analysis.py v 1.1.6.002

This commit is contained in:
art 2019-11-05 12:56:53 -06:00
parent bfa69bf070
commit 3c21512a46

View File

@ -7,165 +7,167 @@
# current benchmark of optimization: 1.33 times faster # current benchmark of optimization: 1.33 times faster
# setup: # setup:
__version__ = "1.1.6.001" __version__ = "1.1.6.002"
# changelog should be viewed using print(analysis.__changelog__) # changelog should be viewed using print(analysis.__changelog__)
__changelog__ = """changelog: __changelog__ = """changelog:
1.1.6.001: 1.1.6.002:
- added additional hyperparameters to decisiontree() - changed layout of __changelog to be vscode friendly
1.1.6.000: 1.1.6.001:
- fixed __version__ - added additional hyperparameters to decisiontree()
- fixed __all__ order 1.1.6.000:
- added decisiontree() - fixed __version__
1.1.5.003: - fixed __all__ order
- added pca - added decisiontree()
1.1.5.002: 1.1.5.003:
- reduced import list - added pca
- added kmeans clustering engine 1.1.5.002:
1.1.5.001: - reduced import list
- simplified regression by using .to(device) - added kmeans clustering engine
1.1.5.000: 1.1.5.001:
- added polynomial regression to regression(); untested - simplified regression by using .to(device)
1.1.4.000: 1.1.5.000:
- added trueskill() - added polynomial regression to regression(); untested
1.1.3.002: 1.1.4.000:
- renamed regression class to Regression, regression_engine() to regression gliko2_engine class to Gliko2 - added trueskill()
1.1.3.001: 1.1.3.002:
- changed glicko2() to return tuple instead of array - renamed regression class to Regression, regression_engine() to regression gliko2_engine class to Gliko2
1.1.3.000: 1.1.3.001:
- added glicko2_engine class and glicko() - changed glicko2() to return tuple instead of array
- verified glicko2() accuracy 1.1.3.000:
1.1.2.003: - added glicko2_engine class and glicko()
- fixed elo() - verified glicko2() accuracy
1.1.2.002: 1.1.2.003:
- added elo() - fixed elo()
- elo() has bugs to be fixed 1.1.2.002:
1.1.2.001: - added elo()
- readded regrression import - elo() has bugs to be fixed
1.1.2.000: 1.1.2.001:
- integrated regression.py as regression class - readded regrression import
- removed regression import 1.1.2.000:
- fixed metadata for regression class - integrated regression.py as regression class
- fixed metadata for analysis class - removed regression import
1.1.1.001: - fixed metadata for regression class
- regression_engine() bug fixes, now actaully regresses - fixed metadata for analysis class
1.1.1.000: 1.1.1.001:
- added regression_engine() - regression_engine() bug fixes, now actaully regresses
- added all regressions except polynomial 1.1.1.000:
1.1.0.007: - added regression_engine()
- updated _init_device() - added all regressions except polynomial
1.1.0.006: 1.1.0.007:
- removed useless try statements - updated _init_device()
1.1.0.005: 1.1.0.006:
- removed impossible outcomes - removed useless try statements
1.1.0.004: 1.1.0.005:
- added performance metrics (r^2, mse, rms) - removed impossible outcomes
1.1.0.003: 1.1.0.004:
- resolved nopython mode for mean, median, stdev, variance - added performance metrics (r^2, mse, rms)
1.1.0.002: 1.1.0.003:
- snapped (removed) majority of uneeded imports - resolved nopython mode for mean, median, stdev, variance
- forced object mode (bad) on all jit 1.1.0.002:
- TODO: stop numba complaining about not being able to compile in nopython mode - snapped (removed) majority of uneeded imports
1.1.0.001: - forced object mode (bad) on all jit
- removed from sklearn import * to resolve uneeded wildcard imports - TODO: stop numba complaining about not being able to compile in nopython mode
1.1.0.000: 1.1.0.001:
- removed c_entities,nc_entities,obstacles,objectives from __all__ - removed from sklearn import * to resolve uneeded wildcard imports
- applied numba.jit to all functions 1.1.0.000:
- depreciated and removed stdev_z_split - removed c_entities,nc_entities,obstacles,objectives from __all__
- cleaned up histo_analysis to include numpy and numba.jit optimizations - applied numba.jit to all functions
- depreciated and removed all regression functions in favor of future pytorch optimizer - depreciated and removed stdev_z_split
- depreciated and removed all nonessential functions (basic_analysis, benchmark, strip_data) - cleaned up histo_analysis to include numpy and numba.jit optimizations
- optimized z_normalize using sklearn.preprocessing.normalize - depreciated and removed all regression functions in favor of future pytorch optimizer
- TODO: implement kernel/function based pytorch regression optimizer - depreciated and removed all nonessential functions (basic_analysis, benchmark, strip_data)
1.0.9.000: - optimized z_normalize using sklearn.preprocessing.normalize
- refactored - TODO: implement kernel/function based pytorch regression optimizer
- numpyed everything 1.0.9.000:
- removed stats in favor of numpy functions - refactored
1.0.8.005: - numpyed everything
- minor fixes - removed stats in favor of numpy functions
1.0.8.004: 1.0.8.005:
- removed a few unused dependencies - minor fixes
1.0.8.003: 1.0.8.004:
- added p_value function - removed a few unused dependencies
1.0.8.002: 1.0.8.003:
- updated __all__ correctly to contain changes made in v 1.0.8.000 and v 1.0.8.001 - added p_value function
1.0.8.001: 1.0.8.002:
- refactors - updated __all__ correctly to contain changes made in v 1.0.8.000 and v 1.0.8.001
- bugfixes 1.0.8.001:
1.0.8.000: - refactors
- depreciated histo_analysis_old - bugfixes
- depreciated debug 1.0.8.000:
- altered basic_analysis to take array data instead of filepath - depreciated histo_analysis_old
- refactor - depreciated debug
- optimization - altered basic_analysis to take array data instead of filepath
1.0.7.002: - refactor
- bug fixes - optimization
1.0.7.001: 1.0.7.002:
- bug fixes - bug fixes
1.0.7.000: 1.0.7.001:
- added tanh_regression (logistical regression) - bug fixes
- bug fixes 1.0.7.000:
1.0.6.005: - added tanh_regression (logistical regression)
- added z_normalize function to normalize dataset - bug fixes
- bug fixes 1.0.6.005:
1.0.6.004: - added z_normalize function to normalize dataset
- bug fixes - bug fixes
1.0.6.003: 1.0.6.004:
- bug fixes - bug fixes
1.0.6.002: 1.0.6.003:
- bug fixes - bug fixes
1.0.6.001: 1.0.6.002:
- corrected __all__ to contain all of the functions - bug fixes
1.0.6.000: 1.0.6.001:
- added calc_overfit, which calculates two measures of overfit, error and performance - corrected __all__ to contain all of the functions
- added calculating overfit to optimize_regression 1.0.6.000:
1.0.5.000: - added calc_overfit, which calculates two measures of overfit, error and performance
- added optimize_regression function, which is a sample function to find the optimal regressions - added calculating overfit to optimize_regression
- optimize_regression function filters out some overfit funtions (functions with r^2 = 1) 1.0.5.000:
- planned addition: overfit detection in the optimize_regression function - added optimize_regression function, which is a sample function to find the optimal regressions
1.0.4.002: - optimize_regression function filters out some overfit funtions (functions with r^2 = 1)
- added __changelog__ - planned addition: overfit detection in the optimize_regression function
- updated debug function with log and exponential regressions 1.0.4.002:
1.0.4.001: - added __changelog__
- added log regressions - updated debug function with log and exponential regressions
- added exponential regressions 1.0.4.001:
- added log_regression and exp_regression to __all__ - added log regressions
1.0.3.008: - added exponential regressions
- added debug function to further consolidate functions - added log_regression and exp_regression to __all__
1.0.3.007: 1.0.3.008:
- added builtin benchmark function - added debug function to further consolidate functions
- added builtin random (linear) data generation function 1.0.3.007:
- added device initialization (_init_device) - added builtin benchmark function
1.0.3.006: - added builtin random (linear) data generation function
- reorganized the imports list to be in alphabetical order - added device initialization (_init_device)
- added search and regurgitate functions to c_entities, nc_entities, obstacles, objectives 1.0.3.006:
1.0.3.005: - reorganized the imports list to be in alphabetical order
- major bug fixes - added search and regurgitate functions to c_entities, nc_entities, obstacles, objectives
- updated historical analysis 1.0.3.005:
- depreciated old historical analysis - major bug fixes
1.0.3.004: - updated historical analysis
- added __version__, __author__, __all__ - depreciated old historical analysis
- added polynomial regression 1.0.3.004:
- added root mean squared function - added __version__, __author__, __all__
- added r squared function - added polynomial regression
1.0.3.003: - added root mean squared function
- bug fixes - added r squared function
- added c_entities 1.0.3.003:
1.0.3.002: - bug fixes
- bug fixes - added c_entities
- added nc_entities, obstacles, objectives 1.0.3.002:
- consolidated statistics.py to analysis.py - bug fixes
1.0.3.001: - added nc_entities, obstacles, objectives
- compiled 1d, column, and row basic stats into basic stats function - consolidated statistics.py to analysis.py
1.0.3.000: 1.0.3.001:
- added historical analysis function - compiled 1d, column, and row basic stats into basic stats function
1.0.2.xxx: 1.0.3.000:
- added z score test - added historical analysis function
1.0.1.xxx: 1.0.2.xxx:
- major bug fixes - added z score test
1.0.0.xxx: 1.0.1.xxx:
- added loading csv - major bug fixes
- added 1d, column, row basic stats 1.0.0.xxx:
- added loading csv
- added 1d, column, row basic stats
""" """
__author__ = ( __author__ = (