3, 9)
[[ 68 63 72 78 16 63 62 107 113]
[ 63 72 78 16 63 62 107 113 16]
[ 72 78 16 63 62 107 113 16 119]]
label 4.0
[[ 0. 0. 0. ..., 0. 0. 0.]
[ 0. 0. 0. ..., 0. 0. 0.]
[ 0. 0. 0. ..., 0. 0. 0.]
...,
[ 0. 0. 0. ..., 0. 0. 0.]
[ 0. 0. 0. ..., 0. 0. 0.]
[ 0. 0. 0. ..., 0. 0. 0.]]
llllll
r
419
had
681
alzheimers
7613
and
r
681
alzheimers
7613
and
2423
they
r
7613
and
2423
they
5340
were
r
2423
they
5340
were
872
going
r
5340
were
872
going
1167
to
r
872
going
1167
to
3469
put
r
1167
to
3469
put
1081
her
r
3469
put
1081
her
1614
into
r
1081
her
1614
into
3518
a
words
(9, 3)
(9, 3)
(3, 9)
[[ 419 681 7613 2423 5340 872 1167 3469 1081]
[ 681 7613 2423 5340 872 1167 3469 1081 1614]
[7613 2423 5340 872 1167 3469 1081 1614 3518]]
pos
(9, 3)
(9, 3)
(3, 9)
[[ 63 72 78 16 63 62 107 113 16]
[ 72 78 16 63 62 107 113 16 119]
[ 78 16 63 62 107 113 16 119 100]]
label 4.0
[[ 0. 0. 0. ..., 0. 0. 0.]
[ 0. 0. 0. ..., 0. 0. 0.]
[ 0. 0. 0. ..., 0. 0. 0.]
...,
[ 0. 0. 0. ..., 0. 0. 0.]
[ 0. 0. 0. ..., 0. 0. 0.]
[ 0. 0. 0. ..., 0. 0. 0.]]
llllll
r
681
alzheimers
7613
and
2423
they
r
7613
and
2423
they
5340
were
r
2423
they
5340
were
872
going
r
5340
were
872
going
1167
to
r
872
going
1167
to
3469
put
r
1167
to
3469
put
1081
her
r
3469
put
1081
her
1614
into
r
1081
her
1614
into
3518
a
r
1614
into
3518
a
3518
a
words
(9, 3)
(9, 3)
(3, 9)
[[ 681 7613 2423 5340 872 1167 3469 1081 1614]
[7613 2423 5340 872 1167 3469 1081 1614 3518]
[2423 5340 872 1167 3469 1081 1614 3518 3518]]
pos
(9, 3)
(9, 3)
(3, 9)
[[ 72 78 16 63 62 107 113 16 119]
[ 78 16 63 62 107 113 16 119 100]
[ 16 63 62 107 113 16 119 100 100]]
label 10.0
[[ 0. 0. 0. ..., 0. 0. 0.]
[ 0. 0. 0. ..., 0. 0. 0.]
[ 0. 0. 0. ..., 0. 0. 0.]
...,
[ 0. 0. 0. ..., 0. 0. 0.]
[ 0. 0. 0. ..., 0. 0. 0.]
[ 0. 0. 0. ..., 0. 0. 0.]]
llllll
r
7613
and
2423
they
5340
were
r
2423
they
5340
were
872
going
r
5340
were
872
going
1167
to
r
872
going
1167
to
3469
put
r
1167
to
3469
put
1081
her
r
3469
put
1081
her
1614
into
r
1081
her
1614
into
3518
a
r
1614
into
3518
a
3518
a
r
3518
a
3518
a
7484
nursing
words
(9, 3)
(9, 3)
(3, 9)
[[7613 2423 5340 872 1167 3469 1081 1614 3518]
[2423 5340 872 1167 3469 1081 1614 3518 3518]
[5340 872 1167 3469 1081 1614 3518 3518 7484]]
pos
(9, 3)
(9, 3)
(3, 9)
[[ 78 16 63 62 107 113 16 119 100]
[ 16 63 62 107 113 16 119 100 100]
[ 63 62 107 113 16 119 100 100 68]]
label 4.0
[[ 0. 0. 0. ..., 0. 0. 0.]
[ 0. 0. 0. ..., 0. 0. 0.]
[ 0. 0. 0. ..., 0. 0. 0.]
...,
[ 0. 0. 0. ..., 0. 0. 0.]
[ 0. 0. 0. ..., 0. 0. 0.]
[ 0. 0. 0. ..., 0. 0. 0.]]
llllll
r
2423
they
5340
were
872
going
r
5340
were
872
going
1167
to
r
872
going
1167
to
3469
put
r
1167
to
3469
put
1081
her
r
3469
put
1081
her
1614
into
r
1081
her
1614
into
3518
a
r
1614
into
3518
a
3518
a
r
3518
a
3518
a
7484
nursing
r
3518
a
7484
nursing
4411
home
words
(9, 3)
(9, 3)
(3, 9)
[[2423 5340 872 1167 3469 1081 1614 3518 3518]
[5340 872 1167 3469 1081 1614 3518 3518 7484]
[ 872 1167 3469 1081 1614 3518 3518 7484 4411]]
pos
(9, 3)
(9, 3)
(3, 9)
[[ 16 63 62 107 113 16 119 100 100]
[ 63 62 107 113 16 119 100 100 68]
[ 62 107 113 16 119 100 100 68 68]]
label 5.0
[[ 0. 0. 0. ..., 0. 0. 0.]
[ 0. 0. 0. ..., 0. 0. 0.]
[ 0. 0. 0. ..., 0. 0. 0.]
...,
[ 0. 0. 0. ..., 0. 0. 0.]
[ 0. 0. 0. ..., 0. 0. 0.]
[ 0. 0. 0. ..., 0. 0. 0.]]
llllll
r
5340
were
872
going
1167
to
r
872
going
1167
to
3469
put
r
1167
to
3469
put
1081
her
r
3469
put
1081
her
1614
into
r
1081
her
1614
into
3518
a
r
1614
into
3518
a
3518
a
r
3518
a
3518
a
7484
nursing
r
3518
a
7484
nursing
4411
home
r
7484
nursing
4411
home
7613
and
words
(9, 3)
(9, 3)
(3, 9)
[[5340 872 1167 3469 1081 1614 3518 3518 7484]
[ 872 1167 3469 1081 1614 3518 3518 7484 4411]
[1167 3469 1081 1614 3518 3518 7484 4411 7613]]
pos
(9, 3)
(9, 3)
(3, 9)
[[ 63 62 107 113 16 119 100 100 68]
[ 62 107 113 16 119 100 100 68 68]
[107 113 16 119 100 100 68 68 78]]
label 6.0
[[ 0. 0. 0. ..., 0. 0. 0.]
[ 0. 0. 0. ..., 0. 0. 0.]
[ 0. 0. 0. ..., 0. 0. 0.]
...,
[ 0. 0. 0. ..., 0. 0. 0.]
[ 0. 0. 0. ..., 0. 0. 0.]
[ 0. 0. 0. ..., 0. 0. 0.]]
llllll
r
872
going
1167
to
3469
put
r
1167
to
3469
put
1081
her
r
3469
put
1081
her
1614
into
r
1081
her
1614
into
3518
a
r
1614
into
3518
a
3518
a
r
3518
a
3518
a
7484
nursing
r
3518
a
7484
nursing
4411
home
r
7484
nursing
4411
home
7613
and
r
4411
home
7613
and
1647
uh
words
(9, 3)
(9, 3)
(3, 9)
[[ 872 1167 3469 1081 1614 3518 3518 7484 4411]
[1167 3469 1081 1614 3518 3518 7484 4411 7613]
[3469 1081 1614 3518 3518 7484 4411 7613 1647]]
pos
(9, 3)
(9, 3)
(3, 9)
[[ 62 107 113 16 119 100 100 68 68]
[107 113 16 119 100 100 68 68 78]
[113 16 119 100 100 68 68 78 57]]
label 0.0
[[ 0. 0. 0. ..., 0. 0. 0.]
[ 0. 0. 0. ..., 0. 0. 0.]
[ 0. 0. 0. ..., 0. 0. 0.]
...,
[ 0. 0. 0. ..., 0. 0. 0.]
[ 0. 0. 0. ..., 0. 0. 0.]
[ 0. 0. 0. ..., 0. 0. 0.]]
llllll
r
1167
to
3469
put
1081
her
r
3469
put
1081
her
1614
into
r
1081
her
1614
into
3518
a
r
1614
into
3518
a
3518
a
r
3518
a
3518
a
7484
nursing
r
3518
a
7484
nursing
4411
home
r
7484
nursing
4411
home
7613
and
r
4411
home
7613
and
1647
uh
r
7613
and
1647
uh
2423
they
words
(9, 3)
(9, 3)
(3, 9)
[[1167 3469 1081 1614 3518 3518 7484 4411 7613]
[3469 1081 1614 3518 3518 7484 4411 7613 1647]
[1081 1614 3518 3518 7484 4411 7613 1647 2423]]
pos
(9, 3)
(9, 3)
(3, 9)
[[107 113 16 119 100 100 68 68 78]
[113 16 119 100 100 68 68 78 57]
[ 16 119 100 100 68 68 78 57 16]]
label 4.0
[[ 0. 0. 0. ..., 0. 0. 0.]
[ 0. 0. 0. ..., 0. 0. 0.]
[ 0. 0. 0. ..., 0. 0. 0.]
...,
[ 0. 0. 0. ..., 0. 0. 0.]
[ 0. 0. 0. ..., 0. 0. 0.]
[ 0. 0. 0. ..., 0. 0. 0.]]
llllll
r
3469
put
1081
her
1614
into
r
1081
her
1614
into
3518
a
r
1614
into
3518
a
3518
a
r
3518
a
3518
a
7484
nursing
r
3518
a
7484
nursing
4411
home
r
7484
nursing
4411
home
7613
and
r
4411
home
7613
and
1647
uh
r
7613
and
1647
uh
2423
they
r
1647
uh
2423
they
7351
when
words
(9, 3)
(9, 3)
(3, 9)
[[3469 1081 1614 3518 3518 7484 4411 7613 1647]
[1081 1614 3518 3518 7484 4411 7613 1647 2423]
[1614 3518 3518 7484 4411 7613 1647 2423 7351]]
pos
(9, 3)
(9, 3)
(3, 9)
[[113 16 119 100 100 68 68 78 57]
[ 16 119 100 100 68 68 78 57 16]
[119 100 100 68 68 78 57 16 22]]
label 12.0
[[ 0. 0. 0. ..., 0. 0. 0.]
[ 0. 0. 0. ..., 0. 0. 0.]
[ 0. 0. 0. ..., 0. 0. 0.]
...,
[ 0. 0. 0. ..., 0. 0. 0.]
[ 0. 0. 0. ..., 0. 0. 0.]
[ 0. 0. 0. ..., 0. 0. 0.]]
llllll
r
1081
her
1614
into
3518
a
r
1614
into
3518
a
3518
a
r
3518
a
3518
a
7484
nursing
r
3518
a
7484
nursing
4411
home
r
7484
nursing
4411
home
7613
and
r
4411
home
7613
and
1647
uh
r
7613
and
1647
uh
2423
they
r
1647
uh
2423
they
7351
when
r
2423
they
7351
when
2423
they
words
(9, 3)
(9, 3)
(3, 9)
[[1081 1614 3518 3518 7484 4411 7613 1647 2423]
[1614 3518 3518 7484 4411 7613 1647 2423 7351]
[3518 3518 7484 4411 7613 1647 2423 7351 2423]]
pos
(9, 3)
(9, 3)
(3, 9)
[[ 16 119 100 100 68 68 78 57 16]
[119 100 100 68 68 78 57 16 22]
[100 100 68 68 78 57 16 22 16]]
label 41.0
[[ 0. 0. 0. ..., 0. 0. 0.]
[ 0. 0. 0. ..., 0. 0. 0.]
[ 0. 0. 0. ..., 0. 0. 0.]
...,
[ 0. 0. 0. ..., 0. 0. 0.]
[ 0. 0. 0. ..., 0. 0. 0.]
[ 0. 0. 0. ..., 0. 0. 0.]]
llllll
r
1614
into
3518
a
3518
a
r
3518
a
3518
a
7484
nursing
r
3518
a
7484
nursing
4411
home
r
7484
nursing
4411
home
7613
and
r
4411
home
7613
and
1647
uh
r
7613
and
1647
uh
2423
they
r
1647
uh
2423
they
7351
when
r
2423
they
7351
when
2423
they
r
7351
when
2423
they
3469
put
words
(9, 3)
(9, 3)
(3, 9)
[[1614 3518 3518 7484 4411 7613 1647 2423 7351]
[3518 3518 7484 4411 7613 1647 2423 7351 2423]
[3518 7484 4411 7613 1647 2423 7351 2423 3469]]
pos
(9, 3)
(9, 3)
(3, 9)
[[119 100 100 68 68 78 57 16 22]
[100 100 68 68 78 57 16 22 16]
[100 68 68 78 57 16 22 16 63]]
label 4.0
[[ 0. 0. 0. ..., 0. 0. 0.]
[ 0. 0. 0. ..., 0. 0. 0.]
[ 0. 0. 0. ..., 0. 0. 0.]
...,
[ 0. 0. 0. ..., 0. 0. 0.]
[ 0. 0. 0. ..., 0. 0. 0.]
[ 0. 0. 0. ..., 0. 0. 0.]]
llllll
r
3518
a
3518
a
7484
nursing
r
3518
a
7484
nursing
4411
home
r
7484
nursing
4411
home
7613
and
r
4411
home
7613
and
1647
uh
r
7613
and
1647
uh
2423
they
r
1647
uh
2423
they
7351
when
r
2423
they
7351
when
2423
they
r
7351
when
2423
they
3469
put
r
2423
they
3469
put
1081
her
words
(9, 3)
(9, 3)
(3, 9)
[[3518 3518 7484 4411 7613 1647 2423 7351 2423]
[3518 7484 4411 7613 1647 2423 7351 2423 3469]
[7484 4411 7613 1647 2423 7351 2423 3469 1081]]
pos
(9, 3)
(9, 3)
(3, 9)
[[100 100 68 68 78 57 16 22 16]
[100 68 68 78 57 16 22 16 63]
[ 68 68 78 57 16 22 16 63 16]]
label 4.0
[[ 0. 0. 0. ..., 0. 0. 0.]
[ 0. 0. 0. ..., 0. 0. 0.]
[ 0. 0. 0. ..., 0. 0. 0.]
...,
[ 0. 0. 0. ..., 0. 0. 0.]
[ 0. 0. 0. ..., 0. 0. 0.]
[ 0. 0. 0. ..., 0. 0. 0.]]
current train_loss (may include reg)is 0.0
In [ ]:
test_file2 = "../data/audio/switchboard/4619B.npy"
In [14]:
dialogues = [("",test_file), ("",test_file2)]
In [15]:
data = np.load(test_file2)
In [16]:
data[:][350+2]
Out[16]:
array([ 57., 57., 124., 57., 0., 58., 119., 57., 100.,
68., 119., 100., 68., 57., 57., 16., 57., 0.,
58., 0., 58., 17., 82., 78., 16., 67., 119.,
16., 78., 71., 119., 4., 68., 57., 78., 57.,
100., 19., 119., 0., 65., 57., 64., 119., 100.,
16., 65., 80., 113., 119., 100., 68., 22., 16.,
63., 100., 68., 119., 68., 78., 57., 22., 16.,
63., 100., 19., 107., 113., 107., 113., 16., 57.,
17., 4., 78., 22., 49., 63., 57., 4., 78.,
4., 22., 49., 63., 4., 78., 4., 19., 57.,
62., 16., 65., 16., 113., 100., 68., 119., 100.,
16., 57., 57., 119., 100., 68., 119., 68., 16.,
65., 65., 16., 113., 119., 119., 100., 68., 119.,
100., 82., 3., 96., 17., 64., 100., 68., 119.,
119., 100., 19., 5., 65., 16., 113., 107., 113.,
5., 80., 113., 119., 17., 4., 68., 57., 57.,
57., 16., 71., 4., 16., 63., 62., 17., 119.,
26., 113., 49., 52., 113., 100., 4., 21., 14.,
68., 57., 119., 17., 62., 107., 113., 119., 46.,
19., 22., 100., 19., 65., 17., 62., 119., 100.,
68., 17., 62., 107., 113., 5., 16., 52., 113.,
119., 17., 4., 68., 57., 57., 17., 16., 67.,
16., 100., 68., 119., 68., 57., 16., 16., 65.,
68., 119., 5., 96., 62., 67., 5., 16., 65.,
65., 64., 16., 63., 107., 17., 113., 119., 100.,
4., 19., 5., 63., 119., 4., 19., 63., 19.,
5., 57., 63., 17., 4., 119., 0., 19., 57.,
4., 78., 4., 119., 16., 18., 113., 113., 119.,
16., 78., 16., 17., 65., 100., 68., 119., 19.,
85., 19., 65., 64., 119., 4., 19., 119., 16.,
65., 16., 65., 57., 100., 68., 119., 4., 19.,
57., 100., 82., 119., 0., 55., 19., 5., 63.,
119., 119., 16., 63., 17., 64., 0., 19., 78.,
16., 63., 119., 17., 119., 17., 22., 16., 63.,
17., 119., 68., 57., 78., 57., 16., 65., 16.,
16., 65., 83., 16., 63., 17., 17., 119., 100.,
19., 119., 16., 63., 62., 119., 16., 52., 113.,
17., 78., 16., 65., 119., 100., 4., 68., 119.,
100., 67., 17., 4., 119., 16., 65., 119., 19.,
119., 119., 113., 16., 119., 17., 78., 49., 54.,
17., 4.])
In [17]:
rnn.fit(dialogues, 0.005, acoustic=True, load_data=True)
---------------------------------------------------------------------------
KeyboardInterrupt Traceback (most recent call last)
<ipython-input-17-a9680c9c5f64> in <module>()
----> 1 rnn.fit(dialogues, 0.005, acoustic=True, load_data=True)
/Users/julianhough/git/simple_rnn_disf/rnn_disf_detection/rnn/elman.py in fit(self, dialogues, lr, acoustic, load_data)
230 test+=1 #TODO for testing
231 #if test > 50: break #TODO for testing
--> 232 x = self.train(lex_data[start:stop+1],pos_data[start:stop+1],np.asarray(acoustic_data[:,start:stop+1],dtype='float32'),labels[stop],lr)
233 loss+=x
234 self.normalize()
/Library/Python/2.7/site-packages/theano/compile/function_module.pyc in __call__(self, *args, **kwargs)
593 t0_fn = time.time()
594 try:
--> 595 outputs = self.fn()
596 except Exception:
597 if hasattr(self.fn, 'position_of_error'):
KeyboardInterrupt:
In [12]:
test = 0
for start,stop in indices:
test+=1 #TODO for testing
if test > 50: break #TODO for testing
#print type(lex_data[start:stop+1]), lex_data[start:stop+1].shape
#print type(pos_data[start:stop+1]), pos_data[start:stop+1].shape
#print type(acoustic_data[:,start:stop+1]), acoustic_data[:,start:stop+1].shape
#print type(labels[stop])
x = rnn.train(lex_data[start:stop+1],pos_data[start:stop+1],np.asarray(acoustic_data[:,start:stop+1],dtype='float32'),labels[stop],0.005)
#loss+=x
rnn.normalize()
In [13]:
#np.swapaxes(acoustic_data[:,start:stop+1], 0, 1).shape
print x
2.20022251803
In [137]:
print fr.shape, ac.shape, le.shape, po.shape, ind.shape, labels.shape
(1047,) (350, 1047) (1047, 3) (1047, 3) (1047, 2) (354,)
In [146]:
start, stop = ind[(100)]
In [148]:
le[start:stop]
Out[148]:
array([[7613, 4904, 6999],
[4904, 6999, 707],
[6999, 707, 355],
[ 707, 355, 3064],
[ 355, 3064, 4258],
[3064, 4258, 1904],
[4258, 1904, 2980],
[1904, 2980, 840]])
In [30]:
groundtruth_filename = "../data/disfluency_detection/switchboard/swbd_heldout_partial_data.csv"
id,ind, gold_words,_,gold_labels = get_tag_data_from_corpus_file(groundtruth_filename, representation="1_trp")
loading data ../data/disfluency_detection/switchboard/swbd_heldout_partial_data.csv
loaded 5732 sequences
In [31]:
turn_end = True
limit = 100
for word, label in zip(sum(gold_words,()),sum(gold_labels,())):
if turn_end == True:
assert("<t" in label)
if turn_end == False:
assert("<c" in label)
if "c>" in label:
turn_end = False
else:
turn_end = True
print word, label
limit-=1
if limit <0 : break
do <f/><tc>
you <f/><cc>
know <f/><cc>
anyone <f/><cc>
that <f/><cc>
uh <e/><cc>
is <rms id="7"/><cc>
is <rps id="7"/><rpnrep id="7"/><cc>
in <f/><cc>
a <f/><cc>
nursing <f/><cc>
home <f/><cc>
or <f/><cc>
has <f/><cc>
ever <f/><cc>
been <f/><cc>
in <f/><cc>
$unc$one <f/><ct>
no <f/><tt>
but <f/><tc>
i <rms id="2"/><cc>
my <rps id="2"/><cc>
grandparents <rpnsub id="2"/><cc>
were <f/><cc>
looking <f/><cc>
into <f/><cc>
it <f/><cc>
before <f/><ct>
so <f/><tc>
i <f/><cc>
know <f/><cc>
what <f/><cc>
theyve <f/><cc>
said <f/><ct>
uh-huh <f/><tt>
uh-huh <f/><tt>
well <e/><tc>
im <f/><cc>
trying <f/><cc>
to <f/><cc>
think <f/><ct>
my <f/><tc>
uh <e/><cc>
uh <e/><cc>
wifes <f/><cc>
grandmother <f/><cc>
had <f/><cc>
alzheimers <f/><ct>
and <f/><tc>
they <f/><cc>
were <f/><cc>
going <f/><cc>
to <f/><cc>
put <f/><cc>
her <f/><cc>
into <f/><cc>
a <rms id="9"/><cc>
a <rps id="9"/><rpnrep id="9"/><cc>
nursing <f/><cc>
home <f/><ct>
and <f/><tc>
uh <e/><cc>
they <rms id="3"/><cc>
when <rps id="3"/><cc>
they <rpnsub id="3"/><cc>
put <f/><cc>
her <f/><cc>
in <f/><cc>
she <f/><cc>
had <f/><cc>
all <f/><cc>
kinds <f/><cc>
of <f/><cc>
trouble <f/><ct>
and <f/><tc>
the <f/><cc>
nursing <f/><cc>
home <f/><cc>
made <f/><cc>
them <f/><cc>
come <f/><cc>
and <f/><cc>
take <f/><cc>
her <f/><cc>
back <f/><cc>
because <f/><cc>
she <f/><cc>
was <f/><cc>
being <f/><cc>
a <rms id="16"/><rms id="19"/><rms id="20"/><cc>
a <rps id="16"/><rm id="19"/><rm id="20"/><rpnrep id="16"/><cc>
you <e/><i id="19"/><cc>
know <e/><i id="19"/><cc>
a <rps id="19"/><rm id="20"/><rpnsub id="19"/><cc>
a <rps id="20"/><rpnsub id="20"/><cc>
nuisance <f/><cc>
or <f/><cc>
worse <f/><cc>
than <f/><cc>
a <f/><cc>
nuisance <f/><ct>
In [ ]:
Content source: dsg-bielefeld/deep_disfluency
Similar notebooks: