Changes between Initial Version and Version 1 of Ticket #1941, comment 21
- Timestamp:
- 05/06/13 18:11:45 (12 years ago)
Legend:
- Unmodified
- Added
- Removed
- Modified
-
Ticket #1941, comment 21
initial v1 7 7 8 8 I just checked the 'Parkinsons' data set and strongly believe that instances should not be shuffled. The data set contains multiple measurements for each patient. So, if the data set is shuffled it should be guaranteed that all measurements from a patient are either in the training partition or in the test partition only. 9 I also verified this in HL using the original and the shuffled data set (leads to much better estimated values).
