Releases: SAFEtoolbox/SAFE-python
Releases · SAFEtoolbox/SAFE-python
SAFEpython v0.2.0
Remove dependency to numba.
SAFEpython v0.1.1
Revise folder structure, dependencies and use lower case for package name (safepython).
SAFEpython v0.1.0
The following changes to the code were applied:
- Modifying the function split_sample (util.py) to ensure that values that are repeated several times belong to the same group (this function is used to split to input-output dataset in the PAWN method);
- Fixing the numba incompatibility issues: making the code compatible with new numba versions in the functions hbv_sim (HBV.py, note that a line of the code needs to be changed to run it with Python 2, see notes in the module), sobol_g_function (sobol_g.py) and empiricalcdf (util.py) and removing the use of numba in FAST.py;
- Small adjustments to the workflows.
SAFEpython v0.0.0
First numbered released containing:
- different Sensitivity Analysis methods:
- Elementary Effect Test (EET.py, method or Morris)
- Regional Sensitivity Analysis (implementation based on threshold RSA_thres.py and on grouping RSA_group.py)
- Variance-Based Sensitivity Analysis (VBSA.py, Sobol' method)
- Fourier Amplitude Sensitivity Test (FAST.py)
- the moment-independent PAWN method (PAWN.py, implementation given data)
- functions to perform the input sampling and Monte Carlo simulations (sampling.py, lhcube.py, and model_execution.py)
- functions to visualize the sensitivity analysis results (plot_functions.py)
- other utility functions (util.py)
- test functions (ishigami_homma.py and sobol_g.py) and model examples (two hydrological models HyMod.py and HBV.py)
- workflows that demonstrate the application of the Sensitivity Analysis methods to the test functions and model examples