Metadata-Version: 2.1
Name: microfilter
Version: 0.0.9
Summary: Filtering noisy data
Home-page: https://github.com/microprediction/microfilter
Author: microprediction
Author-email: pcotton@intechinvestments.com
License: MIT
Description: # microfilter
        
        Some ad-hoc approaches to filtering noisy data that don't appear in textbooks
        
        ![](https://i.imgur.com/b5fAtxr.png)
        
        ## Usage example
        
        Train filter on simulated noisy data
        
            from microfilter.univariate.expnormdist import ExpNormDist
            from microfilter.univariate.noisysim import sim_lagged_values_and_times
            
            lagged_values, lagged_times = sim_lagged_values_and_times
            dist = ExpNormDist()
            dist.hyper_params['max_evals']=500
            dist.fit(lagged_values=lagged_values, lagged_times=lagged_times)
            pprint(dist.params) 
            new_value = 17.0
            dist.update(value=new_value, dt=1.0)
            pprint(dist.state) 
            
        See https://github.com/microprediction/microfilter/blob/master/examples/plot_expnorm.py 
            
             
        
Platform: UNKNOWN
Classifier: License :: OSI Approved :: MIT License
Classifier: Programming Language :: Python :: 3
Classifier: Programming Language :: Python :: 3.7
Description-Content-Type: text/markdown
