Data¶
Uncertainpy stores all results from the uncertainty quantification and
sensitivity analysis in UncertaintyQuantification.data,
as a Data object.
The Data class works similarly to a Python dictionary.
The name of the model or feature is the key,
while the values are DataFeature objects that stores each
statistical metric in in the table below as attributes.
Results can be saved and loaded through
Data.save and Data.load.
| Calculated statistical metric | Symbol | Variable |
|---|---|---|
| Model and feature evaluations | \(U\) | evaluations |
| Model and feature times | \(t\) | time |
| Mean | \(\mathbb{E}\) | mean |
| Variance | \(\mathbb{V}\) | variance |
| 5th percentile | \(P_{5}\) | percentile_5 |
| 95th percentile | \(P_{95}\) | percentile_95 |
| First order Sobol indices | \(S\) | sobol_first |
| Total order Sobol indices | \(S_T\) | sobol_total |
| Average of the first order Sobol indices | \(\widehat{S}\) | sobol_first_average |
| Average of the total order Sobol indices | \(\widehat{S}_{T}\) | sobol_total_average |
An example: if we have performed uncertainty quantification of a spiking neuron model with the number of spikes as one of the features, we get load the data file and get the variance of the number of spikes by typing:
data = un.Data()
data.load("filename")
variance = data["nr_spikes"].variance