Commit graph

4 commits

Author SHA1 Message Date
e9eae205f0 use regression to monitor short term and long term trends
Since the platform calibration inevitably incurs some additional error band,
a linear regresssion over the time series of measurements can additionally be used
to spot ongoing systematic changes below this general error band, while
leveling out local statistical fluctuations.
2024-03-10 23:36:38 +01:00
8f9e54b7c7 implement fitting the platform model by linear regression
* triggered by --calibrate
 * normalise away any known expense factors, but use them as weight
 * calculate simple linear regression from statistic data
2024-03-10 23:36:38 +01:00
015a6ed6f2 add global storage and apply existing platform model
...this is largely just wiring of components built thus far

...TODO build platform model by linear regression
2024-03-10 23:36:38 +01:00
75767a3a97 capture and store individual timings as time series
Note: work-in-progress...
TODO: derive the expense factor and delta
2024-03-10 23:36:38 +01:00