Skip to content

Latest commit

 

History

History
119 lines (75 loc) · 3.3 KB

RELEASE.md

File metadata and controls

119 lines (75 loc) · 3.3 KB

Release log

0.5.6

August 2024

New features

  • Add more example Jupyter notebooks to demonstrate package syntaxes and algorithms

Closed issues

  • SOMATA basic model classes now perform more reliably and consistently, including automatic parsing of parameters to generate components
  • Multitaper spectrogram implementation used functions deprecated in Numpy 2.0
  • Observation matrix and noise covariance ignored if initializing through components

0.5.5

July 2024

Closed issues

  • Remove the use of dependency_links as it is now deprecated and ignored by pip
  • Fix missing dependency of statsmodels required for diagnostic tests
  • Include .stan files from the pac module with build distribution

0.5.4

July 2024

New features

  • Add extra dependency_links for Windows OS install of spectrum and torch

0.5.3

July 2024

New features

  • Improve install instructions for Windows OS and minor README fixes

0.5.2

July 2024

New features

  • Single source dependency specification into requirements-*.txt files
  • Use only pyproject.toml and remove setup.py/.cfg for building with setuptools

Closed issues

  • Failed dependency resolution with pip install on Windows OS

0.5.1

July 2024

New features

  • Add a phase amplitude coupling (PAC) analysis module
  • Add a new class to perform decimated EM learning with state-space models

Closed issues

  • Fix an issue with log likelihood calculations with multivariate observed data y
  • Rename initial state noise covariance matrix at time point t=0 from Q0 to S0
  • Correctly unfill components when calling m_estimate on Ssm class objects
  • Fix how different basic models can be concatenated together (e.g., Arn and Osc)

0.4.1

March 2024

New features

  • Introduce a spectral factorization method to initialize oscillator parameters
  • Introduce DecomposedOscillatorModel to supersede the iOsc algorithm in most applications
  • Rename iterative_oscillator module to oscillator_search
  • Introduce diagnostic plotting and statistical tests for analyzing residuals
  • Introduce dynamic source localization using oscillator models with GPU acceleration

Closed issues

  • Fix an issue with h_t and y lengths in _m_update_r
  • Fix an extra parameter in basic_models_demo
  • Unify the output arguments of spectrum as (psd, freq)
  • Update instruction on editable pip install to work with static code analyzers

0.3.1

June 2023

New features

  • Update iterative oscillator algorithm to use new routines
  • Introduce switching module for switching state-space models

Closed issues

  • Fix an issue of iOsc only adding AR1 component to slow oscillation
  • Fix plotting issues of the multitaper module

0.2.1

November 2022

New features

  • Introduce iterative oscillator algorithm

Closed issues

  • Fix mathjax rendering of state-space equations in README.md

0.1.1

October 2022

New features

  • Introduce four basic state-space model classes: StateSpaceModel, OscillatorModel, AutoRegModel, GeneralSSModel
  • Add exact inference algorithms with Gaussian noise processes for the introduced basic models
  • Add expectation-maximization(EM) learning algorithm using a general run_em() wrapper function

Closed issues

  • StateSpaceModel._initialize_from_components() method now works correctly with the .concat_() method