anomalize
anomalize copied to clipboard
business-science
Tidy anomaly detection
I am trying to use anomalize for heatlhcare data. However, I am only interested in positive anomalies. I have looked in the anomalize documentation package, but I am still not...
Hello, I have the following errors while using MetaVolcanoR Pacakge. Error 1: > meta_degs_vote meta_degs_rem
I am looking at some data that has natural spikes when it rains. I'm having trouble tuning anomalize not to flag these data as anomalies. Any suggestions? 
I've been searching the documentation but can't find the answer. e.g. for the cran tidyverse example is each package analysed separately or is each date considered a multivariate point? Thanks...
Especially for the 1st plot 'observed', and the 3rd plot 'trend', how can I free the y-axis? Thank you.
Hi! using: time_decompose(count,method='twitter') gives me the error... `` Is this the right/intended error message? I'm confused whether the twitter method is used instead of stl at all. Thank you Matt.
Hi, I run the sample code from https://github.com/business-science/anomalize and there's an issue. Please see below:  I tried the same code on my own tibble table and I got...
`anom % group_by(date_only) %>% arrange(date_only) %>% summarize(total = sum(amount)) %>% as_tibble() anom_only % time_decompose(total) %>% anomalize(remainder) %>% time_recompose() %>% filter(anomaly == 'Yes')` This code works as intended on my Mac....
Dear All, Hopefully the reprex is self-explanatory. I plan to use anomalize on non-time series data. It should still work according to the documentation (without the time series decomposition) and...
`time_decompose()` accepts additional parameters `...` which are being passed to the underlying functions `decompose_stl()` and `decompose_twitter()`. However these functions currently don't have ellipsis. This PR suggests to allow additional parameters...
