• correct()
  • Better support for parallel cores
  • classification and regression
  • Make the extract more flexible
  • Integrate figures with labels into one
  • classification across timepints and across trials
  • roll_on_level on time and time_group

  • Fixed an issue when the model in classification won't ouput weights.


  • Added a slot supplement to Analyzed_data
  • Supported multiple (>2) return values for roll_on_levels
  • The output of classification will include weights, and it will be at attribute supplement of Analyzed_data
  • Add default parameters for the topographical plot in matrix_plot
  • If the parameter positions's value of float_plot is "channels", positions will be set as io.load_topolocs('standard-10-5-cap385', None)
  • Make the default values annotation=None and fig=None for float_plot and matrix_plot

  • Changed the Y-axis title to power for Spectrum
  • Changed the printing format of AnalyzeData

  • Fixed a issue in multiple_comparison_correction
  • Fixed a issue of loading AnalyzeData


  • Removed collection from the hierarchical Extracted_epochs. Now there are onlycase and batch
  • New method subtract(raw, between, align)
  • Fixed an issue of trial number when read more than one files of one subject.
  • Changed two_sample test hypothesis from 'two-sides' to 'greater' in classification
  • Added parameter samplerate in load_raw_eeg
  • Set default values for parameter plot_params and ax in plot_waveform, plot_spectrum, plot_heatmap, and plot_topograph
  • Supported to output the scores of different subjects in classification and tanova
  • Changed the smooth to cluster in multiple_comparison_correction

  • Now the AnalyzeData has the repr method
  • Fixed a issue of TANOVA plot
  • Fixed a issue of saving AnalyzeData
  • Renamed the method RMS to GFP

  • Renamed the parameter mode to strategy in the method TANOVA

  • Now Time_frequency supports the power difference between two conditions
  • Fixed a issue when plot the heatmap

0.8.2 (2018-03-05)

  • Removed time_group index which are unnecessary in Analyzed_Result
  • Refine X axis in figure
  • Improved the compatibility of save() and load() of Analyzed_Result
  • New algorithm frequency_topography(). Now we can calculate the topography of frequency.
  • New method convert() in module Basic

0.8.1 (2018-03-01)

  • MASSIVE amount of changes
  • Altered the package name to "easyEEG"

0.8 (2017-12-25)

  • First to public release. Merry Xmax!