diff options
24 files changed, 477 insertions, 60 deletions
diff --git a/.gitmodules b/.gitmodules index 1b17ea57b9..0edca21239 100644 --- a/.gitmodules +++ b/.gitmodules @@ -1,3 +1,3 @@ [submodule "google/protobuf"] path = google/protobuf - url = https://github.com/google/protobuf.git + url = https://github.googlesource.com/google/protobuf.git @@ -21,8 +21,8 @@ new_http_archive( new_http_archive( name = "eigen_archive", - url = "https://bitbucket.org/eigen/eigen/get/8cd7c2c.tar.gz", - sha256 = "30b77010c49a28875c76f5941cab06d0e15c52dc193be9729def53b6ea1fdb57", + url = "https://bitbucket.org/eigen/eigen/get/726c779.tar.gz", + sha256 = "30e0c5d84cfefc6a0bf7ae1e682b22788b5b2e408e7db7d9ea2d2aa9f70a72a9", build_file = "eigen.BUILD", ) diff --git a/eigen.BUILD b/eigen.BUILD index bb78603b7a..084689c6f4 100644 --- a/eigen.BUILD +++ b/eigen.BUILD @@ -1,6 +1,6 @@ package(default_visibility = ["//visibility:public"]) -archive_dir = "eigen-eigen-8cd7c2c6e9e1" +archive_dir = "eigen-eigen-726c779797e8" cc_library( name = "eigen", diff --git a/tensorflow/core/framework/tensor_shape.cc b/tensorflow/core/framework/tensor_shape.cc index 697e170729..a6d540bbe1 100644 --- a/tensorflow/core/framework/tensor_shape.cc +++ b/tensorflow/core/framework/tensor_shape.cc @@ -143,7 +143,7 @@ void TensorShape::AddDim(int64 size) { // to allow REP32. bool can_be_rep32 = (vals.size() <= 3); if (can_be_rep32) { - for (int i = 0; i < vals.size(); i++) { + for (size_t i = 0; i < vals.size(); i++) { if (vals[i] >= kMaxRep32) { can_be_rep32 = false; break; @@ -152,7 +152,7 @@ void TensorShape::AddDim(int64 size) { } if (can_be_rep32) { set_tag(REP32); - for (int d = 0; d < vals.size(); d++) { + for (size_t d = 0; d < vals.size(); d++) { as32()->dims_[d] = static_cast<int32>(vals[d]); } } else { diff --git a/tensorflow/core/graph/equal_graph_def_test.cc b/tensorflow/core/graph/equal_graph_def_test.cc index e7cbb9853f..77ff0d9bce 100644 --- a/tensorflow/core/graph/equal_graph_def_test.cc +++ b/tensorflow/core/graph/equal_graph_def_test.cc @@ -27,7 +27,7 @@ namespace { REGISTER_OP("Input").Output("o: float"); REGISTER_OP("Alternate").Output("o: float"); -REGISTER_OP("Cross").Input("a: float").Input("b: float").Output("o: float"); +REGISTER_OP("Combine").Input("a: float").Input("b: float").Output("o: float"); Node* Input(const GraphDefBuilder::Options& opts) { return ops::SourceOp("Input", opts); @@ -37,9 +37,9 @@ Node* Alternate(const GraphDefBuilder::Options& opts) { return ops::SourceOp("Alternate", opts); } -Node* Cross(ops::NodeOut a, ops::NodeOut b, - const GraphDefBuilder::Options& opts) { - return ops::BinaryOp("Cross", a, b, opts); +Node* Combine(ops::NodeOut a, ops::NodeOut b, + const GraphDefBuilder::Options& opts) { + return ops::BinaryOp("Combine", a, b, opts); } class EqualGraphDefTest : public ::testing::Test { @@ -99,11 +99,11 @@ TEST_F(EqualGraphDefTest, ExtraNode) { TEST_F(EqualGraphDefTest, NodeOrder) { Node* a = Input(e_.opts().WithName("A")); Node* b = Input(e_.opts().WithName("B")); - Cross(a, b, e_.opts().WithName("C")); + Combine(a, b, e_.opts().WithName("C")); b = Input(a_.opts().WithName("B")); a = Input(a_.opts().WithName("A")); - Cross(a, b, a_.opts().WithName("C")); + Combine(a, b, a_.opts().WithName("C")); EXPECT_TRUE(Match()) << diff_; } @@ -141,11 +141,11 @@ TEST_F(EqualGraphDefTest, DeviceMismatch) { TEST_F(EqualGraphDefTest, InputMismatch) { Node* a = Input(e_.opts().WithName("A")); Node* b = Input(e_.opts().WithName("B")); - Cross(a, a, e_.opts().WithName("C")); + Combine(a, a, e_.opts().WithName("C")); a = Input(a_.opts().WithName("A")); b = Input(a_.opts().WithName("B")); - Cross(b, b, a_.opts().WithName("C")); + Combine(b, b, a_.opts().WithName("C")); EXPECT_FALSE(Match()); EXPECT_EQ("Node named 'C' has input 0 'B' that doesn't match expected 'A'", diff_); @@ -154,11 +154,11 @@ TEST_F(EqualGraphDefTest, InputMismatch) { TEST_F(EqualGraphDefTest, InputOrderMismatch) { Node* a = Input(e_.opts().WithName("A")); Node* b = Input(e_.opts().WithName("B")); - Cross(a, b, e_.opts().WithName("C")); + Combine(a, b, e_.opts().WithName("C")); a = Input(a_.opts().WithName("A")); b = Input(a_.opts().WithName("B")); - Cross(b, a, a_.opts().WithName("C")); + Combine(b, a, a_.opts().WithName("C")); EXPECT_FALSE(Match()); EXPECT_EQ("Node named 'C' has input 0 'B' that doesn't match expected 'A'", diff_); @@ -169,21 +169,21 @@ TEST_F(EqualGraphDefTest, ControlInputOrder) { Node* b = Input(e_.opts().WithName("B")); Node* c = Input(e_.opts().WithName("C")); Node* d = Input(e_.opts().WithName("D")); - Cross(a, a, e_.opts() - .WithName("E") - .WithControlInput(b) - .WithControlInput(c) - .WithControlInput(d)); + Combine(a, a, e_.opts() + .WithName("E") + .WithControlInput(b) + .WithControlInput(c) + .WithControlInput(d)); a = Input(a_.opts().WithName("A")); b = Input(a_.opts().WithName("B")); c = Input(a_.opts().WithName("C")); d = Input(a_.opts().WithName("D")); - Cross(a, a, a_.opts() - .WithName("E") - .WithControlInput(c) - .WithControlInput(d) - .WithControlInput(b)); + Combine(a, a, a_.opts() + .WithName("E") + .WithControlInput(c) + .WithControlInput(d) + .WithControlInput(b)); EXPECT_TRUE(Match()) << diff_; } @@ -192,13 +192,15 @@ TEST_F(EqualGraphDefTest, ControlInputMismatch) { Node* b = Input(e_.opts().WithName("B")); Node* c = Input(e_.opts().WithName("C")); Node* d = Input(e_.opts().WithName("D")); - Cross(a, a, e_.opts().WithName("E").WithControlInput(b).WithControlInput(c)); + Combine(a, a, + e_.opts().WithName("E").WithControlInput(b).WithControlInput(c)); a = Input(a_.opts().WithName("A")); b = Input(a_.opts().WithName("B")); c = Input(a_.opts().WithName("C")); d = Input(a_.opts().WithName("D")); - Cross(a, a, a_.opts().WithName("E").WithControlInput(b).WithControlInput(d)); + Combine(a, a, + a_.opts().WithName("E").WithControlInput(b).WithControlInput(d)); EXPECT_FALSE(Match()); EXPECT_EQ("Node named 'E' missing expected control input '^C'", diff_); } @@ -207,12 +209,13 @@ TEST_F(EqualGraphDefTest, ControlInputAdded) { Node* a = Input(e_.opts().WithName("A")); Node* b = Input(e_.opts().WithName("B")); Node* c = Input(e_.opts().WithName("C")); - Cross(a, a, e_.opts().WithName("D").WithControlInput(b)); + Combine(a, a, e_.opts().WithName("D").WithControlInput(b)); a = Input(a_.opts().WithName("A")); b = Input(a_.opts().WithName("B")); c = Input(a_.opts().WithName("C")); - Cross(a, a, a_.opts().WithName("D").WithControlInput(b).WithControlInput(c)); + Combine(a, a, + a_.opts().WithName("D").WithControlInput(b).WithControlInput(c)); EXPECT_FALSE(Match()); EXPECT_EQ( "Node named 'D' has inputs 'A, A, ^B, ^C' that don't match " @@ -224,12 +227,13 @@ TEST_F(EqualGraphDefTest, ControlInputRemoved) { Node* a = Input(e_.opts().WithName("A")); Node* b = Input(e_.opts().WithName("B")); Node* c = Input(e_.opts().WithName("C")); - Cross(a, a, e_.opts().WithName("D").WithControlInput(b).WithControlInput(c)); + Combine(a, a, + e_.opts().WithName("D").WithControlInput(b).WithControlInput(c)); a = Input(a_.opts().WithName("A")); b = Input(a_.opts().WithName("B")); c = Input(a_.opts().WithName("C")); - Cross(a, a, a_.opts().WithName("D").WithControlInput(b)); + Combine(a, a, a_.opts().WithName("D").WithControlInput(b)); EXPECT_FALSE(Match()); EXPECT_EQ( "Node named 'D' has inputs 'A, A, ^B' that don't match " diff --git a/tensorflow/core/graph/graph_partition_test.cc b/tensorflow/core/graph/graph_partition_test.cc index 4bbfc5fe5c..ecdf9ed74f 100644 --- a/tensorflow/core/graph/graph_partition_test.cc +++ b/tensorflow/core/graph/graph_partition_test.cc @@ -131,7 +131,7 @@ void CheckLoopConstruction(const GraphDef& graph_def) { REGISTER_OP("Input").Output("o: float"); REGISTER_OP("BoolInput").Output("o: bool"); -REGISTER_OP("Cross").Input("a: float").Input("b: float").Output("o: float"); +REGISTER_OP("Combine").Input("a: float").Input("b: float").Output("o: float"); Node* Input(const GraphDefBuilder::Options& opts) { return ops::SourceOp("Input", opts); @@ -141,9 +141,9 @@ Node* BoolInput(const GraphDefBuilder::Options& opts) { return ops::SourceOp("BoolInput", opts); } -Node* Cross(ops::NodeOut a, ops::NodeOut b, - const GraphDefBuilder::Options& opts) { - return ops::BinaryOp("Cross", a, b, opts); +Node* Combine(ops::NodeOut a, ops::NodeOut b, + const GraphDefBuilder::Options& opts) { + return ops::BinaryOp("Combine", a, b, opts); } class GraphPartitionTest : public ::testing::Test { @@ -188,13 +188,13 @@ class GraphPartitionTest : public ::testing::Test { TEST_F(GraphPartitionTest, SingleDevice) { using namespace ::tensorflow::ops; // NOLINT(build/namespaces) Node* a1 = Input(in_.opts().WithName("A1")); - Cross(a1, a1, in_.opts().WithName("A2")); + Combine(a1, a1, in_.opts().WithName("A2")); Partition(ToGraphDef(), &partitions_); EXPECT_EQ(1, partitions_.size()); a1 = Input(a_opts_.WithName("A1")); - Cross(a1, a1, a_opts_.WithName("A2")); + Combine(a1, a1, a_opts_.WithName("A2")); ExpectMatchA(); } @@ -202,7 +202,7 @@ TEST_F(GraphPartitionTest, CrossDeviceData) { using namespace ::tensorflow::ops; // NOLINT(build/namespaces) Node* a1 = Input(in_.opts().WithName("A1")); Node* b1 = Input(in_.opts().WithName("B1")); - Cross(a1, b1, in_.opts().WithName("B2")); + Combine(a1, b1, in_.opts().WithName("B2")); Partition(ToGraphDef(), &partitions_); EXPECT_EQ(2, partitions_.size()); @@ -216,7 +216,7 @@ TEST_F(GraphPartitionTest, CrossDeviceData) { b1 = Input(b_opts_.WithName("B1")); Node* recv = _Recv(DT_FLOAT, "edge_1_A1", a, 82, b, b_opts_.WithName("A1/_1")); - Cross(recv, b1, b_opts_.WithName("B2")); + Combine(recv, b1, b_opts_.WithName("B2")); ExpectMatchB(); } @@ -224,7 +224,7 @@ TEST_F(GraphPartitionTest, CrossDeviceControl) { using namespace ::tensorflow::ops; // NOLINT(build/namespaces) Node* a1 = Input(in_.opts().WithName("A1")); Node* b1 = Input(in_.opts().WithName("B1")); - Cross(b1, b1, in_.opts().WithName("B2").WithControlInput(a1)); + Combine(b1, b1, in_.opts().WithName("B2").WithControlInput(a1)); Partition(ToGraphDef(), &partitions_); EXPECT_EQ(2, partitions_.size()); @@ -240,7 +240,7 @@ TEST_F(GraphPartitionTest, CrossDeviceControl) { _Recv(DT_FLOAT, "edge_3_A1", a, 82, b, b_opts_.WithName("A1/_2")); Node* id = Identity(recv, b_opts_.WithName("A1/_3")); b1 = Input(b_opts_.WithName("B1")); - Cross(b1, b1, b_opts_.WithName("B2").WithControlInput(id)); + Combine(b1, b1, b_opts_.WithName("B2").WithControlInput(id)); ExpectMatchB(); } @@ -248,8 +248,8 @@ TEST_F(GraphPartitionTest, CrossDeviceData_MultiUse) { using namespace ::tensorflow::ops; // NOLINT(build/namespaces) Node* a1 = Input(in_.opts().WithName("A1")); Node* b1 = Input(in_.opts().WithName("B1")); - Cross(a1, b1, in_.opts().WithName("B2")); - Cross(a1, a1, in_.opts().WithName("B3")); + Combine(a1, b1, in_.opts().WithName("B2")); + Combine(a1, a1, in_.opts().WithName("B3")); Partition(ToGraphDef(), &partitions_); EXPECT_EQ(2, partitions_.size()); @@ -263,8 +263,8 @@ TEST_F(GraphPartitionTest, CrossDeviceData_MultiUse) { Node* recv = _Recv(DT_FLOAT, "edge_1_A1", a, 82, b, b_opts_.WithName("A1/_1")); b1 = Input(b_opts_.WithName("B1")); - Cross(recv, b1, b_opts_.WithName("B2")); - Cross(recv, recv, b_opts_.WithName("B3")); + Combine(recv, b1, b_opts_.WithName("B2")); + Combine(recv, recv, b_opts_.WithName("B3")); ExpectMatchB(); } @@ -272,7 +272,7 @@ TEST_F(GraphPartitionTest, CrossDeviceControl_MultiUse) { using namespace ::tensorflow::ops; // NOLINT(build/namespaces) Node* a1 = Input(in_.opts().WithName("A1")); Node* b1 = Input(in_.opts().WithName("B1")); - Cross(b1, b1, in_.opts().WithName("B2").WithControlInput(a1)); + Combine(b1, b1, in_.opts().WithName("B2").WithControlInput(a1)); Input(in_.opts().WithName("B3").WithControlInput(a1)); Partition(ToGraphDef(), &partitions_); @@ -289,7 +289,7 @@ TEST_F(GraphPartitionTest, CrossDeviceControl_MultiUse) { _Recv(DT_FLOAT, "edge_1_A1", a, 82, b, b_opts_.WithName("A1/_2")); Node* id = Identity(recv, b_opts_.WithName("A1/_3")); b1 = Input(b_opts_.WithName("B1")); - Cross(b1, b1, b_opts_.WithName("B2").WithControlInput(id)); + Combine(b1, b1, b_opts_.WithName("B2").WithControlInput(id)); Input(b_opts_.WithName("B3").WithControlInput(id)); ExpectMatchB(); } @@ -298,7 +298,7 @@ TEST_F(GraphPartitionTest, CrossDevice_DataControl) { using namespace ::tensorflow::ops; // NOLINT(build/namespaces) Node* a1 = Input(in_.opts().WithName("A1")); Node* b1 = Input(in_.opts().WithName("B1")); - Cross(a1, b1, in_.opts().WithName("B2")); + Combine(a1, b1, in_.opts().WithName("B2")); Input(in_.opts().WithName("B3").WithControlInput(a1)); Partition(ToGraphDef(), &partitions_); @@ -320,7 +320,7 @@ TEST_F(GraphPartitionTest, CrossDevice_DataControl) { Node* recv2 = _Recv(DT_FLOAT, "edge_2_A1", a, 82, b, b_opts_.WithName("A1/_5")); b1 = Input(b_opts_.WithName("B1")); - Cross(recv2, b1, b_opts_.WithName("B2")); + Combine(recv2, b1, b_opts_.WithName("B2")); Input(b_opts_.WithName("B3").WithControlInput(id1)); ExpectMatchB(); } diff --git a/tensorflow/core/kernels/cross_op.cc b/tensorflow/core/kernels/cross_op.cc new file mode 100644 index 0000000000..4564940d0f --- /dev/null +++ b/tensorflow/core/kernels/cross_op.cc @@ -0,0 +1,112 @@ +/* Copyright 2015 Google Inc. All Rights Reserved. + +Licensed under the Apache License, Version 2.0 (the "License"); +you may not use this file except in compliance with the License. +You may obtain a copy of the License at + + http://www.apache.org/licenses/LICENSE-2.0 + +Unless required by applicable law or agreed to in writing, software +distributed under the License is distributed on an "AS IS" BASIS, +WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied. +See the License for the specific language governing permissions and +limitations under the License. +==============================================================================*/ + +// See docs in ../ops/math_ops.cc. +#define EIGEN_USE_THREADS + +#include <algorithm> +#include <cmath> + +#include "third_party/eigen3/unsupported/Eigen/CXX11/Tensor" +#include "tensorflow/core/framework/op_kernel.h" +#include "tensorflow/core/framework/register_types.h" +#include "tensorflow/core/framework/tensor.h" +#include "tensorflow/core/framework/tensor_shape.h" +#include "tensorflow/core/framework/tensor_types.h" +#include "tensorflow/core/framework/types.h" +#include "tensorflow/core/kernels/cross_op.h" +#include "tensorflow/core/lib/core/status.h" +#include "tensorflow/core/platform/logging.h" +#include "tensorflow/core/platform/types.h" + +namespace tensorflow { + +typedef Eigen::ThreadPoolDevice CPUDevice; +typedef Eigen::GpuDevice GPUDevice; + +template <typename Device, typename Type> +class CrossOp : public OpKernel { + public: + explicit CrossOp(OpKernelConstruction* context) : OpKernel(context) {} + + void Compute(OpKernelContext* context) override { + const Tensor& in0 = context->input(0); + const Tensor& in1 = context->input(1); + OP_REQUIRES(context, in0.shape() == in1.shape(), + errors::InvalidArgument("Both inputs must be of same shape: ", + in0.shape().DebugString(), " vs. ", + in1.shape().DebugString())); + OP_REQUIRES(context, in0.dims() >= 1, + errors::InvalidArgument("Input must be at least 1D", + in0.shape().DebugString())); + + // Cross-products only really make sense for three and + // seven dimensions, and the latter is very obscure. If there is + // demand, we could perhaps allow 2D vectors where the last + // element is taken to be zero, but for now, we simply require + // that all are 3D. + auto inner_dim = in0.dim_size(in0.dims() - 1); + OP_REQUIRES(context, inner_dim == 3, + errors::FailedPrecondition( + "Cross-products are only defined for 3-element vectors.")); + + // Create the output Tensor with the same dimensions as the input Tensors. + Tensor* output = nullptr; + OP_REQUIRES_OK(context, context->allocate_output(0, in0.shape(), &output)); + + // Make a canonical tensor, maintaining the last (3-vector) dimension, + // while flattening all others do give the functor easy to work with data. + typename TTypes<Type, 2>::ConstTensor in0_data = + in0.flat_inner_dims<Type>(); + typename TTypes<Type, 2>::ConstTensor in1_data = + in1.flat_inner_dims<Type>(); + typename TTypes<Type, 2>::Tensor output_data = + output->flat_inner_dims<Type>(); + + functor::Cross<Device, Type>()(context->eigen_device<Device>(), in0_data, + in1_data, output_data); + } +}; + +#define REGISTER_CPU_KERNEL(type) \ + REGISTER_KERNEL_BUILDER( \ + Name("Cross").Device(DEVICE_CPU).TypeConstraint<type>("T"), \ + CrossOp<CPUDevice, type>); +TF_CALL_REAL_NUMBER_TYPES(REGISTER_CPU_KERNEL); +#undef REGISTER_CPU_KERNEL + +#if GOOGLE_CUDA +// Forward declarations of the function specializations for GPU (to prevent +// building the GPU versions here, they will be built compiling _gpu.cu.cc). +namespace functor { +#define DECLARE_GPU_KERNEL(type) \ + template <> \ + void Cross<GPUDevice, type>::operator()( \ + const GPUDevice& d, TTypes<type, 2>::ConstTensor in0_data, \ + TTypes<type, 2>::ConstTensor in1_data, \ + TTypes<type, 2>::Tensor output_data); \ + extern template struct Cross<GPUDevice, type>; +TF_CALL_REAL_NUMBER_TYPES(DECLARE_GPU_KERNEL); +#undef DECLARE_GPU_KERNEL +} // namespace functor +#define REGISTER_GPU_KERNEL(type) \ + REGISTER_KERNEL_BUILDER( \ + Name("Cross").Device(DEVICE_GPU).TypeConstraint<type>("T"), \ + CrossOp<GPUDevice, type>); +TF_CALL_REAL_NUMBER_TYPES(REGISTER_GPU_KERNEL); +#undef REGISTER_GPU_KERNEL +#endif + +} // namespace tensorflow diff --git a/tensorflow/core/kernels/cross_op.h b/tensorflow/core/kernels/cross_op.h new file mode 100644 index 0000000000..bac01aeaa7 --- /dev/null +++ b/tensorflow/core/kernels/cross_op.h @@ -0,0 +1,54 @@ +/* Copyright 2015 Google Inc. All Rights Reserved. + +Licensed under the Apache License, Version 2.0 (the "License"); +you may not use this file except in compliance with the License. +You may obtain a copy of the License at + + http://www.apache.org/licenses/LICENSE-2.0 + +Unless required by applicable law or agreed to in writing, software +distributed under the License is distributed on an "AS IS" BASIS, +WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied. +See the License for the specific language governing permissions and +limitations under the License. +==============================================================================*/ + +#ifndef TENSORFLOW_KERNELS_COLORSPACE_OP_H_ +#define TENSORFLOW_KERNELS_COLORSPACE_OP_H_ + +#include "third_party/eigen3/unsupported/Eigen/CXX11/Tensor" +#include "tensorflow/core/framework/tensor_shape.h" +#include "tensorflow/core/framework/tensor_types.h" + +namespace tensorflow { + +namespace functor { + +template <typename Device, typename Type> +struct Cross { + void operator()(const Device &d, + typename TTypes<Type, 2>::ConstTensor in0_data, + typename TTypes<Type, 2>::ConstTensor in1_data, + typename TTypes<Type, 2>::Tensor output_data) { + auto s1 = output_data.template chip<1>(0); + auto s2 = output_data.template chip<1>(1); + auto s3 = output_data.template chip<1>(2); + + auto u1 = in0_data.template chip<1>(0); + auto u2 = in0_data.template chip<1>(1); + auto u3 = in0_data.template chip<1>(2); + + auto v1 = in1_data.template chip<1>(0); + auto v2 = in1_data.template chip<1>(1); + auto v3 = in1_data.template chip<1>(2); + + s1.device(d) = u2 * v3 - u3 * v2; + s2.device(d) = u3 * v1 - u1 * v3; + s3.device(d) = u1 * v2 - u2 * v1; + } +}; + +} // namespace functor +} // namespace tensorflow + +#endif // TENSORFLOW_KERNELS_COLORSPACE_OP_H_ diff --git a/tensorflow/core/kernels/cross_op_gpu.cu.cc b/tensorflow/core/kernels/cross_op_gpu.cu.cc new file mode 100644 index 0000000000..a20ef8cf20 --- /dev/null +++ b/tensorflow/core/kernels/cross_op_gpu.cu.cc @@ -0,0 +1,34 @@ +/* Copyright 2015 Google Inc. All Rights Reserved. + +Licensed under the Apache License, Version 2.0 (the "License"); +you may not use this file except in compliance with the License. +You may obtain a copy of the License at + + http://www.apache.org/licenses/LICENSE-2.0 + +Unless required by applicable law or agreed to in writing, software +distributed under the License is distributed on an "AS IS" BASIS, +WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied. +See the License for the specific language governing permissions and +limitations under the License. +==============================================================================*/ + +#if GOOGLE_CUDA + +#define EIGEN_USE_GPU + +#include "tensorflow/core/kernels/cross_op.h" +#include "tensorflow/core/framework/register_types.h" + +namespace tensorflow { + +typedef Eigen::GpuDevice GPUDevice; + +#define INSTANTIATE_GPU_KERNEL(type) \ + template class functor::Cross<GPUDevice, type>; +TF_CALL_REAL_NUMBER_TYPES(INSTANTIATE_GPU_KERNEL); +#undef INSTANTIATE_GPU_KERNEL + +} // namespace tensorflow + +#endif // GOOGLE_CUDA diff --git a/tensorflow/core/kernels/cross_op_test.cc b/tensorflow/core/kernels/cross_op_test.cc new file mode 100644 index 0000000000..0b179e6d5d --- /dev/null +++ b/tensorflow/core/kernels/cross_op_test.cc @@ -0,0 +1,102 @@ +/* Copyright 2015 Google Inc. All Rights Reserved. + +Licensed under the Apache License, Version 2.0 (the "License"); +you may not use this file except in compliance with the License. +You may obtain a copy of the License at + + http://www.apache.org/licenses/LICENSE-2.0 + +Unless required by applicable law or agreed to in writing, software +distributed under the License is distributed on an "AS IS" BASIS, +WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied. +See the License for the specific language governing permissions and +limitations under the License. +==============================================================================*/ + +#include "tensorflow/core/framework/allocator.h" +#include "tensorflow/core/framework/fake_input.h" +#include "tensorflow/core/framework/graph.pb.h" +#include "tensorflow/core/framework/node_def_builder.h" +#include "tensorflow/core/framework/op_kernel.h" +#include "tensorflow/core/framework/tensor.h" +#include "tensorflow/core/framework/tensor_testutil.h" +#include "tensorflow/core/framework/types.h" +#include "tensorflow/core/framework/types.pb.h" +#include "tensorflow/core/kernels/ops_testutil.h" +#include "tensorflow/core/kernels/ops_util.h" +#include "tensorflow/core/lib/core/status_test_util.h" +#include "tensorflow/core/platform/test.h" + +namespace tensorflow { + +class CrossOpTest : public OpsTestBase { + protected: + CrossOpTest() { + RequireDefaultOps(); + TF_EXPECT_OK(NodeDefBuilder("cross_op", "Cross") + .Input(FakeInput(DT_FLOAT)) + .Input(FakeInput(DT_FLOAT)) + .Finalize(node_def())); + TF_EXPECT_OK(InitOp()); + } +}; + +TEST_F(CrossOpTest, Zero) { + AddInputFromArray<float>(TensorShape({3}), {0, 0, 0}); + AddInputFromArray<float>(TensorShape({3}), {0, 0, 0}); + TF_ASSERT_OK(RunOpKernel()); + + Tensor expected(allocator(), DT_FLOAT, TensorShape({3})); + test::FillValues<float>(&expected, {0, 0, 0}); + test::ExpectTensorEqual<float>(expected, *GetOutput(0)); +} + +TEST_F(CrossOpTest, RightHandRule) { + AddInputFromArray<float>(TensorShape({2, 3}), {1, 0, 0, /**/ 0, 1, 0}); + AddInputFromArray<float>(TensorShape({2, 3}), {0, 1, 0, /**/ 1, 0, 0}); + TF_ASSERT_OK(RunOpKernel()); + + Tensor expected(allocator(), DT_FLOAT, TensorShape({2, 3})); + test::FillValues<float>(&expected, {{0, 0, 1, /**/ 0, 0, -1}}); + test::ExpectTensorEqual<float>(expected, *GetOutput(0)); +} + +TEST_F(CrossOpTest, ArbitraryNonintegral) { + const float u1 = -0.669, u2 = -0.509, u3 = 0.125; + const float v1 = -0.477, v2 = 0.592, v3 = -0.110; + const float s1 = u2 * v3 - u3 * v2; + const float s2 = u3 * v1 - u1 * v3; + const float s3 = u1 * v2 - u2 * v1; + + AddInputFromArray<float>(TensorShape({3}), {u1, u2, u3}); + AddInputFromArray<float>(TensorShape({3}), {v1, v2, v3}); + TF_ASSERT_OK(RunOpKernel()); + + Tensor expected(allocator(), DT_FLOAT, TensorShape({3})); + test::FillValues<float>(&expected, {s1, s2, s3}); + test::ExpectTensorNear<float>(expected, *GetOutput(0), 1e-6); +} + +class CrossOpIntTest : public OpsTestBase { + protected: + CrossOpIntTest() { + RequireDefaultOps(); + TF_EXPECT_OK(NodeDefBuilder("cross_int_op", "Cross") + .Input(FakeInput(DT_INT32)) + .Input(FakeInput(DT_INT32)) + .Finalize(node_def())); + TF_EXPECT_OK(InitOp()); + } +}; + +TEST_F(CrossOpIntTest, RightHandRule) { + AddInputFromArray<int>(TensorShape({2, 3}), {2, 0, 0, /**/ 0, 2, 0}); + AddInputFromArray<int>(TensorShape({2, 3}), {0, 2, 0, /**/ 2, 0, 0}); + TF_ASSERT_OK(RunOpKernel()); + + Tensor expected(allocator(), DT_INT32, TensorShape({2, 3})); + test::FillValues<int>(&expected, {{0, 0, 4, /**/ 0, 0, -4}}); + test::ExpectTensorEqual<int>(expected, *GetOutput(0)); +} + +} // namespace tensorflow diff --git a/tensorflow/core/kernels/transpose_op.cc b/tensorflow/core/kernels/transpose_op.cc index afab885099..e7294cb3e0 100644 --- a/tensorflow/core/kernels/transpose_op.cc +++ b/tensorflow/core/kernels/transpose_op.cc @@ -68,6 +68,12 @@ class InvertPermutationOp : public OpKernel { REGISTER_KERNEL_BUILDER(Name("InvertPermutation").Device(DEVICE_CPU), InvertPermutationOp); +REGISTER_KERNEL_BUILDER(Name("InvertPermutation") + .Device(DEVICE_GPU) + .HostMemory("x") + .HostMemory("y"), + InvertPermutationOp); + // output = TransposeOp(T<any> input, T<int32> perm) takes a tensor // of type T and rank N, and a permutation of 0, 1, ..., N-1. It // shuffles the dimensions of the input tensor according to permutation. diff --git a/tensorflow/core/ops/math_ops.cc b/tensorflow/core/ops/math_ops.cc index 56c6c8e920..603005126d 100644 --- a/tensorflow/core/ops/math_ops.cc +++ b/tensorflow/core/ops/math_ops.cc @@ -1180,4 +1180,23 @@ out: The inverse 2D Fourier Transform of `in`. )doc"); +// -------------------------------------------------------------------------- + +REGISTER_OP("Cross") + .Input("a: T") + .Input("b: T") + .Output("product: T") + .Attr("T: realnumbertype") + .Doc(R"doc( +Compute the pairwise cross product. + +`a` and `b` must be the same shape; they can either be simple 3-element vectors, +or any shape where the innermost dimension is 3. In the latter case, each pair +of corresponding 3-element vectors is cross-multiplied independently. + +a: A tensor containing 3-element vectors. +b: Another tensor, of same type and shape as `a`. +product: Pairwise cross product of the vectors in `a` and `b`. +)doc"); + } // namespace tensorflow diff --git a/tensorflow/core/ops/ops.pbtxt b/tensorflow/core/ops/ops.pbtxt index 1e32f3696d..85eb0965b3 100644 --- a/tensorflow/core/ops/ops.pbtxt +++ b/tensorflow/core/ops/ops.pbtxt @@ -1932,6 +1932,42 @@ op { description: "This operation outputs \"ref\" after the update is done. This makes it\neasier to chain operations that need to use the updated value." } op { + name: "Cross" + input_arg { + name: "a" + description: "A tensor containing 3-element vectors." + type_attr: "T" + } + input_arg { + name: "b" + description: "Another tensor, of same type and shape as `a`." + type_attr: "T" + } + output_arg { + name: "product" + description: "Pairwise cross product of the vectors in `a` and `b`." + type_attr: "T" + } + attr { + name: "T" + type: "type" + allowed_values { + list { + type: DT_FLOAT + type: DT_DOUBLE + type: DT_INT32 + type: DT_INT64 + type: DT_UINT8 + type: DT_INT16 + type: DT_INT8 + type: DT_UINT16 + } + } + } + summary: "Compute the pairwise cross product." + description: "`a` and `b` must be the same shape; they can either be simple 3-element vectors,\nor any shape where the innermost dimension is 3. In the latter case, each pair\nof corresponding 3-element vectors is cross-multiplied independently." +} +op { name: "DecodeCSV" input_arg { name: "records" diff --git a/tensorflow/python/kernel_tests/cross_grad_test.py b/tensorflow/python/kernel_tests/cross_grad_test.py new file mode 100644 index 0000000000..9d29b35129 --- /dev/null +++ b/tensorflow/python/kernel_tests/cross_grad_test.py @@ -0,0 +1,41 @@ +# Copyright 2015 Google Inc. All Rights Reserved. +# +# Licensed under the Apache License, Version 2.0 (the "License"); +# you may not use this file except in compliance with the License. +# You may obtain a copy of the License at +# +# http://www.apache.org/licenses/LICENSE-2.0 +# +# Unless required by applicable law or agreed to in writing, software +# distributed under the License is distributed on an "AS IS" BASIS, +# WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied. +# See the License for the specific language governing permissions and +# limitations under the License. +# ============================================================================== +"""Tests for tensorflow.ops.nn_ops.Cross.""" + +from __future__ import absolute_import +from __future__ import division +from __future__ import print_function + +import tensorflow.python.platform + +import tensorflow as tf + + +class CrossOpTest(tf.test.TestCase): + + def testGradientRandomValues(self): + with self.test_session(): + us = [2, 3] + u = tf.reshape([0.854, -0.616, 0.767, 0.725, -0.927, 0.159], shape=us) + v = tf.reshape([-0.522, 0.755, 0.407, -0.652, 0.241, 0.247], shape=us) + s = tf.cross(u, v) + jacob_u, jacob_v = tf.test.compute_gradient([u, v], [us, us], s, us) + + self.assertAllClose(jacob_u[0], jacob_u[1], rtol=1e-3, atol=1e-3) + self.assertAllClose(jacob_v[0], jacob_v[1], rtol=1e-3, atol=1e-3) + + +if __name__ == "__main__": + tf.test.main() diff --git a/tensorflow/python/kernel_tests/seq2seq_test.py b/tensorflow/python/kernel_tests/seq2seq_test.py index cebc3244d8..b146390682 100644 --- a/tensorflow/python/kernel_tests/seq2seq_test.py +++ b/tensorflow/python/kernel_tests/seq2seq_test.py @@ -433,7 +433,7 @@ class Seq2SeqTest(tf.test.TestCase): perplexities[bucket].append(math.exp(float(res[1]))) for bucket in range(len(buckets)): if len(perplexities[bucket]) > 1: # Assert that perplexity went down. - self.assertLess(perplexities[bucket][1], perplexities[bucket][0]) + self.assertLess(perplexities[bucket][-1], perplexities[bucket][0]) if __name__ == "__main__": diff --git a/tensorflow/python/ops/array_ops.py b/tensorflow/python/ops/array_ops.py index 39ef55ab2a..e9eeb2619b 100644 --- a/tensorflow/python/ops/array_ops.py +++ b/tensorflow/python/ops/array_ops.py @@ -560,8 +560,8 @@ def transpose(a, perm=None, name="transpose"): """ with ops.op_scope([a], name, "transpose") as name: if perm is None: - dims = gen_math_ops._range(0, gen_array_ops.rank(a), 1) - perm = gen_array_ops.reverse(dims, [True]) + rank = gen_array_ops.rank(a) + perm = (rank - 1) - gen_math_ops._range(0, rank, 1) ret = gen_array_ops.transpose(a, perm, name=name) # NOTE(mrry): Setting the shape explicitly because # reverse is not handled by the shape function. diff --git a/tensorflow/python/ops/math_grad.py b/tensorflow/python/ops/math_grad.py index 7a5b5e4c86..8b2779068b 100644 --- a/tensorflow/python/ops/math_grad.py +++ b/tensorflow/python/ops/math_grad.py @@ -567,3 +567,10 @@ def _FFT2DGrad(_, grad): def _IFFT2DGrad(_, grad): rsize = 1. / math_ops.cast(array_ops.size(grad), dtypes.float32) return math_ops.fft2d(grad) * math_ops.complex(rsize, 0.) + + +@ops.RegisterGradient("Cross") +def _CrossGrad(op, grad): + u = op.inputs[0] + v = op.inputs[1] + return (math_ops.cross(v, grad), math_ops.cross(grad, u)) diff --git a/tensorflow/python/ops/math_ops.py b/tensorflow/python/ops/math_ops.py index d878347863..90ed7ad876 100644 --- a/tensorflow/python/ops/math_ops.py +++ b/tensorflow/python/ops/math_ops.py @@ -25,6 +25,7 @@ operators to your graph. @@truediv @@floordiv @@mod +@@cross ## Basic Math Functions @@ -1223,6 +1224,7 @@ ops.RegisterShape("Abs")(common_shapes.unchanged_shape) ops.RegisterShape("Ceil")(common_shapes.unchanged_shape) ops.RegisterShape("Conj")(common_shapes.unchanged_shape) ops.RegisterShape("Cos")(common_shapes.unchanged_shape) +ops.RegisterShape("Cross")(common_shapes.unchanged_shape) ops.RegisterShape("Exp")(common_shapes.unchanged_shape) ops.RegisterShape("Floor")(common_shapes.unchanged_shape) ops.RegisterShape("Imag")(common_shapes.unchanged_shape) diff --git a/third_party/eigen3/Eigen/Cholesky b/third_party/eigen3/Eigen/Cholesky index ef31fc971b..671ec3d4a6 100644 --- a/third_party/eigen3/Eigen/Cholesky +++ b/third_party/eigen3/Eigen/Cholesky @@ -1 +1 @@ -#include "eigen-eigen-8cd7c2c6e9e1/Eigen/Cholesky"
\ No newline at end of file +#include "eigen-eigen-726c779797e8/Eigen/Cholesky" diff --git a/third_party/eigen3/Eigen/Core b/third_party/eigen3/Eigen/Core index a330b6166f..38f45037e6 100644 --- a/third_party/eigen3/Eigen/Core +++ b/third_party/eigen3/Eigen/Core @@ -1 +1 @@ -#include "eigen-eigen-8cd7c2c6e9e1/Eigen/Core" +#include "eigen-eigen-726c779797e8/Eigen/Core" diff --git a/third_party/eigen3/Eigen/Eigenvalues b/third_party/eigen3/Eigen/Eigenvalues index 30158ba1ea..64f4200304 100644 --- a/third_party/eigen3/Eigen/Eigenvalues +++ b/third_party/eigen3/Eigen/Eigenvalues @@ -1 +1,2 @@ -#include "eigen-eigen-8cd7c2c6e9e1/Eigen/Eigenvalues" +#include "eigen-eigen-726c779797e8/Eigen/Eigenvalues" + diff --git a/third_party/eigen3/Eigen/LU b/third_party/eigen3/Eigen/LU index 5637771a51..ab9e6cb4c5 100644 --- a/third_party/eigen3/Eigen/LU +++ b/third_party/eigen3/Eigen/LU @@ -1 +1 @@ -#include "eigen-eigen-8cd7c2c6e9e1/Eigen/LU" +#include "eigen-eigen-726c779797e8/Eigen/LU" diff --git a/third_party/eigen3/Eigen/QR b/third_party/eigen3/Eigen/QR index 360ba8e5e3..9ecf7be16d 100644 --- a/third_party/eigen3/Eigen/QR +++ b/third_party/eigen3/Eigen/QR @@ -1 +1 @@ -#include "eigen-eigen-8cd7c2c6e9e1/Eigen/QR" +#include "eigen-eigen-726c779797e8/Eigen/QR" diff --git a/third_party/eigen3/unsupported/Eigen/CXX11/Tensor b/third_party/eigen3/unsupported/Eigen/CXX11/Tensor index eb293afd04..a80816717b 100644 --- a/third_party/eigen3/unsupported/Eigen/CXX11/Tensor +++ b/third_party/eigen3/unsupported/Eigen/CXX11/Tensor @@ -1,2 +1 @@ - -#include "eigen-eigen-8cd7c2c6e9e1/unsupported/Eigen/CXX11/Tensor" +#include "eigen-eigen-726c779797e8/unsupported/Eigen/CXX11/Tensor" |