From 96d529054dab28c23f3fc1fda94b43b9b051c6da Mon Sep 17 00:00:00 2001
From: Philipp Arras <parras@mpa-garching.mpg.de>
Date: Tue, 21 Jan 2020 20:56:26 +0100
Subject: [PATCH] Formatting

---
 nifty6/minimization/metric_gaussian_kl.py | 4 ++--
 nifty6/operators/sampling_enabler.py      | 2 +-
 2 files changed, 3 insertions(+), 3 deletions(-)

diff --git a/nifty6/minimization/metric_gaussian_kl.py b/nifty6/minimization/metric_gaussian_kl.py
index e665efd2c..ca09b9747 100644
--- a/nifty6/minimization/metric_gaussian_kl.py
+++ b/nifty6/minimization/metric_gaussian_kl.py
@@ -79,7 +79,7 @@ class MetricGaussianKL(Energy):
 
     def __init__(self, mean, hamiltonian, n_samples, constants=[],
                  point_estimates=[], mirror_samples=False,
-                 napprox=0, _samples=None, lh_sampling_dtype = np.float):
+                 napprox=0, _samples=None, lh_sampling_dtype=np.float64):
         super(MetricGaussianKL, self).__init__(mean)
 
         if not isinstance(hamiltonian, StandardHamiltonian):
@@ -101,7 +101,7 @@ class MetricGaussianKL(Energy):
             if napprox > 1:
                 met._approximation = makeOp(approximation2endo(met, napprox))
             _samples = tuple(met.draw_sample(from_inverse=True,
-                                             dtype = lh_sampling_dtype)
+                                             dtype=lh_sampling_dtype)
                              for _ in range(n_samples))
             if mirror_samples:
                 _samples += tuple(-s for s in _samples)
diff --git a/nifty6/operators/sampling_enabler.py b/nifty6/operators/sampling_enabler.py
index 3797f1119..e78846d94 100644
--- a/nifty6/operators/sampling_enabler.py
+++ b/nifty6/operators/sampling_enabler.py
@@ -71,7 +71,7 @@ class SamplingEnabler(EndomorphicOperator):
             else:
                 s = self._prior.draw_sample(from_inverse=True)
                 sp = self._prior(s)
-                nj = self._likelihood.draw_sample(dtype = dtype)
+                nj = self._likelihood.draw_sample(dtype=dtype)
                 energy = QuadraticEnergy(s, self._op, sp + nj,
                                          _grad=self._likelihood(s) - nj)
             inverter = ConjugateGradient(self._ic)
-- 
GitLab