Skip to content

Instantly share code, notes, and snippets.

@thomasbrus

thomasbrus/ann.txt

Created October 15, 2014 18:50
Show Gist options
  • Save thomasbrus/13b150ebbee19d31754c to your computer and use it in GitHub Desktop.
Save thomasbrus/13b150ebbee19d31754c to your computer and use it in GitHub Desktop.
Download ZIP
Predictions of Scrabble game outcomes
Raw
ann.txt
=== Run information ===
Scheme:weka.classifiers.functions.MultilayerPerceptron -L 0.3 -M 0.2 -N 500 -V 0 -S 0 -E 20 -H a
Relation: scrabble-game-states
Instances: 7273
Attributes: 8
number_of_plays
first_player_current_score
second_player_current_score
current_score_difference
first_player_average_score
second_player_average_score
average_score_difference
won_by_first_player
Test mode:split 66.0% train, remainder test
=== Classifier model (full training set) ===
Sigmoid Node 0
Inputs Weights
Threshold -3.6967867902734404
Node 2 1.2819531152694352
Node 3 2.6976139033825586
Node 4 1.1540136937809136
Node 5 2.434043703633358
Sigmoid Node 1
Inputs Weights
Threshold 3.6967867902734355
Node 2 -1.2819531152694343
Node 3 -2.697613903382554
Node 4 -1.154013693780913
Node 5 -2.434043703633358
Sigmoid Node 2
Inputs Weights
Threshold -4.19318814798716
Attrib number_of_plays -1.8868047692736998
Attrib first_player_current_score 9.574260714093745
Attrib second_player_current_score -12.20285791983022
Attrib current_score_difference 24.33688222309119
Attrib first_player_average_score -1.1542083696632148
Attrib second_player_average_score -5.773000736693814
Attrib average_score_difference 3.3352128798022824
Sigmoid Node 3
Inputs Weights
Threshold 0.859709251019633
Attrib number_of_plays -5.80287294462652
Attrib first_player_current_score 5.020199743979442
Attrib second_player_current_score -13.004464318000934
Attrib current_score_difference 20.105453417291802
Attrib first_player_average_score 0.13187264612294883
Attrib second_player_average_score -4.3981810475931
Attrib average_score_difference 3.1859363182942984
Sigmoid Node 4
Inputs Weights
Threshold -3.5818639280457414
Attrib number_of_plays 1.6262929009481193
Attrib first_player_current_score 8.806253768443773
Attrib second_player_current_score -6.535346766709625
Attrib current_score_difference 17.18614422478288
Attrib first_player_average_score 1.875633524088512
Attrib second_player_average_score -6.098564781310641
Attrib average_score_difference 5.443697003271651
Sigmoid Node 5
Inputs Weights
Threshold -5.8324200614739405
Attrib number_of_plays -0.7414136197664812
Attrib first_player_current_score 10.199957080710858
Attrib second_player_current_score -3.0650421827167813
Attrib current_score_difference 14.641791872721141
Attrib first_player_average_score -0.7566679673437983
Attrib second_player_average_score -4.126509540099244
Attrib average_score_difference 2.4431695671142517
Class true
Input
Node 0
Class false
Input
Node 1
Time taken to build model: 4.97 seconds
=== Predictions on test split ===
inst#, actual, predicted, error, probability distribution (number_of_plays,first_player_current_score,second_player_current_score,current_score_difference,first_player_average_score,second_player_average_score,average_score_difference)
12 2:false 2:false 0.465 *0.535 (-1,-0.910387,-0.931689,0.051282,-0.435897,-0.581395,0.084746)
13 2:false 2:false 0.204 *0.796 (-0.925926,-0.910387,-0.685009,-0.226496,-0.717949,-0.034884,-0.5)
14 2:false 2:false 0.205 *0.795 (-0.851852,-0.808554,-0.628083,-0.183761,-0.598291,-0.24031,-0.271186)
15 2:false 2:false 0.205 *0.795 (-0.777778,-0.702648,-0.555977,-0.153846,-0.532051,-0.319767,-0.169492)
16 2:false 2:false 0.204 *0.796 (-0.703704,-0.604888,-0.43074,-0.192308,-0.502564,-0.302326,-0.162712)
17 2:false 2:false 0.175 *0.825 (-0.62963,-0.539715,-0.286528,-0.286325,-0.517094,-0.271318,-0.194915)
18 2:false 2:false 0.188 *0.812 (-0.555556,-0.454175,-0.248577,-0.239316,-0.509158,-0.342193,-0.138015)
19 2:false 2:false 0.068 *0.932 (-0.481481,-0.356415,-0.104364,-0.299145,-0.49359,-0.313953,-0.148305)
20 2:false 2:false 0.01 *0.99 (-0.407407,-0.295316,-0.013283,-0.337607,-0.507123,-0.328165,-0.146893)
21 2:false 2:false 0.007 *0.993 (-0.333333,-0.185336,0.062619,-0.307692,-0.487179,-0.348837,-0.118644)
22 2:false 2:false 0.007 *0.993 (-0.259259,-0.107943,0.108159,-0.277778,-0.48951,-0.382664,-0.095532)
23 2:false 2:false 0.004 *0.996 (-0.185185,-0.022403,0.184061,-0.273504,-0.487179,-0.395349,-0.084746)
24 2:false 2:false 0.003 *0.997 (-0.111111,0.03055,0.244782,-0.286325,-0.500986,-0.413238,-0.080834)
25 2:false 2:false 0.002 *0.998 (-0.037037,0.144603,0.324478,-0.25641,-0.485348,-0.420266,-0.065375)
26 2:false 2:false 0.002 *0.998 (0.037037,0.193483,0.411765,-0.303419,-0.499145,-0.423256,-0.072316)
27 2:false 2:false 0.002 *0.998 (0.111111,0.242363,0.45351,-0.299145,-0.511218,-0.443314,-0.065678)
28 2:false 2:false 0.002 *0.998 (0.185185,0.307536,0.476281,-0.25641,-0.515837,-0.467852,-0.050847)
29 2:false 2:false 0.002 *0.998 (0.259259,0.315682,0.476281,-0.247863,-0.539886,-0.497416,-0.045198)
...
=== Evaluation on test split ===
=== Summary ===
Correctly Classified Instances 1884 76.1828 %
Incorrectly Classified Instances 589 23.8172 %
Kappa statistic 0.525
Mean absolute error 0.2935
Root mean squared error 0.4001
Relative absolute error 58.7733 %
Root relative squared error 79.9396 %
Total Number of Instances 2473
=== Detailed Accuracy By Class ===
TP Rate FP Rate Precision Recall F-Measure ROC Area Class
0.683 0.157 0.819 0.683 0.745 0.851 true
0.843 0.317 0.72 0.843 0.777 0.851 false
Weighted Avg. 0.762 0.235 0.77 0.762 0.76 0.851
=== Confusion Matrix ===
a b <-- classified as
860 399 | a = true
190 1024 | b = false
Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment