Merge branch 'features/240-flipud' of github.com:helmholtz-analytics/…

…heat into features/240-flipud

CHANGELOG.md

@@ -3,9 +3,10 @@
 - [#429](https://github.com/helmholtz-analytics/heat/pull/429) Implementated QR
 - [#429](https://github.com/helmholtz-analytics/heat/pull/429) Added PyTorch Jitter to inner function of matmul for increased speed
 - [#483](https://github.com/helmholtz-analytics/heat/pull/483) Bugfix: Underlying torch tensor moves to the right device on array initialisation
-- [#483](https://github.com/helmholtz-analytics/heat/pull/483) Bugfix:DNDarray.cpu() changes heat device to cpu
+- [#483](https://github.com/helmholtz-analytics/heat/pull/483) Bugfix: DNDarray.cpu() changes heat device to cpu
 - Update documentation theme to "Read the Docs"
 - [#496](https://github.com/helmholtz-analytics/heat/pull/496) New feature: flipud()
+- [#499](https://github.com/helmholtz-analytics/heat/pull/499) Bugfix: MPI datatype mapping: `torch.int16` now maps to `MPI.SHORT` instead of `MPI.SHORT_INT`
 
 # v0.3.0
 

heat/core/communication.py

@@ -51,7 +51,7 @@ class MPICommunication(Communication):
         torch.bool: MPI.BOOL,
         torch.uint8: MPI.UNSIGNED_CHAR,
         torch.int8: MPI.SIGNED_CHAR,
-        torch.int16: MPI.SHORT_INT,
+        torch.int16: MPI.SHORT,
         torch.int32: MPI.INT,
         torch.int64: MPI.LONG,
         torch.float32: MPI.FLOAT,