Version History
v.0.1:
v0.1.0: Implemented.
v0.1.1: Split into different frameworks and adjust changes from based on new version of ebcpy
v0.1.2: Move CalibrationClass from ebcpy and add it to the general module aixcalibuha. Adjust Goals etc. based on changes in ebcpy.
v0.1.3: Remove Continuous Calibration methods and introduce new, better methods for calibration of multiple classes.
Issue 43: Same class now optimizes to one optimum instead of multiple. If an intersection in tuner parameters occurs, the statistics are logged and plotted so the user can better decide with what values to go on.
Issue 42: Visualizer is adjusted to better print the results more readable
Issue 39: Several kwargs are added for better user-interaction and plotting of multiple classes
Issue 46: Current best iterate is stored to ensure an interruption of a calibration won’t yield in a lost optimized value. Keyboard interrupt is now possible.
v0.1.4
Add Goals from ebcpy
Add new tutorial for a better start with the framework. (See Issue 49)
Make changes based on new version 0.1.5 in ebcpy
v0.1.5
Add new scripts in bin folder to ease the setup of the calibration for new users
Add configuration files and save/load classes
Issue 54: Skip failed simulations using two new kwargs in Calibrator class
Issue 53: Save final plots despite abortion of calibration process via STRG+C
Issue 51: Refactor reference_start_time to fix_start_time
Issue 23: Model Wrapper for MoCaTe files.
v0.1.6
Add Re-Calibration code from master thesis of Sebastian Borges
Add fixed_parameters to calibration
Re-add tunerParas from ebcpy
Make changes based on ebcpy v.0.1.7
Split SensivitiyAnalyzer class and use object oriented programming
v0.2.0
Adjust based on ebcpy v 0.2.0
Add examples and fix tutorial
Improve validation output
Fix version of SALib as 1.4 is not working
v0.2.1
Unfix version of SALib as 1.4.0.2 and 1.4.4 are not working
v0.2.2
Issue 21: Fix setup.py by removing the tests packages
v0.2.3
Add workflows
v0.3.0
Issue 20: Add parallelization for calibration and sensitivity analysis
Issue 32: Add example converter to CI/CD
v0.3.1
Issue 41: Fix logging and add kwarg
v1.0.0
Issue 43: Improvement of sensitivity analysis
Enables verbose sensitivity analysis and the reuse of simulations
It is now possible to use verbose sensitivity analysis for an automatic selection of tuner parameters
Enables multiprocessing for the entire sensitivity process
Sensitivity analysis is now usable for large models and data
Add time dependent sensitivity analysis
Ends support for python 3.7