pydev debugger: starting
Building [1, 2, 3, 4, 5, 6, 7, 8, 9, 10]-Gram Models
Computing Perplexities of Test Data, using 1-Gram Model
Perplexity without smoothing of Test Data using the 1-gram English Language Model is: 19.8438187378
Perplexity without smoothing of Test Data using the 1-gram Spanish Language Model is: 23.5110136799
Perplexity without smoothing of Test Data using the 1-gram German Language Model is: 22.3528691816

Perplexity with Laplace smoothing of Test Data using the 1-gram  English Language Model is: 19.8426297183
Perplexity with Laplace smoothing of Test Data using the 1-gram  Spanish Language Model is: 24.0620519762
Perplexity with Laplace smoothing of Test Data using the 1-gram  German Language Model is: 22.8988420846

Perplexity with Backoff  of Test Data using the 1-gram  English Language Model is: 19.8438189083
Perplexity with Backoff  of Test Data using the 1-gram  Spanish Language Model is: 23.5110138978
Perplexity with Backoff  of Test Data using the 1-gram  German Language Model is: 22.3528693473

Perplexity with Backoff smoothing of Test Data using the 1-gram  English Language Model is: 19.8426297183
Perplexity with Backoff smoothing of Test Data using the 1-gram  Spanish Language Model is: 24.0620519762
Perplexity with Backoff smoothing of Test Data using the 1-gram  German Language Model is: 22.8988420846

Perplexity with Weighted Backoff  of Test Data using the 1-gram  English Language Model is: 19.8438189083
Perplexity with Weighted Backoff  of Test Data using the 1-gram  Spanish Language Model is: 23.5110138978
Perplexity with Weighted Backoff  of Test Data using the 1-gram  German Language Model is: 22.3528693473

Perplexity with Frequency Backoff  of Test Data using the 1-gram  English Language Model is: 19.8438189083
Perplexity with Frequency Backoff  of Test Data using the 1-gram  Spanish Language Model is: 23.5110138978
Perplexity with Frequency Backoff  of Test Data using the 1-gram  German Language Model is: 22.3528693473
...........................................................................................

Computing Perplexities of Test Data, using 2-Gram Model
Perplexity without smoothing of Test Data using the 2-gram English Language Model is: 11.472994994
Perplexity without smoothing of Test Data using the 2-gram Spanish Language Model is: 21.6131995586
Perplexity without smoothing of Test Data using the 2-gram German Language Model is: 20.5028813961

Perplexity with Laplace smoothing of Test Data using the 2-gram  English Language Model is: 11.8977864508
Perplexity with Laplace smoothing of Test Data using the 2-gram  Spanish Language Model is: 25.2799895814
Perplexity with Laplace smoothing of Test Data using the 2-gram  German Language Model is: 22.6278713105

Perplexity with Backoff  of Test Data using the 2-gram  English Language Model is: 13.6832745589
Perplexity with Backoff  of Test Data using the 2-gram  Spanish Language Model is: 19.6832244875
Perplexity with Backoff  of Test Data using the 2-gram  German Language Model is: 18.6943357268

Perplexity with Backoff smoothing of Test Data using the 2-gram  English Language Model is: 13.721304978
Perplexity with Backoff smoothing of Test Data using the 2-gram  Spanish Language Model is: 19.9076115761
Perplexity with Backoff smoothing of Test Data using the 2-gram  German Language Model is: 18.9954417949

Perplexity with Weighted Backoff  of Test Data using the 2-gram  English Language Model is: 12.301010824
Perplexity with Weighted Backoff  of Test Data using the 2-gram  Spanish Language Model is: 20.3645436705
Perplexity with Weighted Backoff  of Test Data using the 2-gram  German Language Model is: 19.300225139

Perplexity with Frequency Backoff  of Test Data using the 2-gram  English Language Model is: 12.0590532805
Perplexity with Frequency Backoff  of Test Data using the 2-gram  Spanish Language Model is: 24.6959232474
Perplexity with Frequency Backoff  of Test Data using the 2-gram  German Language Model is: 22.591354448
...........................................................................................

Computing Perplexities of Test Data, using 3-Gram Model
Perplexity without smoothing of Test Data using the 3-gram English Language Model is: 7.81166439802
Perplexity without smoothing of Test Data using the 3-gram Spanish Language Model is: 14.0391323423
Perplexity without smoothing of Test Data using the 3-gram German Language Model is: 14.8478629854

Perplexity with Laplace smoothing of Test Data using the 3-gram  English Language Model is: 9.98399329675
Perplexity with Laplace smoothing of Test Data using the 3-gram  Spanish Language Model is: 28.3312650765
Perplexity with Laplace smoothing of Test Data using the 3-gram  German Language Model is: 27.9319754137

Perplexity with Backoff  of Test Data using the 3-gram  English Language Model is: 10.5024305058
Perplexity with Backoff  of Test Data using the 3-gram  Spanish Language Model is: 18.7522528941
Perplexity with Backoff  of Test Data using the 3-gram  German Language Model is: 17.5628568363

Perplexity with Backoff smoothing of Test Data using the 3-gram  English Language Model is: 10.9587952705
Perplexity with Backoff smoothing of Test Data using the 3-gram  Spanish Language Model is: 18.9053295769
Perplexity with Backoff smoothing of Test Data using the 3-gram  German Language Model is: 18.2766615725

Perplexity with Weighted Backoff  of Test Data using the 3-gram  English Language Model is: 9.21065704719
Perplexity with Weighted Backoff  of Test Data using the 3-gram  Spanish Language Model is: 22.0439424779
Perplexity with Weighted Backoff  of Test Data using the 3-gram  German Language Model is: 20.0617180116

Perplexity with Frequency Backoff  of Test Data using the 3-gram  English Language Model is: 9.61224013886
Perplexity with Frequency Backoff  of Test Data using the 3-gram  Spanish Language Model is: 26.9073714737
Perplexity with Frequency Backoff  of Test Data using the 3-gram  German Language Model is: 25.5276443831
...........................................................................................

Computing Perplexities of Test Data, using 4-Gram Model
Perplexity without smoothing of Test Data using the 4-gram English Language Model is: 4.67169464377
Perplexity without smoothing of Test Data using the 4-gram Spanish Language Model is: 8.25662298234
Perplexity without smoothing of Test Data using the 4-gram German Language Model is: 7.53156525826

Perplexity with Laplace smoothing of Test Data using the 4-gram  English Language Model is: 10.9784646821
Perplexity with Laplace smoothing of Test Data using the 4-gram  Spanish Language Model is: 40.4905958122
Perplexity with Laplace smoothing of Test Data using the 4-gram  German Language Model is: 39.6460204511

Perplexity with Backoff  of Test Data using the 4-gram  English Language Model is: 8.36254648082
Perplexity with Backoff  of Test Data using the 4-gram  Spanish Language Model is: 18.6374331702
Perplexity with Backoff  of Test Data using the 4-gram  German Language Model is: 17.7345654232

Perplexity with Backoff smoothing of Test Data using the 4-gram  English Language Model is: 10.0573495325
Perplexity with Backoff smoothing of Test Data using the 4-gram  Spanish Language Model is: 19.4661787016
Perplexity with Backoff smoothing of Test Data using the 4-gram  German Language Model is: 19.4342122993

Perplexity with Weighted Backoff  of Test Data using the 4-gram  English Language Model is: 7.29784758118
Perplexity with Weighted Backoff  of Test Data using the 4-gram  Spanish Language Model is: 25.1110109294
Perplexity with Weighted Backoff  of Test Data using the 4-gram  German Language Model is: 23.7828162981

Perplexity with Frequency Backoff  of Test Data using the 4-gram  English Language Model is: 8.74325932464
Perplexity with Frequency Backoff  of Test Data using the 4-gram  Spanish Language Model is: 30.0965155275
Perplexity with Frequency Backoff  of Test Data using the 4-gram  German Language Model is: 27.4118282158
...........................................................................................

Computing Perplexities of Test Data, using 5-Gram Model
Perplexity without smoothing of Test Data using the 5-gram English Language Model is: 2.93392817549
Perplexity without smoothing of Test Data using the 5-gram Spanish Language Model is: 3.89606808228
Perplexity without smoothing of Test Data using the 5-gram German Language Model is: 3.56751535106

Perplexity with Laplace smoothing of Test Data using the 5-gram  English Language Model is: 16.0987493061
Perplexity with Laplace smoothing of Test Data using the 5-gram  Spanish Language Model is: 50.725014142
Perplexity with Laplace smoothing of Test Data using the 5-gram  German Language Model is: 51.2963354031

Perplexity with Backoff  of Test Data using the 5-gram  English Language Model is: 7.22489803229
Perplexity with Backoff  of Test Data using the 5-gram  Spanish Language Model is: 18.7066272375
Perplexity with Backoff  of Test Data using the 5-gram  German Language Model is: 17.8402014246

Perplexity with Backoff smoothing of Test Data using the 5-gram  English Language Model is: 10.3026742156
Perplexity with Backoff smoothing of Test Data using the 5-gram  Spanish Language Model is: 20.818514702
Perplexity with Backoff smoothing of Test Data using the 5-gram  German Language Model is: 21.1700155527

Perplexity with Weighted Backoff  of Test Data using the 5-gram  English Language Model is: 6.51919338023
Perplexity with Weighted Backoff  of Test Data using the 5-gram  Spanish Language Model is: 27.427623582
Perplexity with Weighted Backoff  of Test Data using the 5-gram  German Language Model is: 26.0335229832

Perplexity with Frequency Backoff  of Test Data using the 5-gram  English Language Model is: 8.37884264621
Perplexity with Frequency Backoff  of Test Data using the 5-gram  Spanish Language Model is: 30.8518882531
Perplexity with Frequency Backoff  of Test Data using the 5-gram  German Language Model is: 28.051819397
...........................................................................................

Computing Perplexities of Test Data, using 6-Gram Model
Perplexity without smoothing of Test Data using the 6-gram English Language Model is: 2.25149553499
Perplexity without smoothing of Test Data using the 6-gram Spanish Language Model is: 2.01887648542
Perplexity without smoothing of Test Data using the 6-gram German Language Model is: 2.03183417089

Perplexity with Laplace smoothing of Test Data using the 6-gram  English Language Model is: 23.6806402862
Perplexity with Laplace smoothing of Test Data using the 6-gram  Spanish Language Model is: 56.9703724247
Perplexity with Laplace smoothing of Test Data using the 6-gram  German Language Model is: 57.6747926412

Perplexity with Backoff  of Test Data using the 6-gram  English Language Model is: 6.70922356549
Perplexity with Backoff  of Test Data using the 6-gram  Spanish Language Model is: 18.6775386416
Perplexity with Backoff  of Test Data using the 6-gram  German Language Model is: 17.9103834114

Perplexity with Backoff smoothing of Test Data using the 6-gram  English Language Model is: 10.9777486335
Perplexity with Backoff smoothing of Test Data using the 6-gram  Spanish Language Model is: 22.4013343757
Perplexity with Backoff smoothing of Test Data using the 6-gram  German Language Model is: 22.9281625568

Perplexity with Weighted Backoff  of Test Data using the 6-gram  English Language Model is: 6.34502742254
Perplexity with Weighted Backoff  of Test Data using the 6-gram  Spanish Language Model is: 28.298924725
Perplexity with Weighted Backoff  of Test Data using the 6-gram  German Language Model is: 27.0696280226

Perplexity with Frequency Backoff  of Test Data using the 6-gram  English Language Model is: 8.25265135654
Perplexity with Frequency Backoff  of Test Data using the 6-gram  Spanish Language Model is: 30.9787073844
Perplexity with Frequency Backoff  of Test Data using the 6-gram  German Language Model is: 28.3652523613
...........................................................................................

Computing Perplexities of Test Data, using 7-Gram Model
Perplexity without smoothing of Test Data using the 7-gram English Language Model is: 1.9076977204
Perplexity without smoothing of Test Data using the 7-gram Spanish Language Model is: 1.54248226348
Perplexity without smoothing of Test Data using the 7-gram German Language Model is: 1.42999533389

Perplexity with Laplace smoothing of Test Data using the 7-gram  English Language Model is: 31.7419274155
Perplexity with Laplace smoothing of Test Data using the 7-gram  Spanish Language Model is: 60.4565434226
Perplexity with Laplace smoothing of Test Data using the 7-gram  German Language Model is: 60.3811414033

Perplexity with Backoff  of Test Data using the 7-gram  English Language Model is: 6.48723043758
Perplexity with Backoff  of Test Data using the 7-gram  Spanish Language Model is: 18.6698329022
Perplexity with Backoff  of Test Data using the 7-gram  German Language Model is: 17.9235787537

Perplexity with Backoff smoothing of Test Data using the 7-gram  English Language Model is: 11.8123884599
Perplexity with Backoff smoothing of Test Data using the 7-gram  Spanish Language Model is: 23.9696515502
Perplexity with Backoff smoothing of Test Data using the 7-gram  German Language Model is: 24.5763133761

Perplexity with Weighted Backoff  of Test Data using the 7-gram  English Language Model is: 6.40240723594
Perplexity with Weighted Backoff  of Test Data using the 7-gram  Spanish Language Model is: 28.6502895454
Perplexity with Weighted Backoff  of Test Data using the 7-gram  German Language Model is: 27.3936760965

Perplexity with Frequency Backoff  of Test Data using the 7-gram  English Language Model is: 8.21543851212
Perplexity with Frequency Backoff  of Test Data using the 7-gram  Spanish Language Model is: 31.1387404877
Perplexity with Frequency Backoff  of Test Data using the 7-gram  German Language Model is: 28.6098347341
...........................................................................................

Computing Perplexities of Test Data, using 8-Gram Model
Perplexity without smoothing of Test Data using the 8-gram English Language Model is: 1.62013995181
Perplexity without smoothing of Test Data using the 8-gram Spanish Language Model is: 1.29045712632
Perplexity without smoothing of Test Data using the 8-gram German Language Model is: 1.28203158121

Perplexity with Laplace smoothing of Test Data using the 8-gram  English Language Model is: 39.3532862563
Perplexity with Laplace smoothing of Test Data using the 8-gram  Spanish Language Model is: 62.3094914469
Perplexity with Laplace smoothing of Test Data using the 8-gram  German Language Model is: 61.6188912695

Perplexity with Backoff  of Test Data using the 8-gram  English Language Model is: 6.39167923297
Perplexity with Backoff  of Test Data using the 8-gram  Spanish Language Model is: 18.6818587723
Perplexity with Backoff  of Test Data using the 8-gram  German Language Model is: 17.9281039025

Perplexity with Backoff smoothing of Test Data using the 8-gram  English Language Model is: 12.7114100027
Perplexity with Backoff smoothing of Test Data using the 8-gram  Spanish Language Model is: 25.4512680629
Perplexity with Backoff smoothing of Test Data using the 8-gram  German Language Model is: 26.0916212817

Perplexity with Weighted Backoff  of Test Data using the 8-gram  English Language Model is: 6.51410939028
Perplexity with Weighted Backoff  of Test Data using the 8-gram  Spanish Language Model is: 28.8352728388
Perplexity with Weighted Backoff  of Test Data using the 8-gram  German Language Model is: 27.5175082937

Perplexity with Frequency Backoff  of Test Data using the 8-gram  English Language Model is: 8.19504747091
Perplexity with Frequency Backoff  of Test Data using the 8-gram  Spanish Language Model is: 31.2901019326
Perplexity with Frequency Backoff  of Test Data using the 8-gram  German Language Model is: 28.6311277114
...........................................................................................

Computing Perplexities of Test Data, using 9-Gram Model
Perplexity without smoothing of Test Data using the 9-gram English Language Model is: 1.49878250894
Perplexity without smoothing of Test Data using the 9-gram Spanish Language Model is: 1.30790900375
Perplexity without smoothing of Test Data using the 9-gram German Language Model is: 1.42219835417

Perplexity with Laplace smoothing of Test Data using the 9-gram  English Language Model is: 45.6175083672
Perplexity with Laplace smoothing of Test Data using the 9-gram  Spanish Language Model is: 63.2414277016
Perplexity with Laplace smoothing of Test Data using the 9-gram  German Language Model is: 62.2456314296

Perplexity with Backoff  of Test Data using the 9-gram  English Language Model is: 6.35226630677
Perplexity with Backoff  of Test Data using the 9-gram  Spanish Language Model is: 18.6922001221
Perplexity with Backoff  of Test Data using the 9-gram  German Language Model is: 17.9328298955

Perplexity with Backoff smoothing of Test Data using the 9-gram  English Language Model is: 13.6259335573
Perplexity with Backoff smoothing of Test Data using the 9-gram  Spanish Language Model is: 26.8290074329
Perplexity with Backoff smoothing of Test Data using the 9-gram  German Language Model is: 27.4797326294

Perplexity with Weighted Backoff  of Test Data using the 9-gram  English Language Model is: 6.61966705469
Perplexity with Weighted Backoff  of Test Data using the 9-gram  Spanish Language Model is: 28.9211383606
Perplexity with Weighted Backoff  of Test Data using the 9-gram  German Language Model is: 27.5765892403

Perplexity with Frequency Backoff  of Test Data using the 9-gram  English Language Model is: 8.18334512157
Perplexity with Frequency Backoff  of Test Data using the 9-gram  Spanish Language Model is: 31.4402897367
Perplexity with Frequency Backoff  of Test Data using the 9-gram  German Language Model is: 28.6616499576
...........................................................................................

Computing Perplexities of Test Data, using 10-Gram Model
Perplexity without smoothing of Test Data using the 10-gram English Language Model is: 1.41477624902
Perplexity without smoothing of Test Data using the 10-gram Spanish Language Model is: 1.23796101396
Perplexity without smoothing of Test Data using the 10-gram German Language Model is: 1.4722634251

Perplexity with Laplace smoothing of Test Data using the 10-gram  English Language Model is: 50.3983159243
Perplexity with Laplace smoothing of Test Data using the 10-gram  Spanish Language Model is: 63.7304402328
Perplexity with Laplace smoothing of Test Data using the 10-gram  German Language Model is: 62.5928158332

Perplexity with Backoff  of Test Data using the 10-gram  English Language Model is: 6.3397433696
Perplexity with Backoff  of Test Data using the 10-gram  Spanish Language Model is: 18.6973641275
Perplexity with Backoff  of Test Data using the 10-gram  German Language Model is: 17.9356666273

Perplexity with Backoff smoothing of Test Data using the 10-gram  English Language Model is: 14.5276598567
Perplexity with Backoff smoothing of Test Data using the 10-gram  Spanish Language Model is: 28.1043099284
Perplexity with Backoff smoothing of Test Data using the 10-gram  German Language Model is: 28.7529628831

Perplexity with Weighted Backoff  of Test Data using the 10-gram  English Language Model is: 6.70713051866
Perplexity with Weighted Backoff  of Test Data using the 10-gram  Spanish Language Model is: 28.9576928239
Perplexity with Weighted Backoff  of Test Data using the 10-gram  German Language Model is: 27.6057208901

Perplexity with Frequency Backoff  of Test Data using the 10-gram  English Language Model is: 8.19053170107
Perplexity with Frequency Backoff  of Test Data using the 10-gram  Spanish Language Model is: 31.58282259
Perplexity with Frequency Backoff  of Test Data using the 10-gram  German Language Model is: 28.6699779465
...........................................................................................

 Average Perplexity............................................................................

[0.14431927317202375, 0.16802852963660886, 0.1443192729251781, 0.16802852963660886, 0.1443192729251781, 0.1443192729251781]

[0.6069388129657534, 0.6982454034361751, 0.33781939799579797, 0.34870086456642424, 0.47727460738063576, 0.6722870454934877]

[0.6142342206579257, 1.0358858960827173, 0.5469433751350063, 0.5283920413592789, 0.825564446202119, 1.0030438191822482]

[0.523537392321358, 1.2945945628277615, 0.7765808554819396, 0.6595531198053204, 1.2085554212249061, 1.1894089638111431]

[0.23957641761102533, 1.15327765362489, 0.9276330175337322, 0.7118107761214851, 1.4107172056547237, 1.2559159283571324]

[-0.10585463817377616, 0.8840179309518517, 1.0028679103851872, 0.7248728716154493, 1.4729466622107381, 1.2786979735259387]

[-0.2503650532990809, 0.6436623837045716, 1.0366773545632753, 0.7201369595806546, 1.4760638181193977, 1.2900864951736946]

[-0.23079128070879176, 0.45396164754062873, 1.0519642083170488, 0.7066898877262215, 1.4642404283865729, 1.2953681565720232]

[-0.0943363714065471, 0.3187323732785652, 1.05855807028548, 0.6894913472040647, 1.450724903053488, 1.29972408083823]

[-0.04683806859527055, 0.2256986524592688, 1.0607486291214518, 0.6712774721362276, 1.4387582462715198, 1.301253130749999]

 Average Descrimination Power............................................................................

[1, 1, 1, 1, 1, 1]

[1, 1, 1, 1, 1, 1]

[1, 1, 1, 1, 1, 1]

[1, 1, 1, 1, 1, 1]

[1, 1, 1, 1, 1, 1]

[-1, 1, 1, 1, 1, 1]

[-1, 1, 1, 1, 1, 1]

[-1, 1, 1, 1, 1, 1]

[-1, 1, 1, 1, 1, 1]

[0, 1, 1, 1, 1, 1]

