diff --git a/CHANGELOG.md b/CHANGELOG.md
index 2d9091b..1a618dc 100644
--- a/CHANGELOG.md
+++ b/CHANGELOG.md
@@ -6,6 +6,11 @@ and this project adheres to [Semantic Versioning](https://semver.org/spec/v2.0.0
 
 ## [Unreleased]
 ### Added
+### Fixed
+### Removed
+
+## [0.7.2] - 2019-07-30
+### Added
 - Support for multi-gpu execution. Use `parallel=True` when
   constructing a `MSDRegressionModel` or `MSDSegmentationModel`.
 ### Fixed
@@ -26,5 +31,6 @@ and this project adheres to [Semantic Versioning](https://semver.org/spec/v2.0.0
 ### Added
 - Initial release.
 
-[Unreleased]: https://www.github.com/ahendriksen/msd_pytorch/compare/v0.6.2...master
+[Unreleased]: https://www.github.com/ahendriksen/msd_pytorch/compare/v0.7.2...master
+[0.7.2]: https://www.github.com/ahendriksen/msd_pytorch/compare/v0.7.1...v0.7.2
 [0.7.1]: https://www.github.com/ahendriksen/msd_pytorch/compare/v0.6.2...v0.7.1
diff --git a/conda/build.sh b/conda/build.sh
index 3ef8f8c..be3d403 100755
--- a/conda/build.sh
+++ b/conda/build.sh
@@ -35,7 +35,11 @@ export CUDA_HOME=$TMP_NVCC_LOCATION/cuda-$cudatoolkit
 if [ -f $CUDA_HOME/bin/nvcc ] ; then
     echo "-- CUDA already installed";
 else
-    bash $CUDA_INSTALLER --silent --toolkit --toolkitpath=$CUDA_HOME --tmpdir=$TMP_NVCC_LOCATION --override
+    echo "Installing version ${cudatoolkit} "
+    echo "  with ${CUDA_INSTALLER}"
+    echo "  into ${CUDA_HOME}"
+    echo "  with temp space ${TMP_NVCC_LOCATION}"
+    sh $CUDA_INSTALLER --silent --toolkit --toolkitpath=$CUDA_HOME --tmpdir=$TMP_NVCC_LOCATION --override
 fi
 
 echo "-- Setting CUDA_HOME: $CUDA_HOME"
diff --git a/msd_pytorch/VERSION b/msd_pytorch/VERSION
index 7deb86f..d5cc44d 100644
--- a/msd_pytorch/VERSION
+++ b/msd_pytorch/VERSION
@@ -1 +1 @@
-0.7.1
\ No newline at end of file
+0.7.2
\ No newline at end of file