Included evaluation of true coverage
For simulated examples the true coverage is evaluated. For all other datasets create_tabular will show None. The coverage was reversed to use the prediction (mean under the posterior predictive). Note, that the "repeated sampling" assumption is treated indirectly via averaging over the 10 seeds. For simulated data this works decently, for real data one often (but not always, cf. wine) sees a discrepancy.
No related branches found
No related tags found
Showing
- EIVPackage/EIVData/linear.py 19 additions, 9 deletionsEIVPackage/EIVData/linear.py
- EIVPackage/EIVData/quadratic.py 23 additions, 12 deletionsEIVPackage/EIVData/quadratic.py
- EIVPackage/EIVGeneral/coverage_metrics.py 35 additions, 28 deletionsEIVPackage/EIVGeneral/coverage_metrics.py
- Experiments/create_tabular.py 23 additions, 6 deletionsExperiments/create_tabular.py
- Experiments/evaluate_metrics.py 57 additions, 15 deletionsExperiments/evaluate_metrics.py
- Experiments/train_eiv.py 1 addition, 1 deletionExperiments/train_eiv.py
- Experiments/train_noneiv.py 1 addition, 1 deletionExperiments/train_noneiv.py
Loading