statsforecast VS flow-forecast

I can't find the TimeGPT-1 model.

LICENSE Apache-2

https://github.com/Nixtla/statsforecast/blob/main/LICENSE

Mentions ARIMA, ETS, CES, and Theta modeling

If you only have historical data: StatsForecast

I also recommend you check Nixtla's libraries, in particular StatsForecast and HierarchicalForecast. They offer a wide selection of forecasting models, and can work with multiple time series. Given that you're working with many products in a warehouse, I think the hierarchical forecast can be very useful, especially for the short time series (the ones that don't seem to have enough time stamps).

If you are mostly worried about time and use python you could try out Nixtla's statsforecast as it is very snappy. https://github.com/Nixtla/statsforecast

I was researching about using deep learning for time series forecasting applications when I came across two experiments by the Nixtla team. They showed that their traditional statistical ensemble (comprised of AutoARIMA, ETS, CES, and DynamicOptimizedTheta) beat a bunch of deep learning models (link) and also the AWS forecast API (link).

- https://nixtla.github.io/statsforecast/

Leaving these two repos here for anyone interested in trying decision tree regression or statistical forecasting baselines: - https://nixtla.github.io/mlforecast/ - https://github.com/Nixtla/statsforecast

In this reproducible experiment, we compare Amazon Forecast and StatsForecast a python open-source library for statistical methods.

-Flow

How does it compare to Flow Forecast? Honestly people rarely mention FF but I've found it much better than pytorch_forecasting and the like.

PyTorch forecasting has a lot of bugs in it. You should try posting on issue on the actual repository though. Also, I've found Flow Forecast to be an all around much better deep learning for time series forecasting/classification framework.

If you want to use deep learning then Flow Forecast is the best. Many of the latest deep learning models and easy hyper-parameter sweeps.

I haven't personally tried it, but here's a Github Repo called LIME for Time. I'm not sure about the state of attention visualization for timeseries but this repo has several models using attention.

Hey everyone, just released a new Flow Forecast a deep learning for time series forecasting framework written in PyTorch. For this new version we added models, fixed several annoying bugs, and created better error messages. See the improved framework and tutorials Link

This is actually a good case for deep learning techniques that create a learned time series embedding id and/or graph convolutions. The advantage of these methods is the can [learn spatial temporal dependencies across several time series](Cross-Dimensional Self-Attention for Multivariate, Geo-tagged Time Series Imputation). There are some implementations that you can find of these models that you can find in the repository that deep learning for time series repo I maintain.

I mean a lot of deep learning models are more interpretable than you would think. There are a lot of methods to explain model predictions. In the deep learning for time series forecasting framework that I help maintain for example we automatically create SHAP plots to show relevant features. You can also visualize the attention mechanism directly.