pulled out check_arguments and placed it in gpml_extensions repository

mgp.m

@@ -348,89 +348,4 @@
 
   result = sum(B .* A')';
 
-end
-
-% performs argument checks/transformations similar to those found in
-% gp.m from GPML but guaranteed to be compatible with the MGP
-function [theta, inference_method, mean_function, covariance_function, ...
-          likelihood] = ...
-  check_arguments(theta, inference_method, mean_function, covariance_function, ...
-                  likelihood, x)
-
-  % default to exact inference
-  if (isempty(inference_method))
-    inference_method = {@exact_inference};
-  end
-
-  % default to zero mean function
-  if (isempty(mean_function))
-    mean_function = {@zero_mean};
-  end
-
-  % no default covariance function
-  if (isempty(covariance_function))
-    error('mgp:missing_argument', ...
-          'covariance function must be defined!');
-  end
-
-  % default to Gaussian likelihood
-  if (isempty(likelihood))
-    likelihood = {@likGauss};
-  end
-
-  % allow string/function handle input; convert to cell arrays if
-  % necessary
-  if (ischar(inference_method) || ...
-      isa(inference_method, 'function_handle'))
-    inference_method = {inference_method};
-  end
-
-  if (ischar(mean_function) || ...
-      isa(mean_function, 'function_handle'))
-    mean_function = {mean_function};
-  end
-
-  if (ischar(covariance_function) || ...
-      isa(covariance_function, 'function_handle'))
-    covariance_function = {covariance_function};
-  end
-
-  if (ischar(likelihood) || ...
-      isa(likelihood, 'function_handle'))
-    likelihood = {likelihood};
-  end
-
-  % ensure all hyperparameter fields exist
-  for field = {'cov', 'lik', 'mean'}
-    if (~isfield(theta, field{:}))
-      theta.(field{:}) = [];
-    end
-  end
-
-  % check dimension of hyperparameter fields
-  D = size(x, 2);
-
-  expression = feval(mean_function{:});
-  if (numel(theta.mean) ~= eval(expression))
-    error('mgp:incorrect_specification', ...
-          'wrong number of mean hyperparameters! (%i given, %s expected)', ...
-          numel(theta.mean), ...
-          expression);
-  end
-
-  expression = feval(covariance_function{:});
-  if (numel(theta.cov) ~= eval(expression))
-    error('mgp:incorrect_specification', ...
-          'wrong number of covariance hyperparameters! (%i given, %s expected)', ...
-          numel(theta.cov), ...
-          expression);
-  end
-
-  expression = feval(likelihood{:});
-  if (numel(theta.lik) ~= eval(expression))
-    error('mgp:incorrect_specification', ...
-          'wrong number of likelihood hyperparameters! (%i given, 1 expected)', ...
-          numel(theta.lik));
-  end
-
 end