Heiko Strathmann karlnapf

## Spectrum kernel with Shogun.ipynb

      
              1 file
            
          
              0 forks
            
          
              0 comments
            
          
              1 star
            
          
                karlnapf
                / Spectrum kernel with Shogun.ipynb
            
            
              Created
              September 26, 2013 22:01
            
          
      Sorry, something went wrong. Reload?
      Sorry, we cannot display this file.
      Sorry, this file is invalid so it cannot be displayed.
      
          Viewer requires iframe.
      
    
## gamma_median_heuristic.py
def gamma_median_heuristic(Z, num_subsample=1000):
    """
    Computes the median pairwise distance in a random sub-sample of Z.
    Returns a \gamma for k(x,y)=\exp(-\gamma ||x-y||^2), according to the median heuristc,
    i.e. it corresponds to \sigma in k(x,y)=\exp(-0.5*||x-y||^2 / \sigma^2) where
    \sigma is the median distance. \gamma = 0.5/(\sigma^2)
    """
    inds = np.random.permutation(len(Z))[:np.max([num_subsample, len(Z)])]
    dists = squareform(pdist(Z[inds], 'sqeuclidean'))
    median_dist = np.median(dists[dists > 0])

## gist:b24fa3d69aafaa282806
<!DOCTYPE html>
<html>
<head>
<title>Shogun 4.0</title>
</head>

<body>

<p>
The Shogun team just released <a href="http://www.shogun-toolbox.org/new/41" target="_blank">version 4.0</a> of their community driven Machine Learning toolbox.

## large gp regression benchmark.ipynb

      
              1 file
            
          
              0 forks
            
          
              0 comments
            
          
              0 stars
            
          
                karlnapf
                / large gp regression benchmark.ipynb
            
            
              Created
              February 15, 2017 18:09
            
          
      Sorry, something went wrong. Reload?
      Sorry, we cannot display this file.
      Sorry, this file is invalid so it cannot be displayed.
      
          Viewer requires iframe.
      
    
## krr_add_rows.ipynb

      
              1 file
            
          
              0 forks
            
          
              0 comments
            
          
              0 stars
            
          
                karlnapf
                / krr_add_rows.ipynb
            
            
              Created
              February 7, 2017 18:38
            
          
      Sorry, something went wrong. Reload?
      Sorry, we cannot display this file.
      Sorry, this file is invalid so it cannot be displayed.
      
          Viewer requires iframe.
      
    
## krr_add_rows.ipynb

      
              1 file
            
          
              0 forks
            
          
              0 comments
            
          
              0 stars
            
          
                karlnapf
                / krr_add_rows.ipynb
            
            
              Created
              January 31, 2017 05:48
            
          
      Sorry, something went wrong. Reload?
      Sorry, we cannot display this file.
      Sorry, this file is invalid so it cannot be displayed.
      
          Viewer requires iframe.
      
    
## kernel_exp_family_shogun.ipynb

      
              1 file
            
          
              0 forks
            
          
              0 comments
            
          
              0 stars
            
          
                karlnapf
                / kernel_exp_family_shogun.ipynb
            
            
              Created
              September 28, 2016 13:28
            
          
      Sorry, something went wrong. Reload?
      Sorry, we cannot display this file.
      Sorry, this file is invalid so it cannot be displayed.
      
          Viewer requires iframe.
      
    
## sinusoidal_perturbation.ipynb

      
              1 file
            
          
              0 forks
            
          
              0 comments
            
          
              0 stars
            
          
                karlnapf
                / sinusoidal_perturbation.ipynb
            
            
              Created
              July 27, 2016 17:12
            
          
      Sorry, something went wrong. Reload?
      Sorry, we cannot display this file.
      Sorry, this file is invalid so it cannot be displayed.
      
          Viewer requires iframe.
      
    
## errorbars.ipynb

      
              1 file
            
          
              0 forks
            
          
              0 comments
            
          
              0 stars
            
          
                karlnapf
                / errorbars.ipynb
            
            
              Created
              July 27, 2016 12:36
            
          
      Sorry, something went wrong. Reload?
      Sorry, we cannot display this file.
      Sorry, this file is invalid so it cannot be displayed.
      
          Viewer requires iframe.
      
    
## kernel_exp_family_shogun.ipynb

      
              1 file
            
          
              0 forks
            
          
              0 comments
            
          
              0 stars
            
          
                karlnapf
                / kernel_exp_family_shogun.ipynb
            
            
              Created
              June 30, 2016 13:08
            
          
      Sorry, something went wrong. Reload?
      Sorry, we cannot display this file.
      Sorry, this file is invalid so it cannot be displayed.
      
          Viewer requires iframe.
	def gamma_median_heuristic(Z, num_subsample=1000):
	"""
	Computes the median pairwise distance in a random sub-sample of Z.
	Returns a \gamma for k(x,y)=\exp(-\gamma \|\|x-y\|\|^2), according to the median heuristc,
	i.e. it corresponds to \sigma in k(x,y)=\exp(-0.5*\|\|x-y\|\|^2 / \sigma^2) where
	\sigma is the median distance. \gamma = 0.5/(\sigma^2)
	"""
	inds = np.random.permutation(len(Z))[:np.max([num_subsample, len(Z)])]
	dists = squareform(pdist(Z[inds], 'sqeuclidean'))
	median_dist = np.median(dists[dists > 0])
	<!DOCTYPE html>
	<html>
	<head>
	<title>Shogun 4.0</title>
	</head>

	<body>

	<p>
	The Shogun team just released <a href="http://www.shogun-toolbox.org/new/41" target="_blank">version 4.0</a> of their community driven Machine Learning toolbox.