azul
azul
**Describe the bug** Inspired by [this issue](https://github.com/Nixtla/hierarchicalforecast/issues/30), we ran some [experiments](https://github.com/Nixtla/hierarchicalforecast/tree/main/experiments/libs-comparison) using the `AutoETS` model. We found that for some unstable series (with outliers, mainly), the forecast performance of this...
Add CI/CD for tsforecast (the docker image needs to be updated each time).
Check CI/CD behavior on push and pull_request. I think it should be different for docker and lambda. Just think about it.
I'm trying to fit `auto_arima` for the M4-Hourly dataset. You can download it [here](https://auto-arima-results.s3.amazonaws.com/M4-Hourly.csv). I'm using the following code, ``` import time from functools import partial from multiprocessing import Pool,...
- feat: new version GA sklearn syntax - feat: add functional fit/predict - feat: new tests - feat: add fit/predict classes - feat: add repr - feat: add core version...
See #175, and https://github.com/Nixtla/hierarchicalforecast/issues/34
**Is your feature request related to a problem? Please describe.** Enhance the available models, including `mstl` (to handle multiple seasonalities).
Add python 3.10 to `ci.yml` for windows once https://github.com/ray-project/ray/issues/26378 is solved.
**Is your feature request related to a problem? Please describe.** `ray` needs additional dependencies to run with AWS (`boto3` and `botocore`). **Describe the solution you'd like** It would be great...
**Describe the bug** When `ds` is a `object`, the method `forecast` arises an error but at the end of the pipeline, once the forecasts are computed. **Expected behavior** Check `ds`...