Metadata-Version: 2.1
Name: fastash
Version: 0.0.12
Summary: Fast Averaged Shifted Histogram module.
Home-page: https://gitlab.inria.fr/fmazy/fastash
Author: François-Rémi Mazy
Author-email: francois-remi.mazy@inria.fr
License: BSD 3-Clause License
Description: # FastASH
        Fast Averaged Shifted Histograms.
        
        Density estimation which approximate a triangular Kernel Density Estimation (KDE.
        Cython inside.
        
        
        ## Getting Started
        
        Example usage:
        
        For a standard PDF
        ~~~~~~~~~~~~~~~~~~
        
            from scipy import stats
            import numpy as np
            from matplotlib import pyplot as plt
            from fastash import ASH
            
            n = 10**5
            d = 3
            mean = np.zeros(d)
            cov = np.diag(np.ones(d))
            X = stats.multivariate_normal.rvs(mean=mean, cov=cov, size=n)
            
            ash = ASH(q=100)
            ash.fit(X)
            
            Y = np.zeros((300, d))
            Y[:,0] = np.linspace(-4,4,300)
            f = ash.predict(Y)
            
            plt.plot(Y[:,0], stats.multivariate_normal.pdf(Y, mean=mean, cov=cov), label='exact')
            plt.plot(Y[:,0], f, label='FastASH')
            plt.legend()
            plt.show()
        
        ## help for packaging
        
        script files are provided and based on this tutorial :
        https://levelup.gitconnected.com/how-to-deploy-a-cython-package-to-pypi-8217a6581f09
        
        
Platform: UNKNOWN
Description-Content-Type: text/markdown
