Feature Abstraction
Optimize Your Model
Pyhon software coming soon to GitHub
The software package provides end users an innovative way to fine tune feature hyper parameters and optimize model performance for multiclassification problems. Read through the following documentation to find how to get started building better performing models today.
â€‹
Introduction
use case & beyond
Fine tune time series filter hyper parameters to maximize feature performance. The following methodology can be used for other feature optimization methods. The following criteria must be met.
â€‹

The model must be classification

Data must continuous
Optimize Feature
Distribution Analysis
Signal Processing
Calculate Performance
Time Series Filter
ema_optimization
The time series filter consists of three hyper parameters every parameter has the no linear domain [0,1] explained below.
â€‹

Decay a, How rapidly the the filter exponentially decreases

Shift b, The duration of the domain with the maximum weight

Delta c, Smoothing factor between decay and shift
Augment Feature
Distribution Analysis
Signal Processing
Calculate Performance
Time Series Filter Optimization
ema_optimization
To optimize the time series filter hyper parameters bayesian optimization and Gaussian processes are applied to reduce algorithm run time and maximize performance. In addition the algorithm utilizes a batch design from later works which mimics the optimal sequential policy.
Decay
Shift
Delta
Bayesian Optimization
Optimize Score
Augment Feature
Optimize Feature
Signal Processing
Calculate Performance
Distribution Analysis
distribution_optimizer
Transform raw data to a gaussian distributed space using a performance metric. If the data has been found to be gaussian distributed it can be used to:
â€‹

Score feature performance in multiclassification models

Dimensionality reduction via gaussian processes
â€‹
Augment Feature
Optimize Feature
Distribution Analysis
Calculate Performance
Signal Processing
distribution_distillation
The signal processing class uses Bayesian optimization to filter noise and isolate key areas of discontinuity between multiclass distributions.
The area under the curve is then used to weight the importance of the probability the correct multiclass distribution was identified.
weighted by the of the optimization heuristic is derived from discontinuity between classification distributions being able to correctly identify the appropriate classification.