Training Log Archive: Never2Old4SchoolIn the 7 days ending Dec 11, 2016:
| |||||||||||||||||||||||||||||||||||||||||||||||||||||
| « | » |
| » now | ||||||||||
|
|
|
|
|
| ||||||||
| Mo | Tu | We | Th | Fr | Sa | Su | |||||||
Sunday Dec 11, 2016 #
Research (Programming) 3:00:00 [4]
shoes: CISS
A bunch of investigation around the variance bounds and why they still aren't quite right. Bottom line: the exponential assumption is bad.
Saturday Dec 10, 2016 #
Note
(rest day)
Off day for Pere Marquette!
Friday Dec 9, 2016 #
Research (Programming) 2:00:00 [3]
shoes: CISS
Some debugging of the block selection algorithm and validating the final runs. Variance seems a little tight, might have a normality violation that matters.
Thursday Dec 8, 2016 #
Research (Programming) 3:00:00 [3]
shoes: CISS
Good results from full plan dataset. One bug which will require some refactoring, but doesn't really change the result.
Wednesday Dec 7, 2016 #
Research (Programming) 3:00:00 [3]
shoes: CISS
Working through the samplers with the full data set. Validation is going to be a bear (as expected).
Classroom learning 1:00:00 [3]
shoes: General CS
Colloquium by Dr. He. Interesting, but not really relevant to my stuff.
Tuesday Dec 6, 2016 #
Research (Programming) 3:00:00 [3]
shoes: CISS
OK, data is flowing. Oracle's jdbc driver is brain dead, but it works fine if you spell everything out for it.
Monday Dec 5, 2016 #
Research (Programming) 3:00:00 [3]
shoes: CISS
More tuning the data ingestion stuff.