***********************************************************************
ALNfit Pro Automatic Regression and Classification Program by Dendronic
********************** ALNfit setting up the run **********************
The task is regression
A DTREE is to be created using the data file
Opening data file C:\Documents and Settings\arms\My Documents\Dendronic\Madeira Tutorial\Demos\Replace missing outputs\PTVdata.txt for analysis succeeded!
The line read from the file is: 
 Demonstration data for ALNfit Pro			 
The line is assumed to be a header line. Analysis continues.
There were 5 items read in the header line
The line read from the file is: 
 Pressure in millibars 			 
The line is assumed to be a header line. Analysis continues.
There were 3 items read in the header line
The line read from the file is: 
 Temperature in degrees Fahrenheit and Celsius			 
The line is assumed to be a header line. Analysis continues.
There were 6 items read in the header line
The line read from the file is: 
 Volume in Liters			 
The line is assumed to be a header line. Analysis continues.
There were 3 items read in the header line
The line read from the file is: 
 Pressure	TempF	TempC	Volume 
The line is assumed to be a header line. Analysis continues.
There were 4 items read in the header line
The line read from the file is: 
 _______	_______	_______	_______ 
The line is assumed to be a header line. Analysis continues.
There were 4 items read in the header line
First data line to help checking for correctness:
615.5	40.6	4.7	99999
Count of items in first data line = 4
The number of columns in the data file is 4
The number of header rows in the data file is 6
The number of data rows in the data file is 1000
Starting to preprocess the data file
Opening file C:\Documents and Settings\arms\My Documents\Dendronic\Madeira Tutorial\Demos\Replace missing outputs\PTVdata.txt succeeded.
PreprocessedDataFile created with 990 rows
Warning: if the ideal function to be learned is complicated,
Or if there is a lot of noise in the data, there must be MANY times
more rows in the data file as there are ALN inputs!
The number of rows in the test set is 99
The number of ALNs to be averaged in bagging is 10
The dimension of the problem (inputs + one desired output) is 3
The output variable is Volume
Creating the training/validation file
Creating the test file

************** Analysis of TV file begins ********
Dimension nDim is 3, nRowsTV is 891
Side of box in unit cube per point in the TVset = 0.033501
Stdev of Pressure = 346.091782 . Epsilon = 5.797255 
Stdev of TempC = 20.783571 . Epsilon = 0.348138 
Stdev of output variable Volume = 0.922937 . Output tolerance is used for Epsilon. 
Doing regression

****** Linear regression to get upper bound on error begins *****
Linear regression ALN creation succeeded!
Epoch 12 Estimated RMSE 0.372749, Active/Total LTUs 1/1
Training finished.  RMSE = 0.372749 
RMSE on validation set using 48 samples = 0.362190 
Linear regression continues, validation error = 0.362190l 
Epoch 12 Estimated RMSE 0.374005, Active/Total LTUs 1/1
Training finished.  RMSE = 0.374005 
RMSE on validation set using 48 samples = 0.346803 
Linear regression continues, validation error = 0.346803l 
Epoch 12 Estimated RMSE 0.371708, Active/Total LTUs 1/1
Training finished.  RMSE = 0.371708 
RMSE on validation set using 48 samples = 0.291742 
Linear regression continues, validation error = 0.291742l 
Epoch 12 Estimated RMSE 0.275840, Active/Total LTUs 1/1
Training finished.  RMSE = 0.275840 
RMSE on validation set using 48 samples = 0.278657 
Linear regression continues, validation error = 0.278657l 
Epoch 12 Estimated RMSE 0.388425, Active/Total LTUs 1/1
Training finished.  RMSE = 0.388425 
RMSE on validation set using 48 samples = 0.308659 
Linear regression weight on ALN input Pressure is -0.002642
Linear regression weight on ALN input TempC is 0.006743
Linear regression final validation error = 0.278657l 
Linear regression complete

*** Overtraining single ALN to find low validation error begins***
ALN creation succeeded!
Initial validation error = 0.278657l Tolerance = 0.183913l 
Noise estimation ALN training of single ALN, iteration 0
 Epoch 12 Estimated RMSE 0.088491, Active/Total LTUs 3/3
Training finished.  RMSE = 0.088491 
Training succeeded!
RMSE on validation set using 144 samples = 0.090966 
Training iteration finished, validation error = 0.090966l Tolerance = 0.060038l 
Noise estimation ALN training of single ALN, iteration 1
 Epoch 12 Estimated RMSE 0.026324, Active/Total LTUs 4/4
Training finished.  RMSE = 0.026324 
Training succeeded!
RMSE on validation set using 192 samples = 0.030812 
Training iteration finished, validation error = 0.030812l Tolerance = 0.020336l 
Noise estimation ALN training of single ALN, iteration 2
 Epoch 12 Estimated RMSE 0.018861, Active/Total LTUs 9/9
Training finished.  RMSE = 0.018861 
Training succeeded!
RMSE on validation set using 465 samples = 0.026866 
Training iteration finished, validation error = 0.026866l Tolerance = 0.017731l 
Noise estimation ALN training of single ALN, iteration 3
 Epoch 12 Estimated RMSE 0.016158, Active/Total LTUs 17/17
Training finished.  RMSE = 0.016158 
Training succeeded!
RMSE on validation set using 465 samples = 0.021553 
Training iteration finished, validation error = 0.021553l Tolerance = 0.014225l 
Noise estimation ALN training of single ALN, iteration 4
 Epoch 12 Estimated RMSE 0.014732, Active/Total LTUs 31/32
Training finished.  RMSE = 0.014732 
Training succeeded!
RMSE on validation set using 465 samples = 0.021494 
Training iteration finished, validation error = 0.021494l Tolerance = 0.014186l 
Noise estimation ALN training of single ALN, iteration 5
 Epoch 12 Estimated RMSE 0.012275, Active/Total LTUs 39/42
Training finished.  RMSE = 0.012275 
Training succeeded!
RMSE on validation set using 465 samples = 0.018864 
Training iteration finished, validation error = 0.018864l Tolerance = 0.012450l 
Noise estimation ALN training of single ALN, iteration 6
 Epoch 12 Estimated RMSE 0.011790, Active/Total LTUs 44/51
Training finished.  RMSE = 0.011790 
Training succeeded!
RMSE on validation set using 465 samples = 0.020176 
Noise estimation training of single ALN final validation error = 0.018864l 
We set the output tolerance to about 2/3 of the validation error to limit
the splitting of linear pieces that fit better than this error level. 
Noise estimation training of single ALN final tolerance = 0.012450l 

**************Approximation with several ALNs begins ********
Training 10 approximation ALNs starts, using noise to limit splitting
ALN creation succeeded!
----------  Training approximation ALN 0 ------------------
Begin approximation iteration 0 on ALN 0
Epoch 26 Estimated RMSE 0.049487, Active/Total LTUs 4/4
Training finished.  RMSE = 0.049487 
Training succeeded!
RMSE on validation set using 192 samples = 0.051453 
Approximation continues, validation error = 0.051453l Tolerance = 0.012450l 
Begin approximation iteration 1 on ALN 0
Epoch 26 Estimated RMSE 0.019316, Active/Total LTUs 15/15
Training finished.  RMSE = 0.019316 
Training succeeded!
RMSE on validation set using 444 samples = 0.019526 
Approximation continues, validation error = 0.019526l Tolerance = 0.012450l 
Begin approximation iteration 2 on ALN 0
Epoch 26 Estimated RMSE 0.014602, Active/Total LTUs 42/42
Training finished.  RMSE = 0.014602 
Training succeeded!
RMSE on validation set using 444 samples = 0.016874 
Approximation continues, validation error = 0.016874l Tolerance = 0.012450l 
Begin approximation iteration 3 on ALN 0
Epoch 26 Estimated RMSE 0.011876, Active/Total LTUs 67/74
Training finished.  RMSE = 0.011876 
Training succeeded!
RMSE on validation set using 444 samples = 0.019311 
Approximation continues, validation error = 0.019311l Tolerance = 0.012450l 
Begin approximation iteration 4 on ALN 0
Epoch 26 Estimated RMSE 0.010647, Active/Total LTUs 75/84
Training finished.  RMSE = 0.010647 
Training succeeded!
RMSE on validation set using 444 samples = 0.020885 
ALN creation succeeded!
----------  Training approximation ALN 1 ------------------
Begin approximation iteration 0 on ALN 1
Epoch 26 Estimated RMSE 0.058990, Active/Total LTUs 4/4
Training finished.  RMSE = 0.058990 
Training succeeded!
RMSE on validation set using 192 samples = 0.076013 
Approximation continues, validation error = 0.076013l Tolerance = 0.013784l 
Begin approximation iteration 1 on ALN 1
Epoch 26 Estimated RMSE 0.015217, Active/Total LTUs 14/14
Training finished.  RMSE = 0.015217 
Training succeeded!
RMSE on validation set using 445 samples = 0.018902 
Approximation continues, validation error = 0.018902l Tolerance = 0.013784l 
Begin approximation iteration 2 on ALN 1
Epoch 26 Estimated RMSE 0.013259, Active/Total LTUs 38/39
Training finished.  RMSE = 0.013259 
Training succeeded!
RMSE on validation set using 445 samples = 0.018109 
Approximation continues, validation error = 0.018109l Tolerance = 0.013784l 
Begin approximation iteration 3 on ALN 1
Epoch 26 Estimated RMSE 0.010775, Active/Total LTUs 61/65
Training finished.  RMSE = 0.010775 
Training succeeded!
RMSE on validation set using 445 samples = 0.022010 
Approximation continues, validation error = 0.022010l Tolerance = 0.013784l 
Begin approximation iteration 4 on ALN 1
Epoch 26 Estimated RMSE 0.009738, Active/Total LTUs 58/67
Training finished.  RMSE = 0.009738 
Training succeeded!
RMSE on validation set using 445 samples = 0.023418 
ALN creation succeeded!
----------  Training approximation ALN 2 ------------------
Begin approximation iteration 0 on ALN 2
Epoch 26 Estimated RMSE 0.067888, Active/Total LTUs 4/4
Training finished.  RMSE = 0.067888 
Training succeeded!
RMSE on validation set using 192 samples = 0.052668 
Approximation continues, validation error = 0.052668l Tolerance = 0.015456l 
Begin approximation iteration 1 on ALN 2
Epoch 26 Estimated RMSE 0.016499, Active/Total LTUs 14/14
Training finished.  RMSE = 0.016499 
Training succeeded!
RMSE on validation set using 432 samples = 0.016424 
Approximation continues, validation error = 0.016424l Tolerance = 0.015456l 
Begin approximation iteration 2 on ALN 2
Epoch 26 Estimated RMSE 0.014598, Active/Total LTUs 34/36
Training finished.  RMSE = 0.014598 
Training succeeded!
RMSE on validation set using 432 samples = 0.016540 
Approximation continues, validation error = 0.016540l Tolerance = 0.015456l 
Begin approximation iteration 3 on ALN 2
Epoch 26 Estimated RMSE 0.012108, Active/Total LTUs 56/63
Training finished.  RMSE = 0.012108 
Training succeeded!
RMSE on validation set using 432 samples = 0.019135 
Approximation continues, validation error = 0.019135l Tolerance = 0.015456l 
Begin approximation iteration 4 on ALN 2
Epoch 26 Estimated RMSE 0.011039, Active/Total LTUs 65/72
Training finished.  RMSE = 0.011039 
Training succeeded!
RMSE on validation set using 432 samples = 0.020850 
ALN creation succeeded!
----------  Training approximation ALN 3 ------------------
Begin approximation iteration 0 on ALN 3
Epoch 26 Estimated RMSE 0.053527, Active/Total LTUs 4/4
Training finished.  RMSE = 0.053527 
Training succeeded!
RMSE on validation set using 192 samples = 0.057796 
Approximation continues, validation error = 0.057796l Tolerance = 0.013761l 
Begin approximation iteration 1 on ALN 3
Epoch 26 Estimated RMSE 0.015995, Active/Total LTUs 15/15
Training finished.  RMSE = 0.015995 
Training succeeded!
RMSE on validation set using 451 samples = 0.018185 
Approximation continues, validation error = 0.018185l Tolerance = 0.013761l 
Begin approximation iteration 2 on ALN 3
Epoch 26 Estimated RMSE 0.014337, Active/Total LTUs 44/47
Training finished.  RMSE = 0.014337 
Training succeeded!
RMSE on validation set using 451 samples = 0.017203 
Approximation continues, validation error = 0.017203l Tolerance = 0.013761l 
Begin approximation iteration 3 on ALN 3
Epoch 26 Estimated RMSE 0.011498, Active/Total LTUs 61/69
Training finished.  RMSE = 0.011498 
Training succeeded!
RMSE on validation set using 451 samples = 0.020999 
Approximation continues, validation error = 0.020999l Tolerance = 0.013761l 
Begin approximation iteration 4 on ALN 3
Epoch 26 Estimated RMSE 0.010203, Active/Total LTUs 60/71
Training finished.  RMSE = 0.010203 
Training succeeded!
RMSE on validation set using 451 samples = 0.021818 
ALN creation succeeded!
----------  Training approximation ALN 4 ------------------
Begin approximation iteration 0 on ALN 4
Epoch 26 Estimated RMSE 0.077319, Active/Total LTUs 4/4
Training finished.  RMSE = 0.077319 
Training succeeded!
RMSE on validation set using 192 samples = 0.058423 
Approximation continues, validation error = 0.058423l Tolerance = 0.014400l 
Begin approximation iteration 1 on ALN 4
Epoch 26 Estimated RMSE 0.016931, Active/Total LTUs 15/15
Training finished.  RMSE = 0.016931 
Training succeeded!
RMSE on validation set using 433 samples = 0.018003 
Approximation continues, validation error = 0.018003l Tolerance = 0.014400l 
Begin approximation iteration 2 on ALN 4
Epoch 26 Estimated RMSE 0.014986, Active/Total LTUs 36/39
Training finished.  RMSE = 0.014986 
Training succeeded!
RMSE on validation set using 433 samples = 0.016761 
Approximation continues, validation error = 0.016761l Tolerance = 0.014400l 
Begin approximation iteration 3 on ALN 4
Epoch 26 Estimated RMSE 0.012783, Active/Total LTUs 57/62
Training finished.  RMSE = 0.012783 
Training succeeded!
RMSE on validation set using 433 samples = 0.019246 
Approximation continues, validation error = 0.019246l Tolerance = 0.014400l 
Begin approximation iteration 4 on ALN 4
Epoch 26 Estimated RMSE 0.011506, Active/Total LTUs 60/68
Training finished.  RMSE = 0.011506 
Training succeeded!
RMSE on validation set using 433 samples = 0.019410 
ALN creation succeeded!
----------  Training approximation ALN 5 ------------------
Begin approximation iteration 0 on ALN 5
Epoch 26 Estimated RMSE 0.056206, Active/Total LTUs 4/4
Training finished.  RMSE = 0.056206 
Training succeeded!
RMSE on validation set using 192 samples = 0.055594 
Approximation continues, validation error = 0.055594l Tolerance = 0.012811l 
Begin approximation iteration 1 on ALN 5
Epoch 26 Estimated RMSE 0.017614, Active/Total LTUs 15/15
Training finished.  RMSE = 0.017614 
Training succeeded!
RMSE on validation set using 430 samples = 0.018245 
Approximation continues, validation error = 0.018245l Tolerance = 0.012811l 
Begin approximation iteration 2 on ALN 5
Epoch 26 Estimated RMSE 0.014021, Active/Total LTUs 44/46
Training finished.  RMSE = 0.014021 
Training succeeded!
RMSE on validation set using 430 samples = 0.018605 
Approximation continues, validation error = 0.018605l Tolerance = 0.012811l 
Begin approximation iteration 3 on ALN 5
Epoch 26 Estimated RMSE 0.012225, Active/Total LTUs 72/80
Training finished.  RMSE = 0.012225 
Training succeeded!
RMSE on validation set using 430 samples = 0.019427 
Approximation continues, validation error = 0.019427l Tolerance = 0.012811l 
Begin approximation iteration 4 on ALN 5
Epoch 26 Estimated RMSE 0.010670, Active/Total LTUs 84/90
Training finished.  RMSE = 0.010670 
Training succeeded!
RMSE on validation set using 430 samples = 0.019674 
ALN creation succeeded!
----------  Training approximation ALN 6 ------------------
Begin approximation iteration 0 on ALN 6
Epoch 26 Estimated RMSE 0.065153, Active/Total LTUs 4/4
Training finished.  RMSE = 0.065153 
Training succeeded!
RMSE on validation set using 192 samples = 0.056386 
Approximation continues, validation error = 0.056386l Tolerance = 0.012985l 
Begin approximation iteration 1 on ALN 6
Epoch 26 Estimated RMSE 0.016155, Active/Total LTUs 15/15
Training finished.  RMSE = 0.016155 
Training succeeded!
RMSE on validation set using 413 samples = 0.017447 
Approximation continues, validation error = 0.017447l Tolerance = 0.012985l 
Begin approximation iteration 2 on ALN 6
Epoch 26 Estimated RMSE 0.013801, Active/Total LTUs 42/44
Training finished.  RMSE = 0.013801 
Training succeeded!
RMSE on validation set using 413 samples = 0.017432 
Approximation continues, validation error = 0.017432l Tolerance = 0.012985l 
Begin approximation iteration 3 on ALN 6
Epoch 26 Estimated RMSE 0.011365, Active/Total LTUs 70/75
Training finished.  RMSE = 0.011365 
Training succeeded!
RMSE on validation set using 413 samples = 0.021161 
Approximation continues, validation error = 0.021161l Tolerance = 0.012985l 
Begin approximation iteration 4 on ALN 6
Epoch 26 Estimated RMSE 0.009468, Active/Total LTUs 77/88
Training finished.  RMSE = 0.009468 
Training succeeded!
RMSE on validation set using 413 samples = 0.022522 
ALN creation succeeded!
----------  Training approximation ALN 7 ------------------
Begin approximation iteration 0 on ALN 7
Epoch 26 Estimated RMSE 0.067652, Active/Total LTUs 4/4
Training finished.  RMSE = 0.067652 
Training succeeded!
RMSE on validation set using 192 samples = 0.049867 
Approximation continues, validation error = 0.049867l Tolerance = 0.014865l 
Begin approximation iteration 1 on ALN 7
Epoch 26 Estimated RMSE 0.016220, Active/Total LTUs 14/14
Training finished.  RMSE = 0.016220 
Training succeeded!
RMSE on validation set using 436 samples = 0.020135 
Approximation continues, validation error = 0.020135l Tolerance = 0.014865l 
Begin approximation iteration 2 on ALN 7
Epoch 26 Estimated RMSE 0.015398, Active/Total LTUs 33/33
Training finished.  RMSE = 0.015398 
Training succeeded!
RMSE on validation set using 436 samples = 0.020498 
Approximation continues, validation error = 0.020498l Tolerance = 0.014865l 
Begin approximation iteration 3 on ALN 7
Epoch 26 Estimated RMSE 0.013235, Active/Total LTUs 46/52
Training finished.  RMSE = 0.013235 
Training succeeded!
RMSE on validation set using 436 samples = 0.022091 
Approximation continues, validation error = 0.022091l Tolerance = 0.014865l 
Begin approximation iteration 4 on ALN 7
Epoch 26 Estimated RMSE 0.012889, Active/Total LTUs 52/58
Training finished.  RMSE = 0.012889 
Training succeeded!
RMSE on validation set using 436 samples = 0.021756 
Approximation continues, validation error = 0.021756l Tolerance = 0.014865l 
Begin approximation iteration 5 on ALN 7
Epoch 26 Estimated RMSE 0.013047, Active/Total LTUs 51/59
Training finished.  RMSE = 0.013047 
Training succeeded!
RMSE on validation set using 436 samples = 0.021617 
Approximation continues, validation error = 0.021617l Tolerance = 0.014865l 
Begin approximation iteration 6 on ALN 7
Epoch 26 Estimated RMSE 0.012619, Active/Total LTUs 51/59
Training finished.  RMSE = 0.012619 
Training succeeded!
RMSE on validation set using 436 samples = 0.022408 
ALN creation succeeded!
----------  Training approximation ALN 8 ------------------
Begin approximation iteration 0 on ALN 8
Epoch 26 Estimated RMSE 0.042127, Active/Total LTUs 4/4
Training finished.  RMSE = 0.042127 
Training succeeded!
RMSE on validation set using 192 samples = 0.053861 
Approximation continues, validation error = 0.053861l Tolerance = 0.014789l 
Begin approximation iteration 1 on ALN 8
Epoch 26 Estimated RMSE 0.016854, Active/Total LTUs 14/14
Training finished.  RMSE = 0.016854 
Training succeeded!
RMSE on validation set using 458 samples = 0.018160 
Approximation continues, validation error = 0.018160l Tolerance = 0.014789l 
Begin approximation iteration 2 on ALN 8
Epoch 26 Estimated RMSE 0.014253, Active/Total LTUs 31/36
Training finished.  RMSE = 0.014253 
Training succeeded!
RMSE on validation set using 458 samples = 0.020191 
Approximation continues, validation error = 0.020191l Tolerance = 0.014789l 
Begin approximation iteration 3 on ALN 8
Epoch 26 Estimated RMSE 0.012751, Active/Total LTUs 57/62
Training finished.  RMSE = 0.012751 
Training succeeded!
RMSE on validation set using 458 samples = 0.024338 
Approximation continues, validation error = 0.024338l Tolerance = 0.014789l 
Begin approximation iteration 4 on ALN 8
Epoch 26 Estimated RMSE 0.011238, Active/Total LTUs 59/68
Training finished.  RMSE = 0.011238 
Training succeeded!
RMSE on validation set using 458 samples = 0.025480 
ALN creation succeeded!
----------  Training approximation ALN 9 ------------------
Begin approximation iteration 0 on ALN 9
Epoch 26 Estimated RMSE 0.041705, Active/Total LTUs 4/4
Training finished.  RMSE = 0.041705 
Training succeeded!
RMSE on validation set using 192 samples = 0.047385 
Approximation continues, validation error = 0.047385l Tolerance = 0.016817l 
Begin approximation iteration 1 on ALN 9
Epoch 26 Estimated RMSE 0.017407, Active/Total LTUs 12/12
Training finished.  RMSE = 0.017407 
Training succeeded!
RMSE on validation set using 461 samples = 0.020709 
Approximation continues, validation error = 0.020709l Tolerance = 0.016817l 
Begin approximation iteration 2 on ALN 9
Epoch 26 Estimated RMSE 0.014699, Active/Total LTUs 25/26
Training finished.  RMSE = 0.014699 
Training succeeded!
RMSE on validation set using 461 samples = 0.021601 
Approximation continues, validation error = 0.021601l Tolerance = 0.016817l 
Begin approximation iteration 3 on ALN 9
Epoch 26 Estimated RMSE 0.013325, Active/Total LTUs 34/36
Training finished.  RMSE = 0.013325 
Training succeeded!
RMSE on validation set using 461 samples = 0.024936 
Approximation continues, validation error = 0.024936l Tolerance = 0.016817l 
Begin approximation iteration 4 on ALN 9
Epoch 26 Estimated RMSE 0.012518, Active/Total LTUs 38/40
Training finished.  RMSE = 0.012518 
Training succeeded!
RMSE on validation set using 461 samples = 0.026054 

**** Analyzing results on the training/validation set begins ***
Size of datasets PP TV Test 990  891  99 
Root mean square error of the average over 10 ALNS is 0.012986 
Warning: the above result is optimistic, see results on the test set below
Importance of each input variable:
Abs imp = stdev(input var) * average absolute weight / stdev(output var) 
Abs imp is numerical and indicates ups and downs in output when the given input varies.
For example a sawtooth function with six teeth would have importance 12.
First we have to compute the standard deviation of the output variable.

Standard deviation of the output in the TVfile 0.922937
Variable Pressure: stdev = 	346.091782; avg.wt = 	-0.002952; abs imp = 	1.121914
Variable TempC: stdev = 	20.783571; avg.wt = 	0.008092; abs imp = 	0.187246

********** Training an ALN by resampling the average ******
Average ALN creation succeeded!
Tolerance for training the average ALN is 0.006179
Epoch 12 Estimated RMSE 0.288671, Active/Total LTUs 4/4
Training finished.  RMSE = 0.288671 
Average ALN training succeeded!
Epoch 12 Estimated RMSE 0.025449, Active/Total LTUs 14/15
Training finished.  RMSE = 0.025449 
Average ALN training succeeded!
Epoch 12 Estimated RMSE 0.005862, Active/Total LTUs 29/31
Training finished.  RMSE = 0.005862 
Average ALN training succeeded!
Epoch 12 Estimated RMSE 0.003782, Active/Total LTUs 46/50
Training finished.  RMSE = 0.003782 
Average ALN training succeeded!
Epoch 12 Estimated RMSE 0.002874, Active/Total LTUs 51/56
Training finished.  RMSE = 0.002874 
Average ALN training succeeded!
Epoch 12 Estimated RMSE 0.002569, Active/Total LTUs 56/57
Training finished.  RMSE = 0.002569 
Average ALN training succeeded!
Epoch 12 Estimated RMSE 0.002262, Active/Total LTUs 49/57
Training finished.  RMSE = 0.002262 
Average ALN training succeeded!
Epoch 12 Estimated RMSE 0.002398, Active/Total LTUs 50/57
Training finished.  RMSE = 0.002398 
Average ALN training succeeded!
Epoch 14 Estimated RMSE 0.002393, Active/Total LTUs 53/57
Training finished.  RMSE = 0.002393 
Polishing average ALN training succeeded!

***** Constructing an ALN decision tree from the average ALN *****
The DTREE of the average of ALNs 0816PTVdataDTREE.dtr  was written.

*********** Opening the DTREE file ****************
Opening DTREE file 0816PTVdataDTREE.dtr
DTREE succesfully parsed!
Dimension of the DTREE is 3 
Variable Pressure dblMin = 368.590822 dblMax = 1659.509178 
Variable TempC dblMin = -37.178357 dblMax = 40.778357 

******** Testing the DTREE on the test set **********
Creating internal data file OutputData succeeded!
It has 99 rows and 4 columns.
The following is based on a test set with 99 samples
Please be careful interpreting the results for small numbers of samples!
RMS deviation of DTREE output from desired (or from 0 if output column not present) is 0.013193 
Mean absolute deviation of DTREE output from desired is 0.010806
Maximum absolute deviation of DTREE output from desired is 0.039141

Please examine the output file named:
   0816PTVdataTrainScatterPlot.txt
The rightmost column of the output file is the ALN prediction.

Please examine the R or E file named:
   0816PTVdataR.txt
Where possible, missing data file output values have been computed.
The rule at the top shows the input variables and output column used.
Using this file as an input data file, you can do further replacements.