Maryoda2

In the example of raw_data, it is by minute. So 1440 because of 24 hrs * 60 minutes whiwh is equal to 1440

Hi team, Sorry for this up, I mean really. But with this script : library(AnomalyDetection) data(raw_data) data res$plot **Error: Column `x` is a date/time and must be stored as POSIXct,...

Oh thanks a lot !

Well... in fact sometimes it writes : Error in aggregate.data.frame(x[2], format(x[1], "%Y-%m-%d %H:%M:00"), : 'by' must be a list

library(AnomalyDetection) data

> str(raw_data) 'data.frame': 14398 obs. of 2 variables: $ timestamp: POSIXct, format: "1980-09-25 10:01:00" "1980-09-25 10:02:00" ... $ count : num 182 176 184 178 165 ... > str(dems) Classes...

Oh I see, How can I edit the code of twitter/AnomalyDetection, in fact the caijun/Twitter (yours) ?

Well thanks for your answer With isalgueiro it doesn't work, same error. Here is a sample

20.000 is no more a sample, but we can't detect anomalies on dataset less than 20.000 ?

In fact for the moment this is all what I have... So game over then ?