ibis-ml
ibis-ml copied to clipboard
docs: build demo workflows
~We are currently targeting the NVTabular demo on RecSys2020 Challenge as a demo workflow.~ Update: Due to the RecSys2020 demo data being unavailable (and against Twitter's terms to share), we will start with the R nycflights13 dataset. It has been added to Ibis examples to support this.
Major tasks
- Demo Dataset
- nycflights13
- Feature engineering
- Use ibis and ibisml for data preprocessing
- Model training
- XGBoost
- sklearn
- PyTorch
Initial version merged in #60. Some remaining TODOs:
- [ ] Make sure unique key is actually unique
- [ ] Do we want to do something with weather data? We join it, and then throw away all the columns...
- [x] Include
TargetEncodingstep - [x] Clearly document handoff to other modeling frameworks (did test
XGBClassifierworks, but add it to a notebook; also, make sure PyTorch demo is there, maybe something likeskorchwith MLP)... (maybe add model choice to demo notebook) - [x] Once Ibis 9.0 is released, update demo notebook to pull directly from
ibis.examples.nycflights13_*.fetch()instead of local DuckDB
Initial version merged in #60. Some remaining TODOs:
- [ ] Make sure unique key is actually unique
- [ ] Do we want to do something with weather data? We join it, and then throw away all the columns...
- [x] Include
TargetEncodingstep- [x] Clearly document handoff to other modeling frameworks (did test
XGBClassifierworks, but add it to a notebook; also, make sure PyTorch demo is there, maybe something likeskorchwith MLP)... (maybe add model choice to demo notebook)- [x] Once Ibis 9.0 is released, update demo notebook to pull directly from
ibis.examples.nycflights13_*.fetch()instead of local DuckDB
The remaining TODOs don't seem like a pressing priority; will close this as completed.