v1.1.4.1

• Hot fix on requirement.txt to set up fix on conda forge as well
• Reduce number of python versions in unit test

v1.1.4

Core Changes:

• replace PyStan2 by cmdstanpy (#801)
• update requirement and installation procedures (073f03f)
• loggers upgrade

Documentation:

• all tutorial is updated to use cmdstanpy in backend

v1.1.3

Core changes:

• add python 3.8 unit tests (#752)
• optimize interface to be compatible with arviz (#755)
• requirements update (#763)
• code clean up (#765)
• dlt global trend prior adjustment (#786)

Documentation:

Tutorial enhancement:

• tutorial refresh (#795)

Utilities:

• uses tqdm in parameters tuning (#762)
• residuals plot (#758)
• simpler stan compile interface (#769)

v1.1.2

Core changes:

• Add Conda installation option (#679)
• Suppress the lengthy Stan logging message (#696)
• WBIC for pyro SVI sampling and BIC for MAP optimization (#719, #710)
• Backtest module to include confidence intervals (#724)
• Allow configuration for compiled Stan model path (#713)
• Box plot for regression coefficient comparison (#737)
• Bounded logistic growth for DLT model (#712)
• Enhance regression output reporting (#739)å

Documentation:

• Add blacking linting to Github action workflow (#708)
• Tutorial enhancement

Utilities:

• Add a new method make_future_df to prepare data frame for forecasting (#695)

v1.1.2alpha

Core changes:

• Add Conda installation option (#679)
• Suppress the lengthy Stan logging message (#696)
• WBIC for pyro SVI sampling and BIC for MAP optimization (#719, #710)
• Backtest module to include confidence intervals (#724)
• Allow configuration for compiled Stan model path (#713)
• Box plot for regression coefficient comparison (#737)
• Bounded logistic growth for DLT model (#712)
• Enhance regression output reporting (#739)

Documentation:

• Add blacking linting to Github action workflow (#708)
• Tutorial enhancement

Utilities:

• Add a new method make_future_df to prepare data frame for forecasting (#695)

v1.1.1

• fix the .mplstyle file path bug

v1.1.0

Core changes

• Redesign the model class structure with three core components: model template, estimator, and forecaster
  (#506, #507, #508, #513)
• Introduce the Kernel-based Time-varying Regression (KTR) model (#515)
• Implement the negative coefficient for LGT and KTR (#600, #601, #609)
• Allow to handle missing values in response for LGT and DLT (#645)
• Implement WBIC value for model candidate selection (#654)

Documentation

• A new series of tutorials for KTR (#558, #559)
• Migrate the CI from TravisCI to Github Actions (#556)
• Missing value handle tutorial (#645)
• WBIC tutorial (#663)

Utilities

• New Plotting Palette (#571, #589)
• Redesign the diagnostic plotting (#581, #607)
• Raise a warning when date index is not evenly distributed (#639)

v1.0.17

• Core changes:
  • Use global mean instead of median in ktrx model before next major release

v1.0.16

• Core changes
  • Bug fix and code improvement before next major release (#540, #541, #546)
  • lower than matplotlib requirement (#498)

v1.0.15

• Core changes:

  • Prediction functionality refactoring (#430)
  • KTRLite model enhancement and interface cleanup (#440)
  • More flexible scheduling config in Backtester (#447)
  • Allow extraction of training related metrics (e.g. ELBO loss) in Pyro SVI (#443)
  • Add a flag to keep the posterior samples or not in aggregated model (#465)
  • Bug fix and code improvement (#428, #438, #459, #470)

• Documentation:

  • Clean up and standardize example notebooks (#462)
  • Tutorial update and enhancement (#431, #474)

• Utilities:

  • Diagnostic plot with Arviz (#433)
  • Refine plotting palette (#434, #473)
  • Create an orbit-featured plotting style (#434)