pietroppeter
commented
1 year ago - me
- toolsgroup: demand forecasting in supply chain (ML)
- nim
- time series forecasting
- why
- use cases
- demand forecasting
- plot
- with matplotlib/seaborn: passengers
- with vegalite: weather
- with plotly: demand (M5)
- basics
- naive
- exp smoothing
- decomposition: additive/multiplicative
- benchmarking and evaluation
- M4 and M5
- metrics
- list of timeseries libraries (R ecosystem, statsmodels, gluonts, sktime, darts, prophet, ... nixtla)
- nixtla
- repo overview
- business: prediction at scale
- statsforecast
- hierarchical forecast
- tsfeatures
- mlforecast
- neuralforecast
- prophet and the zillow story? likely as a follow up slide
- references
- hyndman (IT!)
- banachewicz on kaggle/youtube
- articles by Nixtla
- articles on prophet?
need to list an overview of slides