fix : v0.3.2

Context

• bump version in version.txt
• fix : package was not tested with wheels on pypi but on local artefacts. This led to

1. Not being sure if the package was going to work with pypi pkg
2. Extremely long release job time

include/aidge/backend/cpu/operator/ConvDepthWiseImpl_kernels.hpp

@@ -133,15 +133,15 @@ void ConvDepthWiseImpl2D_cpu_forward_kernel(const std::array<DimSize_t, 2>& stri
 
 
     // output H size
+    const DimSize_t dilated_kernel_x = dilationDims[0]*(kernelDims[0] - 1) + 1;
     const std::size_t oxSize =
-            static_cast<std::size_t>(std::floor(static_cast<float>(inputDims[2] - kernelDims[0] + strideDims[0]) /
+            static_cast<std::size_t>(std::floor(static_cast<float>(inputDims[2] - dilated_kernel_x - 1 + strideDims[0]) /
                                 static_cast<float>(strideDims[0])));
-    const DimSize_t dilated_kernel_x = dilationDims[0]*(kernelDims[0] - 1) + 1;
     // output W size
+    const DimSize_t dilated_kernel_y = dilationDims[1]*(kernelDims[1] - 1) + 1;
     const std::size_t oySize =
-            static_cast<std::size_t>(std::floor(static_cast<float>(inputDims[3] - kernelDims[1] + strideDims[1]) /
+            static_cast<std::size_t>(std::floor(static_cast<float>(inputDims[3] - dilated_kernel_y - 1 + strideDims[1]) /
                                 static_cast<float>(strideDims[1])));
-    const DimSize_t dilated_kernel_y = dilationDims[1]*(kernelDims[1] - 1) + 1;
 
     // TODO: kernel computation
     // output (batch, outCh, Xout, Yout)