Released on 2023-07-14
Uploaded to PyPI : pip install pyaf
Changelog :
- Python 3.11 Support :
Python 3.11 support #227 - New Hardware Support :
RISC-V Hardware Platform Validation #208 - New Performance Measures :
Outlier-resistant forecasting Performance Measures #209,
Add Differentiable Variant of SMAPE Performance Measure #221 - Model Selection Improvement :
Investigate Model Esthetics for PyAF #212,
Investigate Large Horizon Models #213 ,
Revisit Model Complexity Definition #223,
Use MASE by default for PyAF Model Selection #229 - Signal Transformation Improvements :
Use MaxAbsScaler for some Multiplicative Signal Transformations #235,
Pyaf 5.0 Final Touch 8 : Use an Optimal Choice Rule for the Quantization Signal transform #239 - Generic Modeling :
PyAF 5.0 Final Touch 1 : discard some non-significant components #230,
PyAF 5.0 Final Touch 2: Disable alpha in ridge regressions #231,
Pyaf 5.0 Final Touch 5 : Add more info about Exogenous Data Used in ARX Models #236,
Pyaf 5.0 Final Touch 7 : Improve the Guess of Window Length for Moving Average Trends #238 - Plotting Functions Improvements and Bug Fixes :
Bad plot for shaded area around prediction intervals in hourly data #216,
Forecast Quantiles Plots Improved #225,
Pyaf 5.0 Final Touch 3 : report plot filenames in the logs #232 - New Docs :
Provide some UML docs for PyAF integrators #233 - Bug Fixes :
Failure to build a multiplicative ozone model with Lag1 trend #220 - PyAF "Forecast Tasks" :
Use PyTorch as the reference deep learning framework/architecture for future projects #211,
Automate Prototyping Activities - R-based Models #217 - Recurrent Tasks :
Re-run the Benchmarking process for PyAF 5.0 #222,
Run some Sanity Checks for PyAF 5.0 #224,
Pyaf 5.0 Final Touch 4 : Add More Tests #234,
Pyaf 5.0 Final Touch 6 : Disable Timing Loggers by default #237