Rule#2:First,designandimplementmetrics.
Beforeformalizingwhatyourmachinelearningsystemwilldo,trackasmuchaspossibleinyour
currentsystem.Dothisforthefollowingreasons:
1. Itiseasiertogainpermissionfromthesystem’susersearlieron.
2. Ifyouthinkthatsomethingmightbeaconcerninthefuture,itisbettertogethistorical
datanow.
3. Ifyoudesignyoursystemwithmetricinstrumentationinmind,thingswillgobetterfor
youinthefuture.Specifically,youdon’twanttofindyourselfgreppingforstringsinlogs
toinstrumentyourmetrics!
4. Youwillnoticewhatthingschangeandwhatstaysthesame.Forinstance,supposeyou
wanttodirectlyoptimizeonedayactiveusers.However,duringyourearlymanipulations
ofthesystem,youmaynoticethatdramaticalterationsoftheuserexperiencedon’t
noticeablychangethismetric.
GooglePlusteammeasuresexpandsperread,resharesperread,plusonesperread,
comments/read,commentsperuser,resharesperuser,etc.whichtheyuseincomputingthe
goodnessofapostatservingtime.Also,notethatanexperimentframework,whereyou
cangroupusersintobucketsandaggregatestatisticsbyexperiment,isimportant.See
Rule#12.
Bybeingmoreliberalaboutgatheringmetrics,youcangainabroaderpictureofyoursystem.
Noticeaproblem?Addametrictotrackit!Excitedaboutsomequantitativechangeonthelast
release?Addametrictotrackit!
Rule#3:Choosemachinelearningoveracomplexheuristic.
Asimpleheuristiccangetyourproductoutthedoor.Acomplexheuristicisunmaintainable.
Onceyouhavedataandabasicideaofwhatyouaretryingtoaccomplish,moveontomachine
learning.Asinmostsoftwareengineeringtasks,youwillwanttobeconstantlyupdatingyour
approach,whetheritisaheuristicoramachinelearnedmodel,andyouwillfindthatthe
machinelearnedmodeliseasiertoupdateandmaintain(seeRule#16).
MLPhaseI:YourFirstPipeline
Focusonyoursysteminfrastructureforyourfirstpipeline.Whileitisfuntothinkaboutallthe
imaginativemachinelearningyouaregoingtodo,itwillbehardtofigureoutwhatishappening
ifyoudon’tfirsttrustyourpipeline.
Rule#4:Keepthefirstmodelsimpleandgettheinfrastructureright.
Thefirstmodelprovidesthebiggestboosttoyourproduct,soitdoesn'tneedtobefancy.But
youwillrunintomanymoreinfrastructureissuesthanyouexpect.Beforeanyonecanuseyour
fancynewmachinelearningsystem,youhavetodetermine:
评论0
最新资源