Merge with upstream

Eigen/src/Core/GlobalFunctions.h

@@ -66,6 +66,7 @@ namespace Eigen
   EIGEN_ARRAY_DECLARE_GLOBAL_UNARY(sinh,scalar_sinh_op,hyperbolic sine,\sa ArrayBase::sinh)
   EIGEN_ARRAY_DECLARE_GLOBAL_UNARY(cosh,scalar_cosh_op,hyperbolic cosine,\sa ArrayBase::cosh)
   EIGEN_ARRAY_DECLARE_GLOBAL_UNARY(tanh,scalar_tanh_op,hyperbolic tangent,\sa ArrayBase::tanh)
+  EIGEN_ARRAY_DECLARE_GLOBAL_UNARY(logistic,scalar_logistic_op,logistic function,\sa ArrayBase::logistic)
   EIGEN_ARRAY_DECLARE_GLOBAL_UNARY(lgamma,scalar_lgamma_op,natural logarithm of the gamma function,\sa ArrayBase::lgamma)
   EIGEN_ARRAY_DECLARE_GLOBAL_UNARY(digamma,scalar_digamma_op,derivative of lgamma,\sa ArrayBase::digamma)
   EIGEN_ARRAY_DECLARE_GLOBAL_UNARY(erf,scalar_erf_op,error function,\sa ArrayBase::erf)

Eigen/src/Core/functors/UnaryFunctors.h

@@ -823,6 +823,34 @@ struct functor_traits<scalar_sign_op<Scalar> >
   };
 };
 
+/** \internal
+  * \brief Template functor to compute the logistic function of a scalar
+  * \sa class CwiseUnaryOp, ArrayBase::logistic()
+  */
+template <typename T>
+struct scalar_logistic_op {
+  EIGEN_EMPTY_STRUCT_CTOR(scalar_logistic_op)
+  EIGEN_DEVICE_FUNC EIGEN_STRONG_INLINE T operator()(const T& x) const {
+    const T one = T(1);
+    return one / (one + numext::exp(-x));
+  }
+
+  template <typename Packet> EIGEN_DEVICE_FUNC EIGEN_STRONG_INLINE
+  Packet packetOp(const Packet& x) const {
+    const Packet one = pset1<Packet>(T(1));
+    return pdiv(one, padd(one, pexp(pnegate(x))));
+  }
+};
+template <typename T>
+struct functor_traits<scalar_logistic_op<T> > {
+  enum {
+    Cost = NumTraits<T>::AddCost * 2 + NumTraits<T>::MulCost * 6,
+    PacketAccess = packet_traits<T>::HasAdd && packet_traits<T>::HasDiv &&
+                   packet_traits<T>::HasNegate && packet_traits<T>::HasExp
+  };
+};
+
+
 } // end namespace internal
 
 } // end namespace Eigen

Eigen/src/plugins/ArrayCwiseUnaryOps.h

@@ -21,6 +21,7 @@ typedef CwiseUnaryOp<internal::scalar_acos_op<Scalar>, const Derived> AcosReturn
 typedef CwiseUnaryOp<internal::scalar_asin_op<Scalar>, const Derived> AsinReturnType;
 typedef CwiseUnaryOp<internal::scalar_atan_op<Scalar>, const Derived> AtanReturnType;
 typedef CwiseUnaryOp<internal::scalar_tanh_op<Scalar>, const Derived> TanhReturnType;
+typedef CwiseUnaryOp<internal::scalar_logistic_op<Scalar>, const Derived> LogisticReturnType;
 typedef CwiseUnaryOp<internal::scalar_sinh_op<Scalar>, const Derived> SinhReturnType;
 typedef CwiseUnaryOp<internal::scalar_cosh_op<Scalar>, const Derived> CoshReturnType;
 typedef CwiseUnaryOp<internal::scalar_square_op<Scalar>, const Derived> SquareReturnType;
@@ -335,6 +336,15 @@ cosh() const
   return CoshReturnType(derived());
 }
 
+/** \returns an expression of the coefficient-wise logistic of *this.
+  */
+EIGEN_DEVICE_FUNC
+inline const LogisticReturnType
+logistic() const
+{
+  return LogisticReturnType(derived());
+}
+
 /** \returns an expression of the coefficient-wise inverse of *this.
   *
   * Example: \include Cwise_inverse.cpp

test/array.cpp

@@ -231,6 +231,7 @@ template<typename ArrayType> void array_real(const ArrayType& m)
   VERIFY_IS_APPROX(m1.sinh(), sinh(m1));
   VERIFY_IS_APPROX(m1.cosh(), cosh(m1));
   VERIFY_IS_APPROX(m1.tanh(), tanh(m1));
+  VERIFY_IS_APPROX(m1.logistic(), logistic(m1));
 
   VERIFY_IS_APPROX(m1.arg(), arg(m1));
   VERIFY_IS_APPROX(m1.round(), round(m1));
@@ -266,6 +267,7 @@ template<typename ArrayType> void array_real(const ArrayType& m)
   VERIFY_IS_APPROX(sinh(m1), 0.5*(exp(m1)-exp(-m1)));
   VERIFY_IS_APPROX(cosh(m1), 0.5*(exp(m1)+exp(-m1)));
   VERIFY_IS_APPROX(tanh(m1), (0.5*(exp(m1)-exp(-m1)))/(0.5*(exp(m1)+exp(-m1))));
+  VERIFY_IS_APPROX(logistic(m1), (1.0/(1.0+exp(-m1))));
   VERIFY_IS_APPROX(arg(m1), ((m1<0).template cast<Scalar>())*std::acos(-1.0));
   VERIFY((round(m1) <= ceil(m1) && round(m1) >= floor(m1)).all());
   VERIFY((Eigen::isnan)((m1*0.0)/0.0).all());
@@ -345,6 +347,7 @@ template<typename ArrayType> void array_complex(const ArrayType& m)
   VERIFY_IS_APPROX(m1.sinh(), sinh(m1));
   VERIFY_IS_APPROX(m1.cosh(), cosh(m1));
   VERIFY_IS_APPROX(m1.tanh(), tanh(m1));
+  VERIFY_IS_APPROX(m1.logistic(), logistic(m1));
   VERIFY_IS_APPROX(m1.arg(), arg(m1));
   VERIFY((m1.isNaN() == (Eigen::isnan)(m1)).all());
   VERIFY((m1.isInf() == (Eigen::isinf)(m1)).all());
@@ -368,6 +371,7 @@ template<typename ArrayType> void array_complex(const ArrayType& m)
   VERIFY_IS_APPROX(sinh(m1), 0.5*(exp(m1)-exp(-m1)));
   VERIFY_IS_APPROX(cosh(m1), 0.5*(exp(m1)+exp(-m1)));
   VERIFY_IS_APPROX(tanh(m1), (0.5*(exp(m1)-exp(-m1)))/(0.5*(exp(m1)+exp(-m1))));
+  VERIFY_IS_APPROX(logistic(m1), (1.0/(1.0 + exp(-m1))));
 
   for (Index i = 0; i < m.rows(); ++i)
     for (Index j = 0; j < m.cols(); ++j)

unsupported/Eigen/CXX11/src/Tensor/TensorAssign.h

@@ -187,7 +187,7 @@ struct TensorEvaluator<const TensorAssignOp<LeftArgType, RightArgType>, Device>
 
   EIGEN_DEVICE_FUNC EIGEN_STRONG_INLINE void evalBlock(TensorBlock* block) {
     if (TensorEvaluator<LeftArgType, Device>::RawAccess &&
-        m_leftImpl.data() != nullptr) {
+        m_leftImpl.data() != NULL) {
       TensorBlock left_block(block->first_coeff_index(), block->block_sizes(),
                              block->tensor_strides(), block->tensor_strides(),
                              m_leftImpl.data() + block->first_coeff_index());

unsupported/Eigen/CXX11/src/Tensor/TensorBase.h

@@ -200,9 +200,9 @@ class TensorBase<Derived, ReadOnlyAccessors>
     }
 
     EIGEN_DEVICE_FUNC
-    EIGEN_STRONG_INLINE const TensorCwiseUnaryOp<internal::scalar_sigmoid_op<Scalar>, const Derived>
+    EIGEN_STRONG_INLINE const TensorCwiseUnaryOp<internal::scalar_logistic_op<Scalar>, const Derived>
     sigmoid() const {
-      return unaryExpr(internal::scalar_sigmoid_op<Scalar>());
+      return unaryExpr(internal::scalar_logistic_op<Scalar>());
     }
 
     EIGEN_DEVICE_FUNC

unsupported/Eigen/CXX11/src/Tensor/TensorBlock.h

@@ -155,8 +155,8 @@ struct TensorBlockCopyOp {
     typedef const Eigen::Array<Scalar, Dynamic, 1> Src;
     typedef Eigen::Array<Scalar, Dynamic, 1> Dst;
 
-    typedef Eigen::Map<Src, 0, InnerStride<>> SrcMap;
+    typedef Eigen::Map<Src, 0, InnerStride<> > SrcMap;
-    typedef Eigen::Map<Dst, 0, InnerStride<>> DstMap;
+    typedef Eigen::Map<Dst, 0, InnerStride<> > DstMap;
 
     const SrcMap src(src_base, num_coeff_to_copy, InnerStride<>(src_stride));
     DstMap dst(dst_base, num_coeff_to_copy, InnerStride<>(dst_stride));
@@ -405,9 +405,9 @@ struct TensorBlockCwiseBinaryOp {
     typedef const Eigen::Array<RightScalar, Dynamic, 1> Rhs;
     typedef Eigen::Array<OutputScalar, Dynamic, 1> Out;
 
-    typedef Eigen::Map<Lhs, 0, InnerStride<>> LhsMap;
+    typedef Eigen::Map<Lhs, 0, InnerStride<> > LhsMap;
-    typedef Eigen::Map<Rhs, 0, InnerStride<>> RhsMap;
+    typedef Eigen::Map<Rhs, 0, InnerStride<> > RhsMap;
-    typedef Eigen::Map<Out, 0, InnerStride<>> OutMap;
+    typedef Eigen::Map<Out, 0, InnerStride<> > OutMap;
 
     const LeftScalar* lhs_base = &left_data[left_index];
     const RightScalar* rhs_base = &right_data[right_index];
@@ -501,7 +501,7 @@ struct TensorBlockCwiseBinaryIO {
       if (size == 1) {
         continue;
       }
-      auto& state = block_iter_state[num_squeezed_dims];
+      BlockIteratorState& state = block_iter_state[num_squeezed_dims];
       state.output_stride = block_strides[dim];
       state.left_stride = left_strides[dim];
       state.right_stride = right_strides[dim];
@@ -523,7 +523,7 @@ struct TensorBlockCwiseBinaryIO {
                                     right_stride, right_data);
       // Update index.
       for (int j = 0; j < num_squeezed_dims; ++j) {
-        auto& state = block_iter_state[j];
+        BlockIteratorState& state = block_iter_state[j];
         if (++state.count < state.size) {
           output_index += state.output_stride;
           left_index += state.left_stride;

unsupported/Eigen/CXX11/src/Tensor/TensorDeviceThreadPool.h

@@ -102,7 +102,7 @@ class Allocator {
 // Build a thread pool device on top the an existing pool of threads.
 struct ThreadPoolDevice {
   // The ownership of the thread pool remains with the caller.
-  ThreadPoolDevice(ThreadPoolInterface* pool, int num_cores, Allocator* allocator = nullptr)
+  ThreadPoolDevice(ThreadPoolInterface* pool, int num_cores, Allocator* allocator = NULL)
       : pool_(pool), num_threads_(num_cores), allocator_(allocator) { }
 
   EIGEN_STRONG_INLINE void* allocate(size_t num_bytes) const {
@@ -282,7 +282,7 @@ struct ThreadPoolDevice {
   // Convenience wrapper for parallelFor that does not align blocks.
   void parallelFor(Index n, const TensorOpCost& cost,
                    std::function<void(Index, Index)> f) const {
-    parallelFor(n, cost, nullptr, std::move(f));
+    parallelFor(n, cost, NULL, std::move(f));
   }
 
   // Thread pool accessor.

unsupported/Eigen/CXX11/src/Tensor/TensorExecutor.h

@@ -227,7 +227,7 @@ class TensorExecutor<Expression, ThreadPoolDevice, Vectorizable, Tileable> {
     typedef EvalRange<Evaluator, StorageIndex, Vectorizable> EvalRange;
 
     Evaluator evaluator(expr, device);
-    const bool needs_assign = evaluator.evalSubExprsIfNeeded(nullptr);
+    const bool needs_assign = evaluator.evalSubExprsIfNeeded(NULL);
     if (needs_assign) {
       const StorageIndex PacketSize =
           Vectorizable
@@ -257,7 +257,6 @@ class TensorExecutor<Expression, ThreadPoolDevice, Vectorizable, /*Tileable*/ tr
 
   static EIGEN_STRONG_INLINE void run(const Expression& expr,
                          const ThreadPoolDevice& device) {
-    typedef TensorBlock<ScalarNoConst, StorageIndex, NumDims, Evaluator::Layout> TensorBlock;
     typedef TensorBlockMapper<ScalarNoConst, StorageIndex, NumDims, Evaluator::Layout> TensorBlockMapper;
 
     Evaluator evaluator(expr, device);
@@ -271,7 +270,7 @@ class TensorExecutor<Expression, ThreadPoolDevice, Vectorizable, /*Tileable*/ tr
       return;
     }
 
-    const bool needs_assign = evaluator.evalSubExprsIfNeeded(nullptr);
+    const bool needs_assign = evaluator.evalSubExprsIfNeeded(NULL);
     if (needs_assign) {
       TensorBlockShapeType block_shape = TensorBlockShapeType::kSkewedInnerDims;
       Index block_total_size = 0;

unsupported/Eigen/CXX11/src/Tensor/TensorFunctors.h

@@ -54,36 +54,6 @@ struct functor_traits<scalar_fmod_op<Scalar> > {
          PacketAccess = false };
 };
 
-
-/** \internal
-  * \brief Template functor to compute the sigmoid of a scalar
-  * \sa class CwiseUnaryOp, ArrayBase::sigmoid()
-  */
-template <typename T>
-struct scalar_sigmoid_op {
-  EIGEN_EMPTY_STRUCT_CTOR(scalar_sigmoid_op)
-  EIGEN_DEVICE_FUNC EIGEN_STRONG_INLINE T operator()(const T& x) const {
-    const T one = T(1);
-    return one / (one + numext::exp(-x));
-  }
-
-  template <typename Packet> EIGEN_DEVICE_FUNC EIGEN_STRONG_INLINE
-  Packet packetOp(const Packet& x) const {
-    const Packet one = pset1<Packet>(T(1));
-    return pdiv(one, padd(one, pexp(pnegate(x))));
-  }
-};
-
-template <typename T>
-struct functor_traits<scalar_sigmoid_op<T> > {
-  enum {
-    Cost = NumTraits<T>::AddCost * 2 + NumTraits<T>::MulCost * 6,
-    PacketAccess = packet_traits<T>::HasAdd && packet_traits<T>::HasDiv &&
-                   packet_traits<T>::HasNegate && packet_traits<T>::HasExp
-  };
-};
-
-
 template<typename Reducer, typename Device>
 struct reducer_traits {
   enum {

unsupported/test/cxx11_tensor_thread_pool.cpp

@@ -428,7 +428,7 @@ void test_threadpool_allocate(TestAllocator* allocator)
     void* ptr = device.allocate(512);
     device.deallocate(ptr);
   }
-  VERIFY(allocator != nullptr);
+  VERIFY(allocator != NULL);
   VERIFY_IS_EQUAL(allocator->alloc_count(), num_allocs);
   VERIFY_IS_EQUAL(allocator->dealloc_count(), num_allocs);
 }
@@ -460,7 +460,7 @@ EIGEN_DECLARE_TEST(cxx11_tensor_thread_pool)
   CALL_SUBTEST_6(test_multithread_random());
 
   TestAllocator test_allocator;
-  CALL_SUBTEST_6(test_multithread_shuffle<ColMajor>(nullptr));
+  CALL_SUBTEST_6(test_multithread_shuffle<ColMajor>(NULL));
   CALL_SUBTEST_6(test_multithread_shuffle<RowMajor>(&test_allocator));
   CALL_SUBTEST_6(test_threadpool_allocate(&test_allocator));
 }