Professional Documents
Culture Documents
You can have at most 50 input variables, out of which atmost 40 could be categorical.
Make sure that the number of Input (Cat & Cont) columns exactly match with the number entered in UserInput sheet.
(D) Please make sure that your data does not have blank rows or blank columns.
(E) Continuous Inputs:
Any non-number in Cont column will be treated as missing value.
Application will replace it by the column mean
(E) Categorical Inputs:
Any blank cell or cells containing Excel error in Cat column will be treated as missing value
Application will reaplce it by the most frequently occuring category.
Category labels are case insensitive - lables good, Good, GoOd, GOOD will all be treated as the same categor
There should be at least 2 observations in each category of a Cat column.
If one of the category of a Cat column has only 1 observation, you should do one of the following Remove that observation OR
Rename the category to any other categories of that Cat column.
Step 2: Fill up Model Inputs
(A) Fill up the model inputs in the User Input Page.
(B) Make sure that your inputs are within the range of values allowed by the application.
(C) Click the 'Build Model' button to start modeling.
Step 3: Results of Modeling
(A) A Neural Network model is basically a set of weights between the layers of the net.
At the end of the run, the final set of weights are saved in the Calc sheet.
(B) The output page of this file will show you the values of MSE and ARE on the training and validation set
as the training of the model progresses. Two charts showing training and Validation MSE's
have been already provided in the Output sheet.
(C) In UserInput page if you have asked to save the model in a separate file, then a new file
will be created containing the model inputs, your data, and the fitted model ( i.e. the weights)
You will be able to use this file as a calculator to do prediction, given any new input.
Step 4: Study Profiles
Fitted model is a surface in p -dimension where the number of your inputs is p .
Unless p is 2 or less, it is not possible to show the surface graphically.
Profile plot is the next best way to visualize this fitted surface.
By varying only one predictor between two values and keeping all the others fixed at some pre-specified values
we get the profile plot - which is really a one dimensional cross section of the high dimensional surface.
In the Profile sheet you can specify which predictor to vary and the values at which the other predictors should be held fixe
Click Create Profile button to generate the profile.
If the predictor you choose to vary is categorical then the other info ( #points to be generated, start and end values)
will be ignored and the graph will show you the predicted response for each category of the predictor you have chosen to v
Profile plot lets you study the following things:
(1) Nature of relationship bettween a particular predictor X and the response Y
( E.g. Y increases as X increases OR Y decreases as X increases
OR the relationship is non-linear - Y first increases and then decreases with X etc etc.
(2) Profile plots also lets you study the interaction between predictors.
Suppose there are two predictors X and Z and we are studying the profile of Y as X varies
Suppose we look at the profile by keeping Z fixed at 1 and varying X between -10 and 10.
Now keep Z fixed at 2 instead of 1 and vary X between -10 and 10.
If the shape of the profiles in these two scenarios are drastically different
(e.g. one is increasing and the other is decreasing) then that says thay X and Z has interaction.
In other words, the effect of X on the Response is not same at all levels of Z
To study the effect of X, it matters where Z is set.
# Missing Value
Min
ll as continuous variables.
Max
Average
et of Input variables,
one with Y2 as output.
sd
Intercept
Slope
d as missing value
will all be treated as the same category
dimensional surface.
he other predictors should be held fixed.
le of Y as X varies
ween -10 and 10.
work weights.
Cont. Var.
# Missing Value
Cat. Var.
#Levels
Lables
Values
Dummy
Network Architectu
Training Options
Total #rows in your data
1
2
Training / Validation S
Network ArchitectureOptions
Number of Inputs ( bewtween 2 and 50)
0.4
Training Options
Total #rows in your data ( Minimum 10 )
38
NO
If you want to partition, how do you want to select the Validation set ?
Please choose one option
1
Please fill up the input necessary for the selected option
Save model in a separate workbook?
NO
etween 1 and 10 )
aximum 20 )
( Maximum 500 )
or Sequential )
1
Hidden 1
Hidden 2
2
0.5
50
Sequential
10%
5
Var Type
Var Name
Cont
Omit
Cylinder
Car
8
8
8
8
4
4
4
4
5
6
4
6
6
6
6
6
8
8
8
8
4
6
4
4
4
4
4
4
6
6
4
Output
MPG
Cont
Weight
16.9
15.5
18.5
30
27.5
27.2
30.9
20.3
20.3
17
21.6
16.2
20.6
20.8
18.6
18.1
17
17.6
16.5
18.2
26.5
21.9
34.1
35.1
27.4
31.5
29.5
28.4
28.8
26.8
33.5
4.36
4.054
3.605
3.94
2.155
2.56
2.3
2.23
2.83
3.14
2.795
3.41
3.38
3.07
3.62
3.41
3.84
3.725
3.955
3.83
2.585
2.91
1.975
1.915
2.67
1.99
2.135
2.67
2.595
2.7
2.556
4
4
4
4
6
4
4
Plymouth Horizon
Datsun 210
Fiat Strada
VW Dasher
Datsun 810
BMW 320i
VW Rabbit
34.2
31.8
37.3
30.5
22
21.5
31.9
2.2
2.02
2.13
2.19
2.815
2.6
1.925
m all as Continuous.
Cont
Cont
Drive_Ratio
Cont
Horsepower
2.73
2.26
2.56
2.45
3.7
3.05
3.54
3.37
3.9
3.5
3.77
3.58
2.73
3.08
2.71
2.73
2.41
2.26
2.26
2.45
3.08
3.08
3.73
2.97
3.08
3.78
3.05
2.53
2.69
2.84
2.69
Cat
Displacement
155
142
125
150
68
95
97
75
103
125
115
133
105
85
110
120
130
129
138
135
88
109
65
80
80
71
68
90
115
115
90
Country
350
351
267
360
98
134
119
105
131
163
121
163
231
200
225
258
305
302
351
318
140
171
86
98
121
89
98
151
173
173
151
US
US
US
US
US
Japan
Japan
US
Europe
Europe
Europe
Europe
US
US
US
US
US
US
US
US
US
US
Japan
Japan
US
Europe
Japan
US
US
US
US
Omit
3.37
3.7
3.1
3.7
3.7
3.64
3.78
70
65
69
78
97
110
71
105
85
91
97
146
121
89
US
Japan
Europe
Europe
Japan
Europe
Europe
Omit
Omit
Omit
Omit
Omit
Omit
Omit
Omit
Omit
Omit
X18
X19
Omit
Omit
Omit
Omit
Omit
X20
X21
X22
X23
X24
Omit
Omit
Omit
Omit
Omit
X25
X26
X27
X28
X29
Omit
Omit
Omit
Omit
Omit
X30
X31
X32
X33
X34
Omit
Omit
Omit
Omit
Omit
X35
X36
X37
X38
X39
Omit
Omit
Omit
Omit
Omit
X40
X41
X42
X43
X44
Omit
Omit
Omit
Omit
Omit
X45
X46
X47
X48
X49
Omit
Omit
Omit
Omit
Omit
X50
X51
X52
X53
X54
Omit
Omit
Omit
Omit
Omit
X55
X56
X57
X58
X59
Omit
X60
Created On :
MSE(Training)
MSE(Validation)
13.717
14-Aug-02
6.6302
1
8
RMSE
#VALUE!
Cont
Weight
Cont
Drive_Ratio
20.8014
#DIV/0!
Bias
Raw Input
Cont
Cylinder
1
Bias
Transformed Input
Hdn1_bias
Hdn1_Nrn1
Hdn1_Nrn2
Op_bias
Op_Nrn1
7.9600
Cylinder
Weight
3.0934
Drive_Ratio
0.9900
0.3877
0.0000
0.0000
0.0000
0.0000
0.2943
-1.1508
-0.9277
-0.4073
0.9926
1.0000
-1.1561
0.0774
-1.5765
0.2088
-0.5139
0.0000
0.0000
0.0000
0.0000
-1.8890
1.0000
2.2201
0.2432
2.7873
-1.1353
Category Table
Country
3
1
2
3
2.8629
us
japan
europe
0.5082
ARE
#DIV/0!
Cont
Cont
Cat
Horsepower
Displacement Country
101.7368
177.2895 us
Displacemen
Horsepower
t
Country.us
Country.japa
n
Country.europe
0.4082
0.3356
1.0000
0.0000
0.0000
0.0000
0.0000
0.0000
0.0000
0.0000
0.0000
-1.0479
-0.8085
-0.3676
0.7118
0.1508
-2.4783
-0.9123
-0.8300
0.3429
0.0155
-0.0342
-1.3324
50
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
Epoch
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
ARE (%)
ARE (%)
16.452
16.349
16.245
16.144
16.048
15.957
15.870
15.787
15.707
15.631
15.558
15.487
15.419
15.353
15.289
15.227
15.167
15.108
15.051
14.996
14.942
14.889
14.837
14.786
14.737
14.688
14.641
14.594
14.548
14.503
14.458
14.414
14.371
14.329
14.287
14.246
14.205
14.165
14.125
14.086
14.047
14.009
13.971
13.934
14.33%
14.25%
14.15%
14.05%
13.96%
13.87%
13.78%
13.70%
13.62%
13.54%
13.47%
13.40%
13.34%
13.27%
13.21%
13.15%
13.09%
13.04%
12.98%
12.93%
12.88%
12.83%
12.78%
12.74%
12.69%
12.65%
12.60%
12.56%
12.52%
12.48%
12.44%
12.40%
12.36%
12.33%
12.29%
12.26%
12.22%
12.19%
12.15%
12.12%
12.09%
12.06%
12.03%
11.99%
16.826
16.653
16.326
15.959
15.588
15.221
14.862
14.512
14.171
13.839
13.518
13.205
12.903
12.610
12.326
12.052
11.787
11.531
11.283
11.044
10.813
10.590
10.375
10.167
9.967
9.773
9.586
9.405
9.231
9.063
8.900
8.743
8.591
8.444
8.302
8.164
8.031
7.903
7.778
7.657
7.540
7.427
7.317
7.210
19.74%
19.63%
19.41%
19.17%
18.92%
18.68%
18.44%
18.21%
17.98%
17.75%
17.53%
17.32%
17.11%
16.90%
16.70%
16.50%
16.31%
16.12%
15.94%
15.76%
15.59%
15.42%
15.25%
15.09%
14.93%
14.78%
14.63%
14.48%
14.34%
14.20%
14.06%
13.92%
13.79%
13.67%
13.54%
13.42%
13.30%
13.19%
13.07%
12.96%
12.85%
12.74%
12.64%
12.54%
16.500
16.000
15.500
15.000
14.500
14.000
13.500
0
18.000
16.000
14.000
12.000
10.000
8.000
6.000
4.000
2.000
0.000
0
45
46
47
48
49
50
13.897
13.860
13.824
13.788
13.753
13.717
11.96%
11.93%
11.91%
11.88%
11.85%
11.82%
7.106
7.005
6.908
6.812
6.720
6.630
12.44%
12.34%
12.24%
12.15%
12.06%
11.96%
MSE (Training)
10
20
30
40
50
60
Epoch
MSE (Validation)
10
20
30
Epoch
40
50
60
Predictors
Cylinder
Weight
Drive_Ratio
Horsepower
Displacement
Country
Cylinder
4
4.04
4.08
4.12
4.16
4.2
4.24
4.28
4.32
4.36
4.4
4.44
4.48
4.52
4.56
4.6
4.64
4.68
4.72
4.76
4.8
4.84
4.88
4.92
4.96
5
5.04
5.08
5.12
5.16
5.2
5.24
5.28
5.32
5.36
5.4
5.44
5.48
5.52
5.56
5.6
5.64
5.68
5.72
5.76
Predicted MPG
25.50382
25.43797
25.37239
25.3071
25.24209
25.17737
25.11295
25.04884
24.98503
24.92155
24.85838
24.79555
24.73305
24.67088
24.60906
24.54759
24.48647
24.42571
24.36531
24.30528
24.24562
24.18633
24.12741
24.06888
24.01074
23.95298
23.89561
23.83864
23.78206
23.72588
23.6701
23.61472
23.55975
23.50519
23.45104
23.3973
23.34397
23.29105
23.23855
23.18647
23.1348
23.08355
23.03272
22.98231
22.93231
Predictor
Cylinder
Fixed Value
5.395
30
25
20
15
10
5
0
0
Category Table
Country
3
us
japan
europe
5.8
5.84
5.88
5.92
5.96
6
6.04
6.08
6.12
6.16
6.2
6.24
6.28
6.32
6.36
6.4
6.44
6.48
6.52
6.56
6.6
6.64
6.68
6.72
6.76
6.8
6.84
6.88
6.92
6.96
7
7.04
7.08
7.12
7.16
7.2
7.24
7.28
7.32
7.36
7.4
7.44
7.48
7.52
7.56
7.6
7.64
7.68
7.72
7.76
7.8
7.84
22.88274
22.83359
22.78486
22.73655
22.68866
22.64118
22.59413
22.5475
22.50128
22.45549
22.41011
22.36514
22.32059
22.27646
22.23274
22.18943
22.14653
22.10404
22.06195
22.02028
21.979
21.93813
21.89766
21.85759
21.81792
21.77864
21.73976
21.70127
21.66317
21.62545
21.58812
21.55117
21.51461
21.47842
21.44261
21.40717
21.3721
21.3374
21.30306
21.26909
21.23548
21.20223
21.16934
21.1368
21.1046
21.07276
21.04126
21.0101
20.97929
20.94881
20.91866
20.88885
7.88 20.85936
7.92 20.8302
7.96 20.80136
MPG
data points
Cylinder between
4
and
keeping the other predictors fixed at the specified values
100
Weight
Drive_Ratio
2.863
Horsepower
3.093
Displacement
101.737
177.289
65.00
155.00
Country
us
85.00
360.00
Predicted MPG