finnts
finnts copied to clipboard
microsoft
Microsoft Finance Time Series Forecasting Framework (FinnTS) is a forecasting package that utilizes cutting-edge time series forecasting and parallelization on the cloud to produce accurate forecasts...
prevent very large forecasts which can create errors when writing to disk
The user may or may not specify `date_range`. It's not needed given it can be calculated. By default I propose setting `date_type = NULL` so it happens automatically instead of...
Hi, Thank you for your work on this package. I'm running into an issue with the created Date features when setting `fiscal_year_start` to a value other than 1. For example,...
I have just installed the package and tried smaple code from GitHub Readme: ``` library(finnts) # prepare historical data hist_data % dplyr::rename(Date = date) %>% dplyr::mutate(id = as.character(id)) # call...
Look into adversarial validation to help detect data drift and when to retrain/reselect models.
(disclaim: I'm the author) To minimize the risk of overusing CPUs on multi-tenant and multi-process machines, but also to work in more environments, please consider changing: https://github.com/microsoft/finnts/blob/57207bf18b79949244017c4df8ec1296a9da0752/R/general_parallel.R#L6-L13 to ```r get_cores
This is a great package for automated large scale forecasts! Currently (0.1.0), there are lot of imports. On some systems, it takes a long time and dependency packages will not...
To help the model distinguish between different groups in the data. May only be applicable to global models.
When customers submit a daily forecast, no matter what, we force the lag_periods to be a fixed value in multivariate_prep_recipe_2: ``` r lag_periods_r2
