update to 1.1.0

README.md

@@ -36,7 +36,7 @@ All included operations work on varying data types, are implemented both for CPU
 
 ## Installation
 
-Ensure that at least PyTorch 1.0.0 is installed and verify that `cuda/bin` and `cuda/include` are in your `$PATH` and `$CPATH` respectively, *e.g.*:
+Ensure that at least PyTorch 1.1.0 is installed and verify that `cuda/bin` and `cuda/include` are in your `$PATH` and `$CPATH` respectively, *e.g.*:
 
 ```
 $ python -c "import torch; print(torch.__version__)"

cpu/scatter.cpp

@@ -5,7 +5,7 @@
 void scatter_mul(at::Tensor src, at::Tensor index, at::Tensor out,
                  int64_t dim) {
   int64_t elems_per_row = index.size(dim), i, idx;
-  AT_DISPATCH_ALL_TYPES(src.type(), "scatter_mul", [&] {
+  AT_DISPATCH_ALL_TYPES(src.scalar_type(), "scatter_mul", [&] {
     DIM_APPLY3(scalar_t, src, int64_t, index, scalar_t, out, dim, {
       for (i = 0; i < elems_per_row; i++) {
         idx = index_data[i * index_stride];
@@ -18,7 +18,7 @@ void scatter_mul(at::Tensor src, at::Tensor index, at::Tensor out,
 void scatter_div(at::Tensor src, at::Tensor index, at::Tensor out,
                  int64_t dim) {
   int64_t elems_per_row = index.size(dim), i, idx;
-  AT_DISPATCH_ALL_TYPES(src.type(), "scatter_div", [&] {
+  AT_DISPATCH_ALL_TYPES(src.scalar_type(), "scatter_div", [&] {
     DIM_APPLY3(scalar_t, src, int64_t, index, scalar_t, out, dim, {
       for (i = 0; i < elems_per_row; i++) {
         idx = index_data[i * index_stride];
@@ -31,7 +31,7 @@ void scatter_div(at::Tensor src, at::Tensor index, at::Tensor out,
 void scatter_max(at::Tensor src, at::Tensor index, at::Tensor out,
                  at::Tensor arg, int64_t dim) {
   int64_t elems_per_row = index.size(dim), i, idx;
-  AT_DISPATCH_ALL_TYPES(src.type(), "scatter_max", [&] {
+  AT_DISPATCH_ALL_TYPES(src.scalar_type(), "scatter_max", [&] {
     DIM_APPLY4(scalar_t, src, int64_t, index, scalar_t, out, int64_t, arg, dim,
                {
                  for (i = 0; i < elems_per_row; i++) {
@@ -48,7 +48,7 @@ void scatter_max(at::Tensor src, at::Tensor index, at::Tensor out,
 void scatter_min(at::Tensor src, at::Tensor index, at::Tensor out,
                  at::Tensor arg, int64_t dim) {
   int64_t elems_per_row = index.size(dim), i, idx;
-  AT_DISPATCH_ALL_TYPES(src.type(), "scatter_min", [&] {
+  AT_DISPATCH_ALL_TYPES(src.scalar_type(), "scatter_min", [&] {
     DIM_APPLY4(scalar_t, src, int64_t, index, scalar_t, out, int64_t, arg, dim,
                {
                  for (i = 0; i < elems_per_row; i++) {
@@ -65,7 +65,7 @@ void scatter_min(at::Tensor src, at::Tensor index, at::Tensor out,
 void index_backward(at::Tensor grad, at::Tensor index, at::Tensor arg,
                     at::Tensor out, int64_t dim) {
   int64_t elems_per_row = index.size(dim), i, idx;
-  AT_DISPATCH_ALL_TYPES(grad.type(), "index_backward", [&] {
+  AT_DISPATCH_ALL_TYPES(grad.scalar_type(), "index_backward", [&] {
     DIM_APPLY4(scalar_t, grad, int64_t, index, int64_t, arg, scalar_t, out, dim,
                {
                  for (i = 0; i < elems_per_row; i++) {

cuda/scatter_kernel.cu

@@ -44,7 +44,7 @@ scatter_mul_kernel(at::cuda::detail::TensorInfo<scalar_t, int64_t> src,
 void scatter_mul_cuda(at::Tensor src, at::Tensor index, at::Tensor out,
                       int64_t dim) {
   cudaSetDevice(src.get_device());
-  AT_DISPATCH_ALL_TYPES(src.type(), "scatter_mul_kernel", [&] {
+  AT_DISPATCH_ALL_TYPES(src.scalar_type(), "scatter_mul_kernel", [&] {
     KERNEL_RUN(scatter_mul_kernel, index.dim(), index.numel(),
                at::cuda::detail::getTensorInfo<scalar_t, int64_t>(src),
                at::cuda::detail::getTensorInfo<int64_t, int64_t>(index),
@@ -71,7 +71,7 @@ scatter_div_kernel(at::cuda::detail::TensorInfo<scalar_t, int64_t> src,
 void scatter_div_cuda(at::Tensor src, at::Tensor index, at::Tensor out,
                       int64_t dim) {
   cudaSetDevice(src.get_device());
-  AT_DISPATCH_ALL_TYPES(src.type(), "scatter_div_kernel", [&] {
+  AT_DISPATCH_ALL_TYPES(src.scalar_type(), "scatter_div_kernel", [&] {
     KERNEL_RUN(scatter_div_kernel, index.dim(), index.numel(),
                at::cuda::detail::getTensorInfo<scalar_t, int64_t>(src),
                at::cuda::detail::getTensorInfo<int64_t, int64_t>(index),
@@ -117,7 +117,7 @@ scatter_max_kernel(at::cuda::detail::TensorInfo<scalar_t, int64_t> src,
 void scatter_max_cuda(at::Tensor src, at::Tensor index, at::Tensor out,
                       at::Tensor arg, int64_t dim) {
   cudaSetDevice(src.get_device());
-  AT_DISPATCH_ALL_TYPES(src.type(), "scatter_max_kernel", [&] {
+  AT_DISPATCH_ALL_TYPES(src.scalar_type(), "scatter_max_kernel", [&] {
     auto src_info = at::cuda::detail::getTensorInfo<scalar_t, int64_t>(src);
     auto index_info = at::cuda::detail::getTensorInfo<int64_t, int64_t>(index);
     auto out_info = at::cuda::detail::getTensorInfo<scalar_t, int64_t>(out);
@@ -148,7 +148,7 @@ scatter_min_kernel(at::cuda::detail::TensorInfo<scalar_t, int64_t> src,
 void scatter_min_cuda(at::Tensor src, at::Tensor index, at::Tensor out,
                       at::Tensor arg, int64_t dim) {
   cudaSetDevice(src.get_device());
-  AT_DISPATCH_ALL_TYPES(src.type(), "scatter_min_kernel", [&] {
+  AT_DISPATCH_ALL_TYPES(src.scalar_type(), "scatter_min_kernel", [&] {
     auto src_info = at::cuda::detail::getTensorInfo<scalar_t, int64_t>(src);
     auto index_info = at::cuda::detail::getTensorInfo<int64_t, int64_t>(index);
     auto out_info = at::cuda::detail::getTensorInfo<scalar_t, int64_t>(out);
@@ -184,7 +184,7 @@ index_backward_kernel(at::cuda::detail::TensorInfo<scalar_t, int64_t> grad,
 void index_backward_cuda(at::Tensor grad, at::Tensor index, at::Tensor arg,
                          at::Tensor out, int64_t dim) {
   cudaSetDevice(grad.get_device());
-  AT_DISPATCH_ALL_TYPES(grad.type(), "index_backward_kernel", [&] {
+  AT_DISPATCH_ALL_TYPES(grad.scalar_type(), "index_backward_kernel", [&] {
     KERNEL_RUN(index_backward_kernel, index.dim(), index.numel(),
                at::cuda::detail::getTensorInfo<scalar_t, int64_t>(grad),
                at::cuda::detail::getTensorInfo<int64_t, int64_t>(index),

setup.py

@@ -20,7 +20,7 @@ if CUDA_HOME is not None:
                       ['cuda/scatter.cpp', 'cuda/scatter_kernel.cu'])
     ]
 
-__version__ = '1.1.2'
+__version__ = '1.2.0'
 url = 'https://github.com/rusty1s/pytorch_scatter'
 
 install_requires = []

torch_scatter/__init__.py

@@ -7,7 +7,7 @@ from .std import scatter_std
 from .max import scatter_max
 from .min import scatter_min
 
-__version__ = '1.1.2'
+__version__ = '1.2.0'
 
 __all__ = [
     'scatter_add',