Hello,
I have some output from the score node. However, i dont know how to interpret it. Please can someone help with an explanation?
Please see code below.
Thanks
Dee
*------------------------------------------------------------*
User: user
Date: 13 September 2017
Time: 20:31:56 o'clock
*------------------------------------------------------------*
* Training Output
*------------------------------------------------------------*
Variable Summary
Measurement Frequency
Role Level Count
SEGMENT INTERVAL 1
SEGMENT NOMINAL 1
TARGET BINARY 1
*------------------------------------------------------------*
* Score Output
*------------------------------------------------------------*
Score Input Variables
Used in
Variable Score
Variable Name Role Creator Comment Label Hidden Code
WHERE_VERIFIED INPUT Y Y
Score Output Variables
Variable Name Function Creator Label
EM_CLASSIFICATION CLASSIFICATION Score Prediction for PATIENT_ALIVE_FLAG
EM_EVENTPROBABILITY PREDICT Score Probability for level 1 of PATIENT_ALIVE_FLAG
EM_PROBABILITY PREDICT Score Probability of Classification
EM_SEGMENT TRANSFORM Score Segment
IMP_WHERE_VERIFIED TRANSFORM Impt Imputed WHERE_VERIFIED
I_PATIENT_ALIVE_FLAG CLASSIFICATION EndGrp2 Into: PATIENT_ALIVE_FLAG
P_PATIENT_ALIVE_FLAG0 PREDICT EndGrp2 Predicted: PATIENT_ALIVE_FLAG=0
P_PATIENT_ALIVE_FLAG1 PREDICT EndGrp2 Predicted: PATIENT_ALIVE_FLAG=1
U_PATIENT_ALIVE_FLAG CLASSIFICATION EndGrp2 Unnormalized Into: PATIENT_ALIVE_FLAG
_WARN_ ASSESS EndGrp2 Warnings
_XVAL_ TRANSFORM Impt
b_PATIENT_ALIVE_FLAG TRANSFORM MdlComp
fold TRANSFORM Trans2
*------------------------------------------------------------*
* Report Output
*------------------------------------------------------------*
Class Variable Summary Statistics
Data Role=TRAIN Output Type=CLASSIFICATION
Numeric Formatted Frequency
Variable Value Value Count Percent
I_PATIENT_ALIVE_FLAG . 1 876 100
Data Role=TRAIN Output Type=SEGMENT
Numeric Formatted Frequency
Variable Value Value Count Percent
b_PATIENT_ALIVE_FLAG 1 1 51 5.82192
b_PATIENT_ALIVE_FLAG 2 2 56 6.39269
b_PATIENT_ALIVE_FLAG 3 3 54 6.16438
b_PATIENT_ALIVE_FLAG 4 4 43 4.90868
b_PATIENT_ALIVE_FLAG 5 5 41 4.68037
b_PATIENT_ALIVE_FLAG 6 6 38 4.33790
b_PATIENT_ALIVE_FLAG 7 7 30 3.42466
b_PATIENT_ALIVE_FLAG 8 8 43 4.90868
b_PATIENT_ALIVE_FLAG 9 9 41 4.68037
b_PATIENT_ALIVE_FLAG 10 10 40 4.56621
b_PATIENT_ALIVE_FLAG 11 11 39 4.45205
b_PATIENT_ALIVE_FLAG 12 12 55 6.27854
b_PATIENT_ALIVE_FLAG 13 13 41 4.68037
b_PATIENT_ALIVE_FLAG 14 14 43 4.90868
b_PATIENT_ALIVE_FLAG 15 15 54 6.16438
b_PATIENT_ALIVE_FLAG 16 16 31 3.53881
b_PATIENT_ALIVE_FLAG 17 17 49 5.59361
b_PATIENT_ALIVE_FLAG 18 18 40 4.56621
b_PATIENT_ALIVE_FLAG 19 19 46 5.25114
b_PATIENT_ALIVE_FLAG 20 20 41 4.68037
Data Role=VALIDATE Output Type=CLASSIFICATION
Numeric Formatted Frequency
Variable Value Value Count Percent
I_PATIENT_ALIVE_FLAG . 1 586 100
Data Role=VALIDATE Output Type=SEGMENT
Numeric Formatted Frequency
Variable Value Value Count Percent
b_PATIENT_ALIVE_FLAG 1 1 35 5.97270
b_PATIENT_ALIVE_FLAG 2 2 40 6.82594
b_PATIENT_ALIVE_FLAG 3 3 34 5.80205
b_PATIENT_ALIVE_FLAG 4 4 25 4.26621
b_PATIENT_ALIVE_FLAG 5 5 25 4.26621
b_PATIENT_ALIVE_FLAG 6 6 31 5.29010
b_PATIENT_ALIVE_FLAG 7 7 18 3.07167
b_PATIENT_ALIVE_FLAG 8 8 23 3.92491
b_PATIENT_ALIVE_FLAG 9 9 25 4.26621
b_PATIENT_ALIVE_FLAG 10 10 26 4.43686
b_PATIENT_ALIVE_FLAG 11 11 29 4.94881
b_PATIENT_ALIVE_FLAG 12 12 26 4.43686
b_PATIENT_ALIVE_FLAG 13 13 25 4.26621
b_PATIENT_ALIVE_FLAG 14 14 24 4.09556
b_PATIENT_ALIVE_FLAG 15 15 34 5.80205
b_PATIENT_ALIVE_FLAG 16 16 28 4.77816
b_PATIENT_ALIVE_FLAG 17 17 30 5.11945
b_PATIENT_ALIVE_FLAG 18 18 24 4.09556
b_PATIENT_ALIVE_FLAG 19 19 43 7.33788
b_PATIENT_ALIVE_FLAG 20 20 41 6.99659
Interval Variable Summary Statistics
Variable Name=P_PATIENT_ALIVE_FLAG0
Statistics Label TRAIN VALIDATE
MEAN Mean 0.034 0.038
STD Standard Deviation 0.046 0.049
N Non Missing 876.000 586.000
MIN Minimum 0.000 0.000
P25 25th Percentile 0.012 0.013
MEDIAN Median 0.021 0.024
P75 75th Percentile 0.038 0.042
MAX Maximum 0.751 0.458
Variable Name=P_PATIENT_ALIVE_FLAG1
Statistics Label TRAIN VALIDATE
MEAN Mean 0.966 0.962
STD Standard Deviation 0.046 0.049
N Non Missing 876.000 586.000
MIN Minimum 0.249 0.542
P25 25th Percentile 0.962 0.958
MEDIAN Median 0.979 0.976
P75 75th Percentile 0.988 0.987
MAX Maximum 1.000 1.000
The output from the Score node will contain the following output variables:
Fixed Output Name Definition
EM_PREDICTION The prediction variable for an interval target.
EM_PROBABILITY Posterior probability associated with the predicted classification. That is, it corresponds the maximum of the posterior probabilities, max(P1, P2, ..., Pk).
EM_EVENTPROBABILITY Posterior probability associated with target event.
EM_CLASSIFICATION I_variable, the prediction variable for a class target, the most likely outcome
EM_SEGMENT name of the segment variable.
For your analysis, SAS Enterprise Miner generated
P_PATIENT_ALIVE_FLAG1: the probability that PATIENT_ALIVE_FLAG = 1
P_PATIENT_ALIVE_FLAG0: the probability that PATIENT_ALIVE_FLAG = 0
The assignment to an outcome is ultimately a modeling decision. If you just want to assign each observation to the most likely outcome, this is the value SAS Enterprise Miner has stored in I_PATIENT_ALIVE_FLAG. It is possible, however, that you might not want to predict PATIENT_ALIVE_FLAG = 0 unless there is a very high probability that this was the case (e.g. perhaps only when P_PATIENT_ALIVE_FLAG0 is greater than 0.9). In any case, the software can compute the probabilities but you need to decide what the threshold should be to determine if you predict PATIENT_ALIVE_FLAG=1 or PATIENT_ALIVE_FLAG=0. It is then easy to assign the observations to a category based on the cutoff you choose.
Hope this helps!
Doug
The output from the Score node will contain the following output variables:
Fixed Output Name Definition
EM_PREDICTION The prediction variable for an interval target.
EM_PROBABILITY Posterior probability associated with the predicted classification. That is, it corresponds the maximum of the posterior probabilities, max(P1, P2, ..., Pk).
EM_EVENTPROBABILITY Posterior probability associated with target event.
EM_CLASSIFICATION I_variable, the prediction variable for a class target, the most likely outcome
EM_SEGMENT name of the segment variable.
For your analysis, SAS Enterprise Miner generated
P_PATIENT_ALIVE_FLAG1: the probability that PATIENT_ALIVE_FLAG = 1
P_PATIENT_ALIVE_FLAG0: the probability that PATIENT_ALIVE_FLAG = 0
The assignment to an outcome is ultimately a modeling decision. If you just want to assign each observation to the most likely outcome, this is the value SAS Enterprise Miner has stored in I_PATIENT_ALIVE_FLAG. It is possible, however, that you might not want to predict PATIENT_ALIVE_FLAG = 0 unless there is a very high probability that this was the case (e.g. perhaps only when P_PATIENT_ALIVE_FLAG0 is greater than 0.9). In any case, the software can compute the probabilities but you need to decide what the threshold should be to determine if you predict PATIENT_ALIVE_FLAG=1 or PATIENT_ALIVE_FLAG=0. It is then easy to assign the observations to a category based on the cutoff you choose.
Hope this helps!
Doug
Okay. brilliant thank you very much.
Hi dee2017,
I'm glad you found some useful info! If Doug's reply was the exact solution to your problem, can you "Accept it as a solution"? Or if it was particularly helpful, feel free to "Like" it. This will help other community members who may run into the same issue know what worked.
Thanks!
Anna
Are you ready for the spotlight? We're accepting content ideas for SAS Innovate 2025 to be held May 6-9 in Orlando, FL. The call is open until September 25. Read more here about why you should contribute and what is in it for you!
Use this tutorial as a handy guide to weigh the pros and cons of these commonly used machine learning algorithms.
Find more tutorials on the SAS Users YouTube channel.