GPU Kernel Information
| layer_index | layer_name | layer_type | layer_shape | layer_duration (us) | layer_allocated_bytes | layer_peak_allocated_bytes | layer_allocator_bytes_in_use | layer_allocator_name | layer_host_temp_mem_bytes | layer_device_temp_mem_bytes | layer_host_persistent_mem_bytes | layer_device_persistent_mem_bytes | kernel_name | kernel_duration (us) | kernel_flops | kernel_dram_read_bytes | kernel_dram_write_bytes | kernel_achieved_occupancy (%) | kernel_arithmetic_intensity (flops/byte) | kernel_arithmetic_throughput (GFlops) | kernel_memory_bound | flop_count_sp |
|---|
layer_index | layer_name | layer_type | layer_shape | layer_duration (us) | layer_allocated_bytes | layer_peak_allocated_bytes | layer_allocator_bytes_in_use | layer_allocator_name | layer_host_temp_mem_bytes | layer_device_temp_mem_bytes | layer_host_persistent_mem_bytes | layer_device_persistent_mem_bytes | kernel_name | kernel_duration (us) | kernel_flops | kernel_dram_read_bytes | kernel_dram_write_bytes | kernel_achieved_occupancy (%) | kernel_arithmetic_intensity (flops/byte) | kernel_arithmetic_throughput (GFlops) | kernel_memory_bound | flop_count_sp |
|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|
| 1 | InceptionV2/InceptionV2/Conv2d_1a_7x7/separable_conv2d/depthwise-0-TransposeNHWCToNCHW-LayoutOptimizer | Transpose | [[1 3 224 224]] | 105 | 602112 | 602112 | 45926912 | GPU_0_bfc | 0 | 0 | 0 | 0 | void tensorflow::functor::SwapDimension1And2InTensor3UsingTiles<unsigned int, 1024, 1024, 2, false>(unsigned int const*, tensorflow::functor::Dimension<3>, unsigned int*) | 10.67 | 0 | 0.00 | 0.00 | 0.00 | 0.00 | 0.00 | true | 0;0;0;0;0 |
| 2 | InceptionV2/InceptionV2/Conv2d_1a_7x7/separable_conv2d/depthwise | DepthwiseConv2dNative | [[1 24 112 112]] | 486.333 | 1204224 | 6734080 | 46529024 | GPU_0_bfc | 5529856 | 0 | 0 | 0 | void cudnn::detail::explicit_convolve_sgemm<float, int, 1024, 5, 5, 3, 3, 3, 0, true>(int, int, int, float const*, int, float const*, int, float*, kernel_conv_params, int, int, float, float, int, float*, float*) | 16.00 | 45058048 | 0.00 | 0.00 | 0.00 | 0.00 | 2816.13 | true | 45058048;45058048;45058048;45058048;45058048;45058048;45058048;45058048;45058048;45058048;45058048;45058048;45058048;45058048;45058048 |
| 2 | InceptionV2/InceptionV2/Conv2d_1a_7x7/separable_conv2d/depthwise | DepthwiseConv2dNative | [[1 24 112 112]] | 486.333 | 1204224 | 6734080 | 46529024 | GPU_0_bfc | 5529856 | 0 | 0 | 0 | void cudnn::detail::explicit_convolve_sgemm<float, int, 1024, 5, 5, 3, 3, 3, 0, true>(int, int, int, float const*, int, float const*, int, float*, kernel_conv_params, int, int, float, float, int, float*, float*) | 16.00 | 45058048 | 0.00 | 0.00 | 0.00 | 0.00 | 2816.13 | true | 45058048;45058048;45058048;45058048;45058048;45058048;45058048;45058048;45058048;45058048;45058048;45058048;45058048;45058048;45058048 |
| 2 | InceptionV2/InceptionV2/Conv2d_1a_7x7/separable_conv2d/depthwise | DepthwiseConv2dNative | [[1 24 112 112]] | 486.333 | 1204224 | 6734080 | 46529024 | GPU_0_bfc | 5529856 | 0 | 0 | 0 | void cudnn::detail::explicit_convolve_sgemm<float, int, 1024, 5, 5, 3, 3, 3, 0, true>(int, int, int, float const*, int, float const*, int, float*, kernel_conv_params, int, int, float, float, int, float*, float*) | 16.00 | 45058048 | 0.00 | 0.00 | 0.00 | 0.00 | 2816.13 | true | 45058048;45058048;45058048;45058048;45058048;45058048;45058048;45058048;45058048;45058048;45058048;45058048;45058048;45058048;45058048 |
| 2 | InceptionV2/InceptionV2/Conv2d_1a_7x7/separable_conv2d/depthwise | DepthwiseConv2dNative | [[1 24 112 112]] | 486.333 | 1204224 | 6734080 | 46529024 | GPU_0_bfc | 5529856 | 0 | 0 | 0 | void im2col4d_kernel<float, int>(im2col4d_params, cudnnConvolutionStruct, cudnnTensor4dStruct, float const*, float*, int) | 15.28 | 0 | 0.00 | 0.00 | 0.00 | 0.00 | 0.00 | true | 0;0;0;0;0;0;0;0;0;0;0;0;0;0;0 |
| 2 | InceptionV2/InceptionV2/Conv2d_1a_7x7/separable_conv2d/depthwise | DepthwiseConv2dNative | [[1 24 112 112]] | 486.333 | 1204224 | 6734080 | 46529024 | GPU_0_bfc | 5529856 | 0 | 0 | 0 | void im2col4d_kernel<float, int>(im2col4d_params, cudnnConvolutionStruct, cudnnTensor4dStruct, float const*, float*, int) | 15.22 | 0 | 0.00 | 0.00 | 0.00 | 0.00 | 0.00 | true | 0;0;0;0;0;0;0;0;0;0;0;0;0;0;0 |
| 2 | InceptionV2/InceptionV2/Conv2d_1a_7x7/separable_conv2d/depthwise | DepthwiseConv2dNative | [[1 24 112 112]] | 486.333 | 1204224 | 6734080 | 46529024 | GPU_0_bfc | 5529856 | 0 | 0 | 0 | void im2col4d_kernel<float, int>(im2col4d_params, cudnnConvolutionStruct, cudnnTensor4dStruct, float const*, float*, int) | 15.22 | 0 | 0.00 | 0.00 | 0.00 | 0.00 | 0.00 | true | 0;0;0;0;0;0;0;0;0;0;0;0;0;0;0 |
| 2 | InceptionV2/InceptionV2/Conv2d_1a_7x7/separable_conv2d/depthwise | DepthwiseConv2dNative | [[1 24 112 112]] | 486.333 | 1204224 | 6734080 | 46529024 | GPU_0_bfc | 5529856 | 0 | 0 | 0 | void tensorflow::functor::PadInputCustomKernelNCHW<float, 4>(int, float const*, tensorflow::functor::Dimension<4>, float*, tensorflow::functor::Dimension<4>, tensorflow::functor::Dimension<(4)-(2)>) | 10.00 | 0 | 0.00 | 0.00 | 0.00 | 0.00 | 0.00 | true | 0;0;0;0;0 |
| 2 | InceptionV2/InceptionV2/Conv2d_1a_7x7/separable_conv2d/depthwise | DepthwiseConv2dNative | [[1 24 112 112]] | 486.333 | 1204224 | 6734080 | 46529024 | GPU_0_bfc | 5529856 | 0 | 0 | 0 | void tensorflow::functor::ShuffleInTensor3Simple<float, 2, 1, 0, false>(int, float const*, tensorflow::functor::Dimension<3>, float*) | 4.00 | 0 | 0.00 | 0.00 | 0.00 | 0.00 | 0.00 | true | 0;0;0;0;0 |
| 3 | InceptionV2/InceptionV2/Conv2d_1a_7x7/separable_conv2d | Conv2D | [[1 64 112 112]] | 291.333 | 3211264 | 3292928 | 49138176 | GPU_0_bfc | 81664 | 0 | 0 | 0 | maxwell_scudnn_128x64_relu_interior_nn | 20.33 | 40140800 | 0.00 | 0.00 | 0.00 | 0.00 | 1974.17 | true | 40140800;40140800;40140800;40140800;40140800 |
| 3 | InceptionV2/InceptionV2/Conv2d_1a_7x7/separable_conv2d | Conv2D | [[1 64 112 112]] | 291.333 | 3211264 | 3292928 | 49138176 | GPU_0_bfc | 81664 | 0 | 0 | 0 | void tensorflow::functor::ShuffleInTensor3Simple<float, 2, 1, 0, false>(int, float const*, tensorflow::functor::Dimension<3>, float*) | 5.00 | 0 | 0.00 | 0.00 | 0.00 | 0.00 | 0.00 | true | 0;0;0;0;0 |
| 3 | InceptionV2/InceptionV2/Conv2d_1a_7x7/separable_conv2d | Conv2D | [[1 64 112 112]] | 291.333 | 3211264 | 3292928 | 49138176 | GPU_0_bfc | 81664 | 0 | 0 | 0 | cudnn::maxwell::gemm::computeOffsetsKernel(cudnn::maxwell::gemm::ComputeOffsetsParams) | 3.00 | 0 | 0.00 | 0.00 | 0.00 | 0.00 | 0.00 | true | 0;0;0;0;0 |
| 4 | InceptionV2/InceptionV2/Conv2d_1a_7x7/BiasAdd | BiasAdd | [[1 64 112 112]] | 54.667 | 3211264 | 0 | 47933952 | GPU_0_bfc | 0 | 0 | 0 | 0 | void tensorflow::BiasNCHWKernel<float>(int, float const*, float const*, float*, int, int) | 17.00 | 802816 | 0.00 | 0.00 | 0.00 | 0.00 | 47.22 | true | 802816;802816;802816;802816;802816 |
| 5 | InceptionV2/InceptionV2/Conv2d_1a_7x7/Relu | Relu | [[1 64 112 112]] | 34 | 3211264 | 0 | 47933952 | GPU_0_bfc | 0 | 0 | 0 | 0 | void Eigen::internal::EigenMetaKernel<Eigen::TensorEvaluator<Eigen::TensorAssignOp<Eigen::TensorMap<Eigen::Tensor<float, 1, 1, long>, 16, Eigen::MakePointer>, Eigen::TensorCwiseBinaryOp<Eigen::internal::scalar_max_op<float const, float const>, Eigen::TensorMap<Eigen::Tensor<float const, 1, 1, long>, 16, Eigen::MakePointer> const, Eigen::TensorCwiseNullaryOp<Eigen::internal::scalar_constant_op<float const>, Eigen::TensorMap<Eigen::Tensor<float const, 1, 1, long>, 16, Eigen::MakePointer> const> const> const> const, Eigen::GpuDevice>, long>(Eigen::TensorEvaluator<Eigen::TensorAssignOp<Eigen::TensorMap<Eigen::Tensor<float, 1, 1, long>, 16, Eigen::MakePointer>, Eigen::TensorCwiseBinaryOp<Eigen::internal::scalar_max_op<float const, float const>, Eigen::TensorMap<Eigen::Tensor<float const, 1, 1, long>, 16, Eigen::MakePointer> const, Eigen::TensorCwiseNullaryOp<Eigen::internal::scalar_constant_op<float const>, Eigen::TensorMap<Eigen::Tensor<float const, 1, 1, long>, 16, Eigen::MakePointer> const> const> const> const, Eigen::GpuDevice>, long) | 7.00 | 0 | 0.00 | 0.00 | 0.00 | 0.00 | 0.00 | true | 0;0;0;0;0 |
| 6 | InceptionV2/InceptionV2/MaxPool_2a_3x3/MaxPool | MaxPool | [[1 64 56 56]] | 80.667 | 802816 | 802816 | 48736768 | GPU_0_bfc | 0 | 0 | 0 | 0 | void cudnn::detail::pooling_fw_4d_kernel<float, float, cudnn::detail::maxpooling_func<float, (cudnnNanPropagation_t)0>, 0, false>(cudnnTensorStruct, float const*, cudnnTensorStruct, float*, cudnnPoolingStruct, float, float, int, cudnn::reduced_divisor, cudnn::reduced_divisor) | 19.00 | 200704 | 0.00 | 0.00 | 0.00 | 0.00 | 10.56 | true | 200704;200704;200704;200704;200704 |
| 7 | InceptionV2/InceptionV2/Conv2d_2b_1x1/BatchNorm/batchnorm/mul | Conv2D | [[1 64 56 56]] | 148 | 802816 | 838144 | 46328320 | GPU_0_bfc | 35328 | 0 | 0 | 0 | maxwell_scudnn_128x64_relu_interior_nn | 24.00 | 26624000 | 0.00 | 0.00 | 0.00 | 0.00 | 1109.33 | true | 26624000;26624000;26624000;26624000;26624000 |
| 7 | InceptionV2/InceptionV2/Conv2d_2b_1x1/BatchNorm/batchnorm/mul | Conv2D | [[1 64 56 56]] | 148 | 802816 | 838144 | 46328320 | GPU_0_bfc | 35328 | 0 | 0 | 0 | void tensorflow::functor::ShuffleInTensor3Simple<float, 2, 1, 0, false>(int, float const*, tensorflow::functor::Dimension<3>, float*) | 4.33 | 0 | 0.00 | 0.00 | 0.00 | 0.00 | 0.00 | true | 0;0;0;0;0 |
| 7 | InceptionV2/InceptionV2/Conv2d_2b_1x1/BatchNorm/batchnorm/mul | Conv2D | [[1 64 56 56]] | 148 | 802816 | 838144 | 46328320 | GPU_0_bfc | 35328 | 0 | 0 | 0 | cudnn::maxwell::gemm::computeOffsetsKernel(cudnn::maxwell::gemm::ComputeOffsetsParams) | 2.50 | 0 | 0.00 | 0.00 | 0.00 | 0.00 | 0.00 | true | 0;0;0;0;0 |
| 8 | InceptionV2/InceptionV2/Conv2d_2b_1x1/BatchNorm/batchnorm/add_1 | Add | [[1 64 56 56]] | 47.333 | 802816 | 0 | 45525504 | GPU_0_bfc | 0 | 0 | 0 | 0 | void Eigen::internal::EigenMetaKernel<Eigen::TensorEvaluator<Eigen::TensorAssignOp<Eigen::TensorMap<Eigen::Tensor<float, 2, 1, int>, 16, Eigen::MakePointer>, Eigen::TensorCwiseBinaryOp<Eigen::internal::scalar_sum_op<float, float>, Eigen::TensorMap<Eigen::Tensor<float const, 2, 1, int>, 16, Eigen::MakePointer> const, Eigen::TensorBroadcastingOp<Eigen::array<long, 2ul> const, Eigen::TensorMap<Eigen::Tensor<float const, 2, 1, int>, 16, Eigen::MakePointer> const> const> const> const, Eigen::GpuDevice>, int>(Eigen::TensorEvaluator<Eigen::TensorAssignOp<Eigen::TensorMap<Eigen::Tensor<float, 2, 1, int>, 16, Eigen::MakePointer>, Eigen::TensorCwiseBinaryOp<Eigen::internal::scalar_sum_op<float, float>, Eigen::TensorMap<Eigen::Tensor<float const, 2, 1, int>, 16, Eigen::MakePointer> const, Eigen::TensorBroadcastingOp<Eigen::array<long, 2ul> const, Eigen::TensorMap<Eigen::Tensor<float const, 2, 1, int>, 16, Eigen::MakePointer> const> const> const> const, Eigen::GpuDevice>, int) | 4.50 | 200704 | 0.00 | 0.00 | 0.00 | 0.00 | 44.60 | true | 200704;200704;200704;200704;200704 |
| 9 | InceptionV2/InceptionV2/Conv2d_2b_1x1/Relu | Relu | [[1 64 56 56]] | 27.667 | 802816 | 0 | 45525504 | GPU_0_bfc | 0 | 0 | 0 | 0 | void Eigen::internal::EigenMetaKernel<Eigen::TensorEvaluator<Eigen::TensorAssignOp<Eigen::TensorMap<Eigen::Tensor<float, 1, 1, long>, 16, Eigen::MakePointer>, Eigen::TensorCwiseBinaryOp<Eigen::internal::scalar_max_op<float const, float const>, Eigen::TensorMap<Eigen::Tensor<float const, 1, 1, long>, 16, Eigen::MakePointer> const, Eigen::TensorCwiseNullaryOp<Eigen::internal::scalar_constant_op<float const>, Eigen::TensorMap<Eigen::Tensor<float const, 1, 1, long>, 16, Eigen::MakePointer> const> const> const> const, Eigen::GpuDevice>, long>(Eigen::TensorEvaluator<Eigen::TensorAssignOp<Eigen::TensorMap<Eigen::Tensor<float, 1, 1, long>, 16, Eigen::MakePointer>, Eigen::TensorCwiseBinaryOp<Eigen::internal::scalar_max_op<float const, float const>, Eigen::TensorMap<Eigen::Tensor<float const, 1, 1, long>, 16, Eigen::MakePointer> const, Eigen::TensorCwiseNullaryOp<Eigen::internal::scalar_constant_op<float const>, Eigen::TensorMap<Eigen::Tensor<float const, 1, 1, long>, 16, Eigen::MakePointer> const> const> const> const, Eigen::GpuDevice>, long) | 4.00 | 0 | 0.00 | 0.00 | 0.00 | 0.00 | 0.00 | true | 0;0;0;0;0 |
| 10 | InceptionV2/InceptionV2/Conv2d_2c_3x3/BatchNorm/batchnorm/mul | Conv2D | [[1 192 56 56]] | 287.667 | 2408448 | 4440832 | 47933952 | GPU_0_bfc | 2032384 | 0 | 0 | 0 | maxwell_scudnn_winograd_128x128_ldg1_ldg4_tile148n_nt | 77.00 | 372621312 | 0.00 | 0.00 | 0.00 | 0.00 | 4839.24 | true | 372621312;372621312;372621312;372621312;372621312 |
| 10 | InceptionV2/InceptionV2/Conv2d_2c_3x3/BatchNorm/batchnorm/mul | Conv2D | [[1 192 56 56]] | 287.667 | 2408448 | 4440832 | 47933952 | GPU_0_bfc | 2032384 | 0 | 0 | 0 | void tensorflow::functor::ShuffleInTensor3Simple<float, 2, 1, 0, false>(int, float const*, tensorflow::functor::Dimension<3>, float*) | 8.00 | 0 | 0.00 | 0.00 | 0.00 | 0.00 | 0.00 | true | 0;0;0;0;0 |
| 10 | InceptionV2/InceptionV2/Conv2d_2c_3x3/BatchNorm/batchnorm/mul | Conv2D | [[1 192 56 56]] | 287.667 | 2408448 | 4440832 | 47933952 | GPU_0_bfc | 2032384 | 0 | 0 | 0 | void cudnn::winograd::generateWinogradTilesKernel<0, float, float>(cudnn::winograd::GenerateWinogradTilesParams<float, float>) | 7.00 | 712704 | 0.00 | 0.00 | 0.00 | 0.00 | 101.81 | true | 712704;712704;712704;712704;712704 |
| 11 | InceptionV2/InceptionV2/Conv2d_2c_3x3/BatchNorm/batchnorm/add_1 | Add | [[1 192 56 56]] | 40.333 | 2408448 | 0 | 47131136 | GPU_0_bfc | 0 | 0 | 0 | 0 | void Eigen::internal::EigenMetaKernel<Eigen::TensorEvaluator<Eigen::TensorAssignOp<Eigen::TensorMap<Eigen::Tensor<float, 2, 1, int>, 16, Eigen::MakePointer>, Eigen::TensorCwiseBinaryOp<Eigen::internal::scalar_sum_op<float, float>, Eigen::TensorMap<Eigen::Tensor<float const, 2, 1, int>, 16, Eigen::MakePointer> const, Eigen::TensorBroadcastingOp<Eigen::array<long, 2ul> const, Eigen::TensorMap<Eigen::Tensor<float const, 2, 1, int>, 16, Eigen::MakePointer> const> const> const> const, Eigen::GpuDevice>, int>(Eigen::TensorEvaluator<Eigen::TensorAssignOp<Eigen::TensorMap<Eigen::Tensor<float, 2, 1, int>, 16, Eigen::MakePointer>, Eigen::TensorCwiseBinaryOp<Eigen::internal::scalar_sum_op<float, float>, Eigen::TensorMap<Eigen::Tensor<float const, 2, 1, int>, 16, Eigen::MakePointer> const, Eigen::TensorBroadcastingOp<Eigen::array<long, 2ul> const, Eigen::TensorMap<Eigen::Tensor<float const, 2, 1, int>, 16, Eigen::MakePointer> const> const> const> const, Eigen::GpuDevice>, int) | 8.00 | 602112 | 0.00 | 0.00 | 0.00 | 0.00 | 75.26 | true | 602112;602112;602112;602112;602112 |
| 12 | InceptionV2/InceptionV2/Conv2d_2c_3x3/Relu | Relu | [[1 192 56 56]] | 31.667 | 2408448 | 0 | 47131136 | GPU_0_bfc | 0 | 0 | 0 | 0 | void Eigen::internal::EigenMetaKernel<Eigen::TensorEvaluator<Eigen::TensorAssignOp<Eigen::TensorMap<Eigen::Tensor<float, 1, 1, long>, 16, Eigen::MakePointer>, Eigen::TensorCwiseBinaryOp<Eigen::internal::scalar_max_op<float const, float const>, Eigen::TensorMap<Eigen::Tensor<float const, 1, 1, long>, 16, Eigen::MakePointer> const, Eigen::TensorCwiseNullaryOp<Eigen::internal::scalar_constant_op<float const>, Eigen::TensorMap<Eigen::Tensor<float const, 1, 1, long>, 16, Eigen::MakePointer> const> const> const> const, Eigen::GpuDevice>, long>(Eigen::TensorEvaluator<Eigen::TensorAssignOp<Eigen::TensorMap<Eigen::Tensor<float, 1, 1, long>, 16, Eigen::MakePointer>, Eigen::TensorCwiseBinaryOp<Eigen::internal::scalar_max_op<float const, float const>, Eigen::TensorMap<Eigen::Tensor<float const, 1, 1, long>, 16, Eigen::MakePointer> const, Eigen::TensorCwiseNullaryOp<Eigen::internal::scalar_constant_op<float const>, Eigen::TensorMap<Eigen::Tensor<float const, 1, 1, long>, 16, Eigen::MakePointer> const> const> const> const, Eigen::GpuDevice>, long) | 6.00 | 0 | 0.00 | 0.00 | 0.00 | 0.00 | 0.00 | true | 0;0;0;0;0 |
| 13 | InceptionV2/InceptionV2/MaxPool_3a_3x3/MaxPool | MaxPool | [[1 192 28 28]] | 65.667 | 602112 | 602112 | 47733248 | GPU_0_bfc | 0 | 0 | 0 | 0 | void cudnn::detail::pooling_fw_4d_kernel<float, float, cudnn::detail::maxpooling_func<float, (cudnnNanPropagation_t)0>, 0, false>(cudnnTensorStruct, float const*, cudnnTensorStruct, float*, cudnnPoolingStruct, float, float, int, cudnn::reduced_divisor, cudnn::reduced_divisor) | 16.00 | 150528 | 0.00 | 0.00 | 0.00 | 0.00 | 9.41 | true | 150528;150528;150528;150528;150528 |
| 14 | InceptionV2/InceptionV2/Mixed_3b/Branch_3/AvgPool_0a_3x3/AvgPool | AvgPool | [[1 192 28 28]] | 65.333 | 602112 | 602112 | 45926912 | GPU_0_bfc | 0 | 0 | 0 | 0 | void cudnn::detail::pooling_fw_4d_kernel<float, float, cudnn::detail::averpooling_func<float>, 2, false>(cudnnTensorStruct, float const*, cudnnTensorStruct, float*, cudnnPoolingStruct, float, float, int, cudnn::reduced_divisor, cudnn::reduced_divisor) | 19.00 | 3956397 | 0.00 | 0.00 | 0.00 | 0.00 | 208.23 | true | 3956397;3956397;3956397;3956397;3956397 |
| 15 | InceptionV2/InceptionV2/Mixed_3b/Branch_2/Conv2d_0a_1x1/BatchNorm/batchnorm/mul | Conv2D | [[1 64 28 28]] | 169.333 | 200704 | 254720 | 46127616 | GPU_0_bfc | 54016 | 0 | 0 | 0 | maxwell_scudnn_128x64_relu_interior_nn | 51.00 | 22134784 | 0.00 | 0.00 | 0.00 | 0.00 | 434.02 | true | 22134784;22134784;22134784;22134784;22134784 |
| 15 | InceptionV2/InceptionV2/Mixed_3b/Branch_2/Conv2d_0a_1x1/BatchNorm/batchnorm/mul | Conv2D | [[1 64 28 28]] | 169.333 | 200704 | 254720 | 46127616 | GPU_0_bfc | 54016 | 0 | 0 | 0 | void tensorflow::functor::ShuffleInTensor3Simple<float, 2, 1, 0, false>(int, float const*, tensorflow::functor::Dimension<3>, float*) | 4.50 | 0 | 0.00 | 0.00 | 0.00 | 0.00 | 0.00 | true | 0;0;0;0;0 |
| 15 | InceptionV2/InceptionV2/Mixed_3b/Branch_2/Conv2d_0a_1x1/BatchNorm/batchnorm/mul | Conv2D | [[1 64 28 28]] | 169.333 | 200704 | 254720 | 46127616 | GPU_0_bfc | 54016 | 0 | 0 | 0 | cudnn::maxwell::gemm::computeOffsetsKernel(cudnn::maxwell::gemm::ComputeOffsetsParams) | 2.83 | 0 | 0.00 | 0.00 | 0.00 | 0.00 | 0.00 | true | 0;0;0;0;0 |
| 16 | InceptionV2/InceptionV2/Mixed_3b/Branch_1/Conv2d_0a_1x1/BatchNorm/batchnorm/mul | Conv2D | [[1 64 28 28]] | 164.667 | 200704 | 254720 | 46328320 | GPU_0_bfc | 54016 | 0 | 0 | 0 | maxwell_scudnn_128x64_relu_interior_nn | 51.00 | 22134784 | 0.00 | 0.00 | 0.00 | 0.00 | 434.02 | true | 22134784;22134784;22134784;22134784;22134784 |
| 16 | InceptionV2/InceptionV2/Mixed_3b/Branch_1/Conv2d_0a_1x1/BatchNorm/batchnorm/mul | Conv2D | [[1 64 28 28]] | 164.667 | 200704 | 254720 | 46328320 | GPU_0_bfc | 54016 | 0 | 0 | 0 | void tensorflow::functor::ShuffleInTensor3Simple<float, 2, 1, 0, false>(int, float const*, tensorflow::functor::Dimension<3>, float*) | 5.00 | 0 | 0.00 | 0.00 | 0.00 | 0.00 | 0.00 | true | 0;0;0;0;0 |
| 16 | InceptionV2/InceptionV2/Mixed_3b/Branch_1/Conv2d_0a_1x1/BatchNorm/batchnorm/mul | Conv2D | [[1 64 28 28]] | 164.667 | 200704 | 254720 | 46328320 | GPU_0_bfc | 54016 | 0 | 0 | 0 | cudnn::maxwell::gemm::computeOffsetsKernel(cudnn::maxwell::gemm::ComputeOffsetsParams) | 2.50 | 0 | 0.00 | 0.00 | 0.00 | 0.00 | 0.00 | true | 0;0;0;0;0 |
| 17 | InceptionV2/InceptionV2/Mixed_3b/Branch_0/Conv2d_0a_1x1/BatchNorm/batchnorm/mul | Conv2D | [[1 64 28 28]] | 164.333 | 200704 | 254720 | 46529024 | GPU_0_bfc | 54016 | 0 | 0 | 0 | maxwell_scudnn_128x64_relu_interior_nn | 51.00 | 22134784 | 0.00 | 0.00 | 0.00 | 0.00 | 434.02 | true | 22134784;22134784;22134784;22134784;22134784 |
| 17 | InceptionV2/InceptionV2/Mixed_3b/Branch_0/Conv2d_0a_1x1/BatchNorm/batchnorm/mul | Conv2D | [[1 64 28 28]] | 164.333 | 200704 | 254720 | 46529024 | GPU_0_bfc | 54016 | 0 | 0 | 0 | void tensorflow::functor::ShuffleInTensor3Simple<float, 2, 1, 0, false>(int, float const*, tensorflow::functor::Dimension<3>, float*) | 5.00 | 0 | 0.00 | 0.00 | 0.00 | 0.00 | 0.00 | true | 0;0;0;0;0 |
| 17 | InceptionV2/InceptionV2/Mixed_3b/Branch_0/Conv2d_0a_1x1/BatchNorm/batchnorm/mul | Conv2D | [[1 64 28 28]] | 164.333 | 200704 | 254720 | 46529024 | GPU_0_bfc | 54016 | 0 | 0 | 0 | cudnn::maxwell::gemm::computeOffsetsKernel(cudnn::maxwell::gemm::ComputeOffsetsParams) | 2.50 | 0 | 0.00 | 0.00 | 0.00 | 0.00 | 0.00 | true | 0;0;0;0;0 |
| 18 | InceptionV2/InceptionV2/Mixed_3b/Branch_3/Conv2d_0b_1x1/BatchNorm/batchnorm/mul | Conv2D | [[1 32 28 28]] | 157 | 100352 | 129792 | 46027264 | GPU_0_bfc | 29440 | 0 | 0 | 0 | maxwell_scudnn_128x32_relu_interior_nn | 40.67 | 11067392 | 0.00 | 0.00 | 0.00 | 0.00 | 272.15 | true | 11067392;11067392;11067392;11067392;11067392 |
| 18 | InceptionV2/InceptionV2/Mixed_3b/Branch_3/Conv2d_0b_1x1/BatchNorm/batchnorm/mul | Conv2D | [[1 32 28 28]] | 157 | 100352 | 129792 | 46027264 | GPU_0_bfc | 29440 | 0 | 0 | 0 | void tensorflow::functor::ShuffleInTensor3Simple<float, 2, 1, 0, false>(int, float const*, tensorflow::functor::Dimension<3>, float*) | 5.00 | 0 | 0.00 | 0.00 | 0.00 | 0.00 | 0.00 | true | 0;0;0;0;0 |
| 18 | InceptionV2/InceptionV2/Mixed_3b/Branch_3/Conv2d_0b_1x1/BatchNorm/batchnorm/mul | Conv2D | [[1 32 28 28]] | 157 | 100352 | 129792 | 46027264 | GPU_0_bfc | 29440 | 0 | 0 | 0 | cudnn::maxwell::gemm::computeOffsetsKernel(cudnn::maxwell::gemm::ComputeOffsetsParams) | 2.67 | 0 | 0.00 | 0.00 | 0.00 | 0.00 | 0.00 | true | 0;0;0;0;0 |
| 19 | InceptionV2/InceptionV2/Mixed_3b/Branch_2/Conv2d_0a_1x1/BatchNorm/batchnorm/add_1 | Add | [[1 64 28 28]] | 37 | 200704 | 0 | 45425152 | GPU_0_bfc | 0 | 0 | 0 | 0 | void Eigen::internal::EigenMetaKernel<Eigen::TensorEvaluator<Eigen::TensorAssignOp<Eigen::TensorMap<Eigen::Tensor<float, 2, 1, int>, 16, Eigen::MakePointer>, Eigen::TensorCwiseBinaryOp<Eigen::internal::scalar_sum_op<float, float>, Eigen::TensorMap<Eigen::Tensor<float const, 2, 1, int>, 16, Eigen::MakePointer> const, Eigen::TensorBroadcastingOp<Eigen::array<long, 2ul> const, Eigen::TensorMap<Eigen::Tensor<float const, 2, 1, int>, 16, Eigen::MakePointer> const> const> const> const, Eigen::GpuDevice>, int>(Eigen::TensorEvaluator<Eigen::TensorAssignOp<Eigen::TensorMap<Eigen::Tensor<float, 2, 1, int>, 16, Eigen::MakePointer>, Eigen::TensorCwiseBinaryOp<Eigen::internal::scalar_sum_op<float, float>, Eigen::TensorMap<Eigen::Tensor<float const, 2, 1, int>, 16, Eigen::MakePointer> const, Eigen::TensorBroadcastingOp<Eigen::array<long, 2ul> const, Eigen::TensorMap<Eigen::Tensor<float const, 2, 1, int>, 16, Eigen::MakePointer> const> const> const> const, Eigen::GpuDevice>, int) | 4.00 | 50176 | 0.00 | 0.00 | 0.00 | 0.00 | 12.54 | true | 50176;50176;50176;50176;50176 |
| 20 | InceptionV2/InceptionV2/Mixed_3b/Branch_1/Conv2d_0a_1x1/BatchNorm/batchnorm/add_1 | Add | [[1 64 28 28]] | 29.333 | 200704 | 0 | 45425152 | GPU_0_bfc | 0 | 0 | 0 | 0 | void Eigen::internal::EigenMetaKernel<Eigen::TensorEvaluator<Eigen::TensorAssignOp<Eigen::TensorMap<Eigen::Tensor<float, 2, 1, int>, 16, Eigen::MakePointer>, Eigen::TensorCwiseBinaryOp<Eigen::internal::scalar_sum_op<float, float>, Eigen::TensorMap<Eigen::Tensor<float const, 2, 1, int>, 16, Eigen::MakePointer> const, Eigen::TensorBroadcastingOp<Eigen::array<long, 2ul> const, Eigen::TensorMap<Eigen::Tensor<float const, 2, 1, int>, 16, Eigen::MakePointer> const> const> const> const, Eigen::GpuDevice>, int>(Eigen::TensorEvaluator<Eigen::TensorAssignOp<Eigen::TensorMap<Eigen::Tensor<float, 2, 1, int>, 16, Eigen::MakePointer>, Eigen::TensorCwiseBinaryOp<Eigen::internal::scalar_sum_op<float, float>, Eigen::TensorMap<Eigen::Tensor<float const, 2, 1, int>, 16, Eigen::MakePointer> const, Eigen::TensorBroadcastingOp<Eigen::array<long, 2ul> const, Eigen::TensorMap<Eigen::Tensor<float const, 2, 1, int>, 16, Eigen::MakePointer> const> const> const> const, Eigen::GpuDevice>, int) | 4.00 | 50176 | 0.00 | 0.00 | 0.00 | 0.00 | 12.54 | true | 50176;50176;50176;50176;50176 |
| 21 | InceptionV2/InceptionV2/Mixed_3b/Branch_0/Conv2d_0a_1x1/BatchNorm/batchnorm/add_1 | Add | [[1 64 28 28]] | 32 | 200704 | 0 | 45425152 | GPU_0_bfc | 0 | 0 | 0 | 0 | void Eigen::internal::EigenMetaKernel<Eigen::TensorEvaluator<Eigen::TensorAssignOp<Eigen::TensorMap<Eigen::Tensor<float, 2, 1, int>, 16, Eigen::MakePointer>, Eigen::TensorCwiseBinaryOp<Eigen::internal::scalar_sum_op<float, float>, Eigen::TensorMap<Eigen::Tensor<float const, 2, 1, int>, 16, Eigen::MakePointer> const, Eigen::TensorBroadcastingOp<Eigen::array<long, 2ul> const, Eigen::TensorMap<Eigen::Tensor<float const, 2, 1, int>, 16, Eigen::MakePointer> const> const> const> const, Eigen::GpuDevice>, int>(Eigen::TensorEvaluator<Eigen::TensorAssignOp<Eigen::TensorMap<Eigen::Tensor<float, 2, 1, int>, 16, Eigen::MakePointer>, Eigen::TensorCwiseBinaryOp<Eigen::internal::scalar_sum_op<float, float>, Eigen::TensorMap<Eigen::Tensor<float const, 2, 1, int>, 16, Eigen::MakePointer> const, Eigen::TensorBroadcastingOp<Eigen::array<long, 2ul> const, Eigen::TensorMap<Eigen::Tensor<float const, 2, 1, int>, 16, Eigen::MakePointer> const> const> const> const, Eigen::GpuDevice>, int) | 4.00 | 50176 | 0.00 | 0.00 | 0.00 | 0.00 | 12.54 | true | 50176;50176;50176;50176;50176 |
| 22 | InceptionV2/InceptionV2/Mixed_3b/Branch_3/Conv2d_0b_1x1/BatchNorm/batchnorm/add_1 | Add | [[1 32 28 28]] | 28 | 100352 | 0 | 45425152 | GPU_0_bfc | 0 | 0 | 0 | 0 | void Eigen::internal::EigenMetaKernel<Eigen::TensorEvaluator<Eigen::TensorAssignOp<Eigen::TensorMap<Eigen::Tensor<float, 2, 1, int>, 16, Eigen::MakePointer>, Eigen::TensorCwiseBinaryOp<Eigen::internal::scalar_sum_op<float, float>, Eigen::TensorMap<Eigen::Tensor<float const, 2, 1, int>, 16, Eigen::MakePointer> const, Eigen::TensorBroadcastingOp<Eigen::array<long, 2ul> const, Eigen::TensorMap<Eigen::Tensor<float const, 2, 1, int>, 16, Eigen::MakePointer> const> const> const> const, Eigen::GpuDevice>, int>(Eigen::TensorEvaluator<Eigen::TensorAssignOp<Eigen::TensorMap<Eigen::Tensor<float, 2, 1, int>, 16, Eigen::MakePointer>, Eigen::TensorCwiseBinaryOp<Eigen::internal::scalar_sum_op<float, float>, Eigen::TensorMap<Eigen::Tensor<float const, 2, 1, int>, 16, Eigen::MakePointer> const, Eigen::TensorBroadcastingOp<Eigen::array<long, 2ul> const, Eigen::TensorMap<Eigen::Tensor<float const, 2, 1, int>, 16, Eigen::MakePointer> const> const> const> const, Eigen::GpuDevice>, int) | 4.00 | 25088 | 0.00 | 0.00 | 0.00 | 0.00 | 6.27 | true | 25088;25088;25088;25088;25088 |
| 23 | InceptionV2/InceptionV2/Mixed_3b/Branch_2/Conv2d_0a_1x1/Relu | Relu | [[1 64 28 28]] | 26.667 | 200704 | 0 | 45425152 | GPU_0_bfc | 0 | 0 | 0 | 0 | void Eigen::internal::EigenMetaKernel<Eigen::TensorEvaluator<Eigen::TensorAssignOp<Eigen::TensorMap<Eigen::Tensor<float, 1, 1, long>, 16, Eigen::MakePointer>, Eigen::TensorCwiseBinaryOp<Eigen::internal::scalar_max_op<float const, float const>, Eigen::TensorMap<Eigen::Tensor<float const, 1, 1, long>, 16, Eigen::MakePointer> const, Eigen::TensorCwiseNullaryOp<Eigen::internal::scalar_constant_op<float const>, Eigen::TensorMap<Eigen::Tensor<float const, 1, 1, long>, 16, Eigen::MakePointer> const> const> const> const, Eigen::GpuDevice>, long>(Eigen::TensorEvaluator<Eigen::TensorAssignOp<Eigen::TensorMap<Eigen::Tensor<float, 1, 1, long>, 16, Eigen::MakePointer>, Eigen::TensorCwiseBinaryOp<Eigen::internal::scalar_max_op<float const, float const>, Eigen::TensorMap<Eigen::Tensor<float const, 1, 1, long>, 16, Eigen::MakePointer> const, Eigen::TensorCwiseNullaryOp<Eigen::internal::scalar_constant_op<float const>, Eigen::TensorMap<Eigen::Tensor<float const, 1, 1, long>, 16, Eigen::MakePointer> const> const> const> const, Eigen::GpuDevice>, long) | 3.00 | 0 | 0.00 | 0.00 | 0.00 | 0.00 | 0.00 | true | 0;0;0;0;0 |
| 24 | InceptionV2/InceptionV2/Mixed_3b/Branch_1/Conv2d_0a_1x1/Relu | Relu | [[1 64 28 28]] | 24.333 | 200704 | 0 | 45425152 | GPU_0_bfc | 0 | 0 | 0 | 0 | void Eigen::internal::EigenMetaKernel<Eigen::TensorEvaluator<Eigen::TensorAssignOp<Eigen::TensorMap<Eigen::Tensor<float, 1, 1, long>, 16, Eigen::MakePointer>, Eigen::TensorCwiseBinaryOp<Eigen::internal::scalar_max_op<float const, float const>, Eigen::TensorMap<Eigen::Tensor<float const, 1, 1, long>, 16, Eigen::MakePointer> const, Eigen::TensorCwiseNullaryOp<Eigen::internal::scalar_constant_op<float const>, Eigen::TensorMap<Eigen::Tensor<float const, 1, 1, long>, 16, Eigen::MakePointer> const> const> const> const, Eigen::GpuDevice>, long>(Eigen::TensorEvaluator<Eigen::TensorAssignOp<Eigen::TensorMap<Eigen::Tensor<float, 1, 1, long>, 16, Eigen::MakePointer>, Eigen::TensorCwiseBinaryOp<Eigen::internal::scalar_max_op<float const, float const>, Eigen::TensorMap<Eigen::Tensor<float const, 1, 1, long>, 16, Eigen::MakePointer> const, Eigen::TensorCwiseNullaryOp<Eigen::internal::scalar_constant_op<float const>, Eigen::TensorMap<Eigen::Tensor<float const, 1, 1, long>, 16, Eigen::MakePointer> const> const> const> const, Eigen::GpuDevice>, long) | 3.00 | 0 | 0.00 | 0.00 | 0.00 | 0.00 | 0.00 | true | 0;0;0;0;0 |
| 25 | InceptionV2/InceptionV2/Mixed_3b/Branch_2/Conv2d_0b_3x3/BatchNorm/batchnorm/mul | Conv2D | [[1 96 28 28]] | 198 | 301056 | 1137152 | 45726208 | GPU_0_bfc | 836096 | 0 | 0 | 0 | maxwell_scudnn_winograd_128x128_ldg1_ldg4_tile148n_nt | 26.00 | 53231616 | 0.00 | 0.00 | 0.00 | 0.00 | 2047.37 | true | 53231616;53231616;53231616;53231616;53231616 |
| 25 | InceptionV2/InceptionV2/Mixed_3b/Branch_2/Conv2d_0b_3x3/BatchNorm/batchnorm/mul | Conv2D | [[1 96 28 28]] | 198 | 301056 | 1137152 | 45726208 | GPU_0_bfc | 836096 | 0 | 0 | 0 | void tensorflow::functor::ShuffleInTensor3Simple<float, 2, 1, 0, false>(int, float const*, tensorflow::functor::Dimension<3>, float*) | 5.83 | 0 | 0.00 | 0.00 | 0.00 | 0.00 | 0.00 | true | 0;0;0;0;0 |
| 25 | InceptionV2/InceptionV2/Mixed_3b/Branch_2/Conv2d_0b_3x3/BatchNorm/batchnorm/mul | Conv2D | [[1 96 28 28]] | 198 | 301056 | 1137152 | 45726208 | GPU_0_bfc | 836096 | 0 | 0 | 0 | void cudnn::winograd::generateWinogradTilesKernel<0, float, float>(cudnn::winograd::GenerateWinogradTilesParams<float, float>) | 4.00 | 356352 | 0.00 | 0.00 | 0.00 | 0.00 | 89.09 | true | 356352;356352;356352;356352;356352 |
| 26 | InceptionV2/InceptionV2/Mixed_3b/Branch_1/Conv2d_0b_3x3/BatchNorm/batchnorm/mul | Conv2D | [[1 64 28 28]] | 194.667 | 200704 | 1003520 | 45726208 | GPU_0_bfc | 802816 | 0 | 0 | 0 | maxwell_scudnn_winograd_128x128_ldg1_ldg4_tile148n_nt | 25.83 | 35487744 | 0.00 | 0.00 | 0.00 | 0.00 | 1373.74 | true | 35487744;35487744;35487744;35487744;35487744 |
| 26 | InceptionV2/InceptionV2/Mixed_3b/Branch_1/Conv2d_0b_3x3/BatchNorm/batchnorm/mul | Conv2D | [[1 64 28 28]] | 194.667 | 200704 | 1003520 | 45726208 | GPU_0_bfc | 802816 | 0 | 0 | 0 | void tensorflow::functor::ShuffleInTensor3Simple<float, 2, 1, 0, false>(int, float const*, tensorflow::functor::Dimension<3>, float*) | 5.00 | 0 | 0.00 | 0.00 | 0.00 | 0.00 | 0.00 | true | 0;0;0;0;0 |
| 26 | InceptionV2/InceptionV2/Mixed_3b/Branch_1/Conv2d_0b_3x3/BatchNorm/batchnorm/mul | Conv2D | [[1 64 28 28]] | 194.667 | 200704 | 1003520 | 45726208 | GPU_0_bfc | 802816 | 0 | 0 | 0 | void cudnn::winograd::generateWinogradTilesKernel<0, float, float>(cudnn::winograd::GenerateWinogradTilesParams<float, float>) | 4.00 | 237568 | 0.00 | 0.00 | 0.00 | 0.00 | 59.39 | true | 237568;237568;237568;237568;237568 |
| 27 | InceptionV2/InceptionV2/Mixed_3b/Branch_2/Conv2d_0b_3x3/BatchNorm/batchnorm/add_1 | Add | [[1 96 28 28]] | 35.667 | 301056 | 0 | 45525504 | GPU_0_bfc | 0 | 0 | 0 | 0 | void Eigen::internal::EigenMetaKernel<Eigen::TensorEvaluator<Eigen::TensorAssignOp<Eigen::TensorMap<Eigen::Tensor<float, 2, 1, int>, 16, Eigen::MakePointer>, Eigen::TensorCwiseBinaryOp<Eigen::internal::scalar_sum_op<float, float>, Eigen::TensorMap<Eigen::Tensor<float const, 2, 1, int>, 16, Eigen::MakePointer> const, Eigen::TensorBroadcastingOp<Eigen::array<long, 2ul> const, Eigen::TensorMap<Eigen::Tensor<float const, 2, 1, int>, 16, Eigen::MakePointer> const> const> const> const, Eigen::GpuDevice>, int>(Eigen::TensorEvaluator<Eigen::TensorAssignOp<Eigen::TensorMap<Eigen::Tensor<float, 2, 1, int>, 16, Eigen::MakePointer>, Eigen::TensorCwiseBinaryOp<Eigen::internal::scalar_sum_op<float, float>, Eigen::TensorMap<Eigen::Tensor<float const, 2, 1, int>, 16, Eigen::MakePointer> const, Eigen::TensorBroadcastingOp<Eigen::array<long, 2ul> const, Eigen::TensorMap<Eigen::Tensor<float const, 2, 1, int>, 16, Eigen::MakePointer> const> const> const> const, Eigen::GpuDevice>, int) | 4.00 | 75264 | 0.00 | 0.00 | 0.00 | 0.00 | 18.82 | true | 75264;75264;75264;75264;75264 |
| 28 | InceptionV2/InceptionV2/Mixed_3b/Branch_1/Conv2d_0b_3x3/BatchNorm/batchnorm/add_1 | Add | [[1 64 28 28]] | 28.667 | 200704 | 0 | 45525504 | GPU_0_bfc | 0 | 0 | 0 | 0 | void Eigen::internal::EigenMetaKernel<Eigen::TensorEvaluator<Eigen::TensorAssignOp<Eigen::TensorMap<Eigen::Tensor<float, 2, 1, int>, 16, Eigen::MakePointer>, Eigen::TensorCwiseBinaryOp<Eigen::internal::scalar_sum_op<float, float>, Eigen::TensorMap<Eigen::Tensor<float const, 2, 1, int>, 16, Eigen::MakePointer> const, Eigen::TensorBroadcastingOp<Eigen::array<long, 2ul> const, Eigen::TensorMap<Eigen::Tensor<float const, 2, 1, int>, 16, Eigen::MakePointer> const> const> const> const, Eigen::GpuDevice>, int>(Eigen::TensorEvaluator<Eigen::TensorAssignOp<Eigen::TensorMap<Eigen::Tensor<float, 2, 1, int>, 16, Eigen::MakePointer>, Eigen::TensorCwiseBinaryOp<Eigen::internal::scalar_sum_op<float, float>, Eigen::TensorMap<Eigen::Tensor<float const, 2, 1, int>, 16, Eigen::MakePointer> const, Eigen::TensorBroadcastingOp<Eigen::array<long, 2ul> const, Eigen::TensorMap<Eigen::Tensor<float const, 2, 1, int>, 16, Eigen::MakePointer> const> const> const> const, Eigen::GpuDevice>, int) | 4.00 | 50176 | 0.00 | 0.00 | 0.00 | 0.00 | 12.54 | true | 50176;50176;50176;50176;50176 |
| 29 | InceptionV2/InceptionV2/Mixed_3b/Branch_2/Conv2d_0b_3x3/Relu | Relu | [[1 96 28 28]] | 27.667 | 301056 | 0 | 45525504 | GPU_0_bfc | 0 | 0 | 0 | 0 | void Eigen::internal::EigenMetaKernel<Eigen::TensorEvaluator<Eigen::TensorAssignOp<Eigen::TensorMap<Eigen::Tensor<float, 1, 1, long>, 16, Eigen::MakePointer>, Eigen::TensorCwiseBinaryOp<Eigen::internal::scalar_max_op<float const, float const>, Eigen::TensorMap<Eigen::Tensor<float const, 1, 1, long>, 16, Eigen::MakePointer> const, Eigen::TensorCwiseNullaryOp<Eigen::internal::scalar_constant_op<float const>, Eigen::TensorMap<Eigen::Tensor<float const, 1, 1, long>, 16, Eigen::MakePointer> const> const> const> const, Eigen::GpuDevice>, long>(Eigen::TensorEvaluator<Eigen::TensorAssignOp<Eigen::TensorMap<Eigen::Tensor<float, 1, 1, long>, 16, Eigen::MakePointer>, Eigen::TensorCwiseBinaryOp<Eigen::internal::scalar_max_op<float const, float const>, Eigen::TensorMap<Eigen::Tensor<float const, 1, 1, long>, 16, Eigen::MakePointer> const, Eigen::TensorCwiseNullaryOp<Eigen::internal::scalar_constant_op<float const>, Eigen::TensorMap<Eigen::Tensor<float const, 1, 1, long>, 16, Eigen::MakePointer> const> const> const> const, Eigen::GpuDevice>, long) | 3.00 | 0 | 0.00 | 0.00 | 0.00 | 0.00 | 0.00 | true | 0;0;0;0;0 |
| 30 | InceptionV2/InceptionV2/Mixed_3b/Branch_2/Conv2d_0c_3x3/BatchNorm/batchnorm/mul | Conv2D | [[1 96 28 28]] | 204 | 301056 | 1554944 | 45826560 | GPU_0_bfc | 1253888 | 0 | 0 | 0 | maxwell_scudnn_winograd_128x128_ldg1_ldg4_tile148n_nt | 34.00 | 79380480 | 0.00 | 0.00 | 0.00 | 0.00 | 2334.72 | true | 79380480;79380480;79380480;79380480;79380480 |
| 30 | InceptionV2/InceptionV2/Mixed_3b/Branch_2/Conv2d_0c_3x3/BatchNorm/batchnorm/mul | Conv2D | [[1 96 28 28]] | 204 | 301056 | 1554944 | 45826560 | GPU_0_bfc | 1253888 | 0 | 0 | 0 | void tensorflow::functor::ShuffleInTensor3Simple<float, 2, 1, 0, false>(int, float const*, tensorflow::functor::Dimension<3>, float*) | 7.00 | 0 | 0.00 | 0.00 | 0.00 | 0.00 | 0.00 | true | 0;0;0;0;0 |
| 30 | InceptionV2/InceptionV2/Mixed_3b/Branch_2/Conv2d_0c_3x3/BatchNorm/batchnorm/mul | Conv2D | [[1 96 28 28]] | 204 | 301056 | 1554944 | 45826560 | GPU_0_bfc | 1253888 | 0 | 0 | 0 | void cudnn::winograd::generateWinogradTilesKernel<0, float, float>(cudnn::winograd::GenerateWinogradTilesParams<float, float>) | 4.33 | 534528 | 0.00 | 0.00 | 0.00 | 0.00 | 123.36 | true | 534528;534528;534528;534528;534528 |
| 31 | InceptionV2/InceptionV2/Mixed_3b/Branch_2/Conv2d_0c_3x3/BatchNorm/batchnorm/add_1 | Add | [[1 96 28 28]] | 35.333 | 301056 | 0 | 45525504 | GPU_0_bfc | 0 | 0 | 0 | 0 | void Eigen::internal::EigenMetaKernel<Eigen::TensorEvaluator<Eigen::TensorAssignOp<Eigen::TensorMap<Eigen::Tensor<float, 2, 1, int>, 16, Eigen::MakePointer>, Eigen::TensorCwiseBinaryOp<Eigen::internal::scalar_sum_op<float, float>, Eigen::TensorMap<Eigen::Tensor<float const, 2, 1, int>, 16, Eigen::MakePointer> const, Eigen::TensorBroadcastingOp<Eigen::array<long, 2ul> const, Eigen::TensorMap<Eigen::Tensor<float const, 2, 1, int>, 16, Eigen::MakePointer> const> const> const> const, Eigen::GpuDevice>, int>(Eigen::TensorEvaluator<Eigen::TensorAssignOp<Eigen::TensorMap<Eigen::Tensor<float, 2, 1, int>, 16, Eigen::MakePointer>, Eigen::TensorCwiseBinaryOp<Eigen::internal::scalar_sum_op<float, float>, Eigen::TensorMap<Eigen::Tensor<float const, 2, 1, int>, 16, Eigen::MakePointer> const, Eigen::TensorBroadcastingOp<Eigen::array<long, 2ul> const, Eigen::TensorMap<Eigen::Tensor<float const, 2, 1, int>, 16, Eigen::MakePointer> const> const> const> const, Eigen::GpuDevice>, int) | 4.00 | 75264 | 0.00 | 0.00 | 0.00 | 0.00 | 18.82 | true | 75264;75264;75264;75264;75264 |
| 33 | InceptionV2/InceptionV2/Mixed_3b/Branch_0/Conv2d_0a_1x1/Relu | Relu | [[1 256 28 28]] | 32.333 | 802816 | 0 | 45525504 | GPU_0_bfc | 0 | 0 | 0 | 0 | void Eigen::internal::EigenMetaKernel<Eigen::TensorEvaluator<Eigen::TensorAssignOp<Eigen::TensorMap<Eigen::Tensor<float, 1, 1, long>, 16, Eigen::MakePointer>, Eigen::TensorCwiseBinaryOp<Eigen::internal::scalar_max_op<float const, float const>, Eigen::TensorMap<Eigen::Tensor<float const, 1, 1, long>, 16, Eigen::MakePointer> const, Eigen::TensorCwiseNullaryOp<Eigen::internal::scalar_constant_op<float const>, Eigen::TensorMap<Eigen::Tensor<float const, 1, 1, long>, 16, Eigen::MakePointer> const> const> const> const, Eigen::GpuDevice>, long>(Eigen::TensorEvaluator<Eigen::TensorAssignOp<Eigen::TensorMap<Eigen::Tensor<float, 1, 1, long>, 16, Eigen::MakePointer>, Eigen::TensorCwiseBinaryOp<Eigen::internal::scalar_max_op<float const, float const>, Eigen::TensorMap<Eigen::Tensor<float const, 1, 1, long>, 16, Eigen::MakePointer> const, Eigen::TensorCwiseNullaryOp<Eigen::internal::scalar_constant_op<float const>, Eigen::TensorMap<Eigen::Tensor<float const, 1, 1, long>, 16, Eigen::MakePointer> const> const> const> const, Eigen::GpuDevice>, long) | 4.00 | 0 | 0.00 | 0.00 | 0.00 | 0.00 | 0.00 | true | 0;0;0;0;0 |
| 34 | InceptionV2/InceptionV2/Mixed_3c/Branch_3/AvgPool_0a_3x3/AvgPool | AvgPool | [[1 256 28 28]] | 74 | 802816 | 802816 | 46328320 | GPU_0_bfc | 0 | 0 | 0 | 0 | void cudnn::detail::pooling_fw_4d_kernel<float, float, cudnn::detail::averpooling_func<float>, 2, false>(cudnnTensorStruct, float const*, cudnnTensorStruct, float*, cudnnPoolingStruct, float, float, int, cudnn::reduced_divisor, cudnn::reduced_divisor) | 26.17 | 4827031 | 0.00 | 0.00 | 0.00 | 0.00 | 184.47 | true | 4827031;4827031;4827031;4827031;4827031 |
| 35 | InceptionV2/InceptionV2/Mixed_3c/Branch_2/Conv2d_0a_1x1/BatchNorm/batchnorm/mul | Conv2D | [[1 64 28 28]] | 182.333 | 200704 | 271104 | 46529024 | GPU_0_bfc | 70400 | 0 | 0 | 0 | maxwell_scudnn_128x64_relu_interior_nn | 64.00 | 29474816 | 0.00 | 0.00 | 0.00 | 0.00 | 460.54 | true | 29474816;29474816;29474816;29474816;29474816 |
| 35 | InceptionV2/InceptionV2/Mixed_3c/Branch_2/Conv2d_0a_1x1/BatchNorm/batchnorm/mul | Conv2D | [[1 64 28 28]] | 182.333 | 200704 | 271104 | 46529024 | GPU_0_bfc | 70400 | 0 | 0 | 0 | void tensorflow::functor::ShuffleInTensor3Simple<float, 2, 1, 0, false>(int, float const*, tensorflow::functor::Dimension<3>, float*) | 5.00 | 0 | 0.00 | 0.00 | 0.00 | 0.00 | 0.00 | true | 0;0;0;0;0 |
| 35 | InceptionV2/InceptionV2/Mixed_3c/Branch_2/Conv2d_0a_1x1/BatchNorm/batchnorm/mul | Conv2D | [[1 64 28 28]] | 182.333 | 200704 | 271104 | 46529024 | GPU_0_bfc | 70400 | 0 | 0 | 0 | cudnn::maxwell::gemm::computeOffsetsKernel(cudnn::maxwell::gemm::ComputeOffsetsParams) | 2.17 | 0 | 0.00 | 0.00 | 0.00 | 0.00 | 0.00 | true | 0;0;0;0;0 |
| 36 | InceptionV2/InceptionV2/Mixed_3c/Branch_1/Conv2d_0a_1x1/BatchNorm/batchnorm/mul | Conv2D | [[1 64 28 28]] | 178 | 200704 | 271104 | 46729728 | GPU_0_bfc | 70400 | 0 | 0 | 0 | maxwell_scudnn_128x64_relu_interior_nn | 64.00 | 29474816 | 0.00 | 0.00 | 0.00 | 0.00 | 460.54 | true | 29474816;29474816;29474816;29474816;29474816 |
| 36 | InceptionV2/InceptionV2/Mixed_3c/Branch_1/Conv2d_0a_1x1/BatchNorm/batchnorm/mul | Conv2D | [[1 64 28 28]] | 178 | 200704 | 271104 | 46729728 | GPU_0_bfc | 70400 | 0 | 0 | 0 | void tensorflow::functor::ShuffleInTensor3Simple<float, 2, 1, 0, false>(int, float const*, tensorflow::functor::Dimension<3>, float*) | 5.00 | 0 | 0.00 | 0.00 | 0.00 | 0.00 | 0.00 | true | 0;0;0;0;0 |
| 36 | InceptionV2/InceptionV2/Mixed_3c/Branch_1/Conv2d_0a_1x1/BatchNorm/batchnorm/mul | Conv2D | [[1 64 28 28]] | 178 | 200704 | 271104 | 46729728 | GPU_0_bfc | 70400 | 0 | 0 | 0 | cudnn::maxwell::gemm::computeOffsetsKernel(cudnn::maxwell::gemm::ComputeOffsetsParams) | 2.67 | 0 | 0.00 | 0.00 | 0.00 | 0.00 | 0.00 | true | 0;0;0;0;0 |
| 37 | InceptionV2/InceptionV2/Mixed_3c/Branch_0/Conv2d_0a_1x1/BatchNorm/batchnorm/mul | Conv2D | [[1 64 28 28]] | 177.667 | 200704 | 271104 | 46930432 | GPU_0_bfc | 70400 | 0 | 0 | 0 | maxwell_scudnn_128x64_relu_interior_nn | 64.00 | 29474816 | 0.00 | 0.00 | 0.00 | 0.00 | 460.54 | true | 29474816;29474816;29474816;29474816;29474816 |
| 37 | InceptionV2/InceptionV2/Mixed_3c/Branch_0/Conv2d_0a_1x1/BatchNorm/batchnorm/mul | Conv2D | [[1 64 28 28]] | 177.667 | 200704 | 271104 | 46930432 | GPU_0_bfc | 70400 | 0 | 0 | 0 | void tensorflow::functor::ShuffleInTensor3Simple<float, 2, 1, 0, false>(int, float const*, tensorflow::functor::Dimension<3>, float*) | 5.00 | 0 | 0.00 | 0.00 | 0.00 | 0.00 | 0.00 | true | 0;0;0;0;0 |
| 37 | InceptionV2/InceptionV2/Mixed_3c/Branch_0/Conv2d_0a_1x1/BatchNorm/batchnorm/mul | Conv2D | [[1 64 28 28]] | 177.667 | 200704 | 271104 | 46930432 | GPU_0_bfc | 70400 | 0 | 0 | 0 | cudnn::maxwell::gemm::computeOffsetsKernel(cudnn::maxwell::gemm::ComputeOffsetsParams) | 2.50 | 0 | 0.00 | 0.00 | 0.00 | 0.00 | 0.00 | true | 0;0;0;0;0 |
| 38 | InceptionV2/InceptionV2/Mixed_3c/Branch_3/Conv2d_0b_1x1/BatchNorm/batchnorm/mul | Conv2D | [[1 64 28 28]] | 178 | 200704 | 271104 | 46328320 | GPU_0_bfc | 70400 | 0 | 0 | 0 | maxwell_scudnn_128x64_relu_interior_nn | 64.00 | 29474816 | 0.00 | 0.00 | 0.00 | 0.00 | 460.54 | true | 29474816;29474816;29474816;29474816;29474816 |
| 38 | InceptionV2/InceptionV2/Mixed_3c/Branch_3/Conv2d_0b_1x1/BatchNorm/batchnorm/mul | Conv2D | [[1 64 28 28]] | 178 | 200704 | 271104 | 46328320 | GPU_0_bfc | 70400 | 0 | 0 | 0 | void tensorflow::functor::ShuffleInTensor3Simple<float, 2, 1, 0, false>(int, float const*, tensorflow::functor::Dimension<3>, float*) | 5.00 | 0 | 0.00 | 0.00 | 0.00 | 0.00 | 0.00 | true | 0;0;0;0;0 |
| 38 | InceptionV2/InceptionV2/Mixed_3c/Branch_3/Conv2d_0b_1x1/BatchNorm/batchnorm/mul | Conv2D | [[1 64 28 28]] | 178 | 200704 | 271104 | 46328320 | GPU_0_bfc | 70400 | 0 | 0 | 0 | cudnn::maxwell::gemm::computeOffsetsKernel(cudnn::maxwell::gemm::ComputeOffsetsParams) | 2.33 | 0 | 0.00 | 0.00 | 0.00 | 0.00 | 0.00 | true | 0;0;0;0;0 |
| 39 | InceptionV2/InceptionV2/Mixed_3c/Branch_2/Conv2d_0a_1x1/BatchNorm/batchnorm/add_1 | Add | [[1 64 28 28]] | 39.667 | 200704 | 0 | 45525504 | GPU_0_bfc | 0 | 0 | 0 | 0 | void Eigen::internal::EigenMetaKernel<Eigen::TensorEvaluator<Eigen::TensorAssignOp<Eigen::TensorMap<Eigen::Tensor<float, 2, 1, int>, 16, Eigen::MakePointer>, Eigen::TensorCwiseBinaryOp<Eigen::internal::scalar_sum_op<float, float>, Eigen::TensorMap<Eigen::Tensor<float const, 2, 1, int>, 16, Eigen::MakePointer> const, Eigen::TensorBroadcastingOp<Eigen::array<long, 2ul> const, Eigen::TensorMap<Eigen::Tensor<float const, 2, 1, int>, 16, Eigen::MakePointer> const> const> const> const, Eigen::GpuDevice>, int>(Eigen::TensorEvaluator<Eigen::TensorAssignOp<Eigen::TensorMap<Eigen::Tensor<float, 2, 1, int>, 16, Eigen::MakePointer>, Eigen::TensorCwiseBinaryOp<Eigen::internal::scalar_sum_op<float, float>, Eigen::TensorMap<Eigen::Tensor<float const, 2, 1, int>, 16, Eigen::MakePointer> const, Eigen::TensorBroadcastingOp<Eigen::array<long, 2ul> const, Eigen::TensorMap<Eigen::Tensor<float const, 2, 1, int>, 16, Eigen::MakePointer> const> const> const> const, Eigen::GpuDevice>, int) | 4.00 | 50176 | 0.00 | 0.00 | 0.00 | 0.00 | 12.54 | true | 50176;50176;50176;50176;50176 |
| 40 | InceptionV2/InceptionV2/Mixed_3c/Branch_1/Conv2d_0a_1x1/BatchNorm/batchnorm/add_1 | Add | [[1 64 28 28]] | 30 | 200704 | 0 | 45525504 | GPU_0_bfc | 0 | 0 | 0 | 0 | void Eigen::internal::EigenMetaKernel<Eigen::TensorEvaluator<Eigen::TensorAssignOp<Eigen::TensorMap<Eigen::Tensor<float, 2, 1, int>, 16, Eigen::MakePointer>, Eigen::TensorCwiseBinaryOp<Eigen::internal::scalar_sum_op<float, float>, Eigen::TensorMap<Eigen::Tensor<float const, 2, 1, int>, 16, Eigen::MakePointer> const, Eigen::TensorBroadcastingOp<Eigen::array<long, 2ul> const, Eigen::TensorMap<Eigen::Tensor<float const, 2, 1, int>, 16, Eigen::MakePointer> const> const> const> const, Eigen::GpuDevice>, int>(Eigen::TensorEvaluator<Eigen::TensorAssignOp<Eigen::TensorMap<Eigen::Tensor<float, 2, 1, int>, 16, Eigen::MakePointer>, Eigen::TensorCwiseBinaryOp<Eigen::internal::scalar_sum_op<float, float>, Eigen::TensorMap<Eigen::Tensor<float const, 2, 1, int>, 16, Eigen::MakePointer> const, Eigen::TensorBroadcastingOp<Eigen::array<long, 2ul> const, Eigen::TensorMap<Eigen::Tensor<float const, 2, 1, int>, 16, Eigen::MakePointer> const> const> const> const, Eigen::GpuDevice>, int) | 4.00 | 50176 | 0.00 | 0.00 | 0.00 | 0.00 | 12.54 | true | 50176;50176;50176;50176;50176 |
| 41 | InceptionV2/InceptionV2/Mixed_3c/Branch_0/Conv2d_0a_1x1/BatchNorm/batchnorm/add_1 | Add | [[1 64 28 28]] | 28 | 200704 | 0 | 45525504 | GPU_0_bfc | 0 | 0 | 0 | 0 | void Eigen::internal::EigenMetaKernel<Eigen::TensorEvaluator<Eigen::TensorAssignOp<Eigen::TensorMap<Eigen::Tensor<float, 2, 1, int>, 16, Eigen::MakePointer>, Eigen::TensorCwiseBinaryOp<Eigen::internal::scalar_sum_op<float, float>, Eigen::TensorMap<Eigen::Tensor<float const, 2, 1, int>, 16, Eigen::MakePointer> const, Eigen::TensorBroadcastingOp<Eigen::array<long, 2ul> const, Eigen::TensorMap<Eigen::Tensor<float const, 2, 1, int>, 16, Eigen::MakePointer> const> const> const> const, Eigen::GpuDevice>, int>(Eigen::TensorEvaluator<Eigen::TensorAssignOp<Eigen::TensorMap<Eigen::Tensor<float, 2, 1, int>, 16, Eigen::MakePointer>, Eigen::TensorCwiseBinaryOp<Eigen::internal::scalar_sum_op<float, float>, Eigen::TensorMap<Eigen::Tensor<float const, 2, 1, int>, 16, Eigen::MakePointer> const, Eigen::TensorBroadcastingOp<Eigen::array<long, 2ul> const, Eigen::TensorMap<Eigen::Tensor<float const, 2, 1, int>, 16, Eigen::MakePointer> const> const> const> const, Eigen::GpuDevice>, int) | 4.00 | 50176 | 0.00 | 0.00 | 0.00 | 0.00 | 12.54 | true | 50176;50176;50176;50176;50176 |
| 42 | InceptionV2/InceptionV2/Mixed_3c/Branch_3/Conv2d_0b_1x1/BatchNorm/batchnorm/add_1 | Add | [[1 64 28 28]] | 27.667 | 200704 | 0 | 45525504 | GPU_0_bfc | 0 | 0 | 0 | 0 | void Eigen::internal::EigenMetaKernel<Eigen::TensorEvaluator<Eigen::TensorAssignOp<Eigen::TensorMap<Eigen::Tensor<float, 2, 1, int>, 16, Eigen::MakePointer>, Eigen::TensorCwiseBinaryOp<Eigen::internal::scalar_sum_op<float, float>, Eigen::TensorMap<Eigen::Tensor<float const, 2, 1, int>, 16, Eigen::MakePointer> const, Eigen::TensorBroadcastingOp<Eigen::array<long, 2ul> const, Eigen::TensorMap<Eigen::Tensor<float const, 2, 1, int>, 16, Eigen::MakePointer> const> const> const> const, Eigen::GpuDevice>, int>(Eigen::TensorEvaluator<Eigen::TensorAssignOp<Eigen::TensorMap<Eigen::Tensor<float, 2, 1, int>, 16, Eigen::MakePointer>, Eigen::TensorCwiseBinaryOp<Eigen::internal::scalar_sum_op<float, float>, Eigen::TensorMap<Eigen::Tensor<float const, 2, 1, int>, 16, Eigen::MakePointer> const, Eigen::TensorBroadcastingOp<Eigen::array<long, 2ul> const, Eigen::TensorMap<Eigen::Tensor<float const, 2, 1, int>, 16, Eigen::MakePointer> const> const> const> const, Eigen::GpuDevice>, int) | 4.00 | 50176 | 0.00 | 0.00 | 0.00 | 0.00 | 12.54 | true | 50176;50176;50176;50176;50176 |
| 43 | InceptionV2/InceptionV2/Mixed_3c/Branch_2/Conv2d_0a_1x1/Relu | Relu | [[1 64 28 28]] | 26.667 | 200704 | 0 | 45525504 | GPU_0_bfc | 0 | 0 | 0 | 0 | void Eigen::internal::EigenMetaKernel<Eigen::TensorEvaluator<Eigen::TensorAssignOp<Eigen::TensorMap<Eigen::Tensor<float, 1, 1, long>, 16, Eigen::MakePointer>, Eigen::TensorCwiseBinaryOp<Eigen::internal::scalar_max_op<float const, float const>, Eigen::TensorMap<Eigen::Tensor<float const, 1, 1, long>, 16, Eigen::MakePointer> const, Eigen::TensorCwiseNullaryOp<Eigen::internal::scalar_constant_op<float const>, Eigen::TensorMap<Eigen::Tensor<float const, 1, 1, long>, 16, Eigen::MakePointer> const> const> const> const, Eigen::GpuDevice>, long>(Eigen::TensorEvaluator<Eigen::TensorAssignOp<Eigen::TensorMap<Eigen::Tensor<float, 1, 1, long>, 16, Eigen::MakePointer>, Eigen::TensorCwiseBinaryOp<Eigen::internal::scalar_max_op<float const, float const>, Eigen::TensorMap<Eigen::Tensor<float const, 1, 1, long>, 16, Eigen::MakePointer> const, Eigen::TensorCwiseNullaryOp<Eigen::internal::scalar_constant_op<float const>, Eigen::TensorMap<Eigen::Tensor<float const, 1, 1, long>, 16, Eigen::MakePointer> const> const> const> const, Eigen::GpuDevice>, long) | 3.00 | 0 | 0.00 | 0.00 | 0.00 | 0.00 | 0.00 | true | 0;0;0;0;0 |
| 44 | InceptionV2/InceptionV2/Mixed_3c/Branch_1/Conv2d_0a_1x1/Relu | Relu | [[1 64 28 28]] | 24.333 | 200704 | 0 | 45525504 | GPU_0_bfc | 0 | 0 | 0 | 0 | void Eigen::internal::EigenMetaKernel<Eigen::TensorEvaluator<Eigen::TensorAssignOp<Eigen::TensorMap<Eigen::Tensor<float, 1, 1, long>, 16, Eigen::MakePointer>, Eigen::TensorCwiseBinaryOp<Eigen::internal::scalar_max_op<float const, float const>, Eigen::TensorMap<Eigen::Tensor<float const, 1, 1, long>, 16, Eigen::MakePointer> const, Eigen::TensorCwiseNullaryOp<Eigen::internal::scalar_constant_op<float const>, Eigen::TensorMap<Eigen::Tensor<float const, 1, 1, long>, 16, Eigen::MakePointer> const> const> const> const, Eigen::GpuDevice>, long>(Eigen::TensorEvaluator<Eigen::TensorAssignOp<Eigen::TensorMap<Eigen::Tensor<float, 1, 1, long>, 16, Eigen::MakePointer>, Eigen::TensorCwiseBinaryOp<Eigen::internal::scalar_max_op<float const, float const>, Eigen::TensorMap<Eigen::Tensor<float const, 1, 1, long>, 16, Eigen::MakePointer> const, Eigen::TensorCwiseNullaryOp<Eigen::internal::scalar_constant_op<float const>, Eigen::TensorMap<Eigen::Tensor<float const, 1, 1, long>, 16, Eigen::MakePointer> const> const> const> const, Eigen::GpuDevice>, long) | 3.00 | 0 | 0.00 | 0.00 | 0.00 | 0.00 | 0.00 | true | 0;0;0;0;0 |
| 45 | InceptionV2/InceptionV2/Mixed_3c/Branch_2/Conv2d_0b_3x3/BatchNorm/batchnorm/mul | Conv2D | [[1 96 28 28]] | 197.333 | 301056 | 1137152 | 45826560 | GPU_0_bfc | 836096 | 0 | 0 | 0 | maxwell_scudnn_winograd_128x128_ldg1_ldg4_tile148n_nt | 26.00 | 53231616 | 0.00 | 0.00 | 0.00 | 0.00 | 2047.37 | true | 53231616;53231616;53231616;53231616;53231616 |
| 45 | InceptionV2/InceptionV2/Mixed_3c/Branch_2/Conv2d_0b_3x3/BatchNorm/batchnorm/mul | Conv2D | [[1 96 28 28]] | 197.333 | 301056 | 1137152 | 45826560 | GPU_0_bfc | 836096 | 0 | 0 | 0 | void tensorflow::functor::ShuffleInTensor3Simple<float, 2, 1, 0, false>(int, float const*, tensorflow::functor::Dimension<3>, float*) | 5.33 | 0 | 0.00 | 0.00 | 0.00 | 0.00 | 0.00 | true | 0;0;0;0;0 |
| 45 | InceptionV2/InceptionV2/Mixed_3c/Branch_2/Conv2d_0b_3x3/BatchNorm/batchnorm/mul | Conv2D | [[1 96 28 28]] | 197.333 | 301056 | 1137152 | 45826560 | GPU_0_bfc | 836096 | 0 | 0 | 0 | void cudnn::winograd::generateWinogradTilesKernel<0, float, float>(cudnn::winograd::GenerateWinogradTilesParams<float, float>) | 4.00 | 356352 | 0.00 | 0.00 | 0.00 | 0.00 | 89.09 | true | 356352;356352;356352;356352;356352 |
| 46 | InceptionV2/InceptionV2/Mixed_3c/Branch_1/Conv2d_0b_3x3/BatchNorm/batchnorm/mul | Conv2D | [[1 96 28 28]] | 196 | 301056 | 1317376 | 45926912 | GPU_0_bfc | 1016320 | 0 | 0 | 0 | maxwell_scudnn_winograd_128x128_ldg1_ldg4_tile148n_nt | 26.00 | 53231616 | 0.00 | 0.00 | 0.00 | 0.00 | 2047.37 | true | 53231616;53231616;53231616;53231616;53231616 |
| 46 | InceptionV2/InceptionV2/Mixed_3c/Branch_1/Conv2d_0b_3x3/BatchNorm/batchnorm/mul | Conv2D | [[1 96 28 28]] | 196 | 301056 | 1317376 | 45926912 | GPU_0_bfc | 1016320 | 0 | 0 | 0 | void tensorflow::functor::ShuffleInTensor3Simple<float, 2, 1, 0, false>(int, float const*, tensorflow::functor::Dimension<3>, float*) | 5.17 | 0 | 0.00 | 0.00 | 0.00 | 0.00 | 0.00 | true | 0;0;0;0;0 |
| 46 | InceptionV2/InceptionV2/Mixed_3c/Branch_1/Conv2d_0b_3x3/BatchNorm/batchnorm/mul | Conv2D | [[1 96 28 28]] | 196 | 301056 | 1317376 | 45926912 | GPU_0_bfc | 1016320 | 0 | 0 | 0 | void cudnn::winograd::generateWinogradTilesKernel<0, float, float>(cudnn::winograd::GenerateWinogradTilesParams<float, float>) | 4.00 | 356352 | 0.00 | 0.00 | 0.00 | 0.00 | 89.09 | true | 356352;356352;356352;356352;356352 |
| 47 | InceptionV2/InceptionV2/Mixed_3c/Branch_2/Conv2d_0b_3x3/BatchNorm/batchnorm/add_1 | Add | [[1 96 28 28]] | 35 | 301056 | 0 | 45726208 | GPU_0_bfc | 0 | 0 | 0 | 0 | void Eigen::internal::EigenMetaKernel<Eigen::TensorEvaluator<Eigen::TensorAssignOp<Eigen::TensorMap<Eigen::Tensor<float, 2, 1, int>, 16, Eigen::MakePointer>, Eigen::TensorCwiseBinaryOp<Eigen::internal::scalar_sum_op<float, float>, Eigen::TensorMap<Eigen::Tensor<float const, 2, 1, int>, 16, Eigen::MakePointer> const, Eigen::TensorBroadcastingOp<Eigen::array<long, 2ul> const, Eigen::TensorMap<Eigen::Tensor<float const, 2, 1, int>, 16, Eigen::MakePointer> const> const> const> const, Eigen::GpuDevice>, int>(Eigen::TensorEvaluator<Eigen::TensorAssignOp<Eigen::TensorMap<Eigen::Tensor<float, 2, 1, int>, 16, Eigen::MakePointer>, Eigen::TensorCwiseBinaryOp<Eigen::internal::scalar_sum_op<float, float>, Eigen::TensorMap<Eigen::Tensor<float const, 2, 1, int>, 16, Eigen::MakePointer> const, Eigen::TensorBroadcastingOp<Eigen::array<long, 2ul> const, Eigen::TensorMap<Eigen::Tensor<float const, 2, 1, int>, 16, Eigen::MakePointer> const> const> const> const, Eigen::GpuDevice>, int) | 4.00 | 75264 | 0.00 | 0.00 | 0.00 | 0.00 | 18.82 | true | 75264;75264;75264;75264;75264 |
| 48 | InceptionV2/InceptionV2/Mixed_3c/Branch_1/Conv2d_0b_3x3/BatchNorm/batchnorm/add_1 | Add | [[1 96 28 28]] | 29 | 301056 | 0 | 45726208 | GPU_0_bfc | 0 | 0 | 0 | 0 | void Eigen::internal::EigenMetaKernel<Eigen::TensorEvaluator<Eigen::TensorAssignOp<Eigen::TensorMap<Eigen::Tensor<float, 2, 1, int>, 16, Eigen::MakePointer>, Eigen::TensorCwiseBinaryOp<Eigen::internal::scalar_sum_op<float, float>, Eigen::TensorMap<Eigen::Tensor<float const, 2, 1, int>, 16, Eigen::MakePointer> const, Eigen::TensorBroadcastingOp<Eigen::array<long, 2ul> const, Eigen::TensorMap<Eigen::Tensor<float const, 2, 1, int>, 16, Eigen::MakePointer> const> const> const> const, Eigen::GpuDevice>, int>(Eigen::TensorEvaluator<Eigen::TensorAssignOp<Eigen::TensorMap<Eigen::Tensor<float, 2, 1, int>, 16, Eigen::MakePointer>, Eigen::TensorCwiseBinaryOp<Eigen::internal::scalar_sum_op<float, float>, Eigen::TensorMap<Eigen::Tensor<float const, 2, 1, int>, 16, Eigen::MakePointer> const, Eigen::TensorBroadcastingOp<Eigen::array<long, 2ul> const, Eigen::TensorMap<Eigen::Tensor<float const, 2, 1, int>, 16, Eigen::MakePointer> const> const> const> const, Eigen::GpuDevice>, int) | 4.00 | 75264 | 0.00 | 0.00 | 0.00 | 0.00 | 18.82 | true | 75264;75264;75264;75264;75264 |
| 49 | InceptionV2/InceptionV2/Mixed_3c/Branch_2/Conv2d_0b_3x3/Relu | Relu | [[1 96 28 28]] | 26.667 | 301056 | 0 | 45726208 | GPU_0_bfc | 0 | 0 | 0 | 0 | void Eigen::internal::EigenMetaKernel<Eigen::TensorEvaluator<Eigen::TensorAssignOp<Eigen::TensorMap<Eigen::Tensor<float, 1, 1, long>, 16, Eigen::MakePointer>, Eigen::TensorCwiseBinaryOp<Eigen::internal::scalar_max_op<float const, float const>, Eigen::TensorMap<Eigen::Tensor<float const, 1, 1, long>, 16, Eigen::MakePointer> const, Eigen::TensorCwiseNullaryOp<Eigen::internal::scalar_constant_op<float const>, Eigen::TensorMap<Eigen::Tensor<float const, 1, 1, long>, 16, Eigen::MakePointer> const> const> const> const, Eigen::GpuDevice>, long>(Eigen::TensorEvaluator<Eigen::TensorAssignOp<Eigen::TensorMap<Eigen::Tensor<float, 1, 1, long>, 16, Eigen::MakePointer>, Eigen::TensorCwiseBinaryOp<Eigen::internal::scalar_max_op<float const, float const>, Eigen::TensorMap<Eigen::Tensor<float const, 1, 1, long>, 16, Eigen::MakePointer> const, Eigen::TensorCwiseNullaryOp<Eigen::internal::scalar_constant_op<float const>, Eigen::TensorMap<Eigen::Tensor<float const, 1, 1, long>, 16, Eigen::MakePointer> const> const> const> const, Eigen::GpuDevice>, long) | 3.00 | 0 | 0.00 | 0.00 | 0.00 | 0.00 | 0.00 | true | 0;0;0;0;0 |
| 50 | InceptionV2/InceptionV2/Mixed_3c/Branch_2/Conv2d_0c_3x3/BatchNorm/batchnorm/mul | Conv2D | [[1 96 28 28]] | 203 | 301056 | 1554944 | 46027264 | GPU_0_bfc | 1253888 | 0 | 0 | 0 | maxwell_scudnn_winograd_128x128_ldg1_ldg4_tile148n_nt | 34.00 | 79380480 | 0.00 | 0.00 | 0.00 | 0.00 | 2334.72 | true | 79380480;79380480;79380480;79380480;79380480 |
| 50 | InceptionV2/InceptionV2/Mixed_3c/Branch_2/Conv2d_0c_3x3/BatchNorm/batchnorm/mul | Conv2D | [[1 96 28 28]] | 203 | 301056 | 1554944 | 46027264 | GPU_0_bfc | 1253888 | 0 | 0 | 0 | void tensorflow::functor::ShuffleInTensor3Simple<float, 2, 1, 0, false>(int, float const*, tensorflow::functor::Dimension<3>, float*) | 7.00 | 0 | 0.00 | 0.00 | 0.00 | 0.00 | 0.00 | true | 0;0;0;0;0 |
| 50 | InceptionV2/InceptionV2/Mixed_3c/Branch_2/Conv2d_0c_3x3/BatchNorm/batchnorm/mul | Conv2D | [[1 96 28 28]] | 203 | 301056 | 1554944 | 46027264 | GPU_0_bfc | 1253888 | 0 | 0 | 0 | void cudnn::winograd::generateWinogradTilesKernel<0, float, float>(cudnn::winograd::GenerateWinogradTilesParams<float, float>) | 4.33 | 534528 | 0.00 | 0.00 | 0.00 | 0.00 | 123.36 | true | 534528;534528;534528;534528;534528 |
| 51 | InceptionV2/InceptionV2/Mixed_3c/Branch_2/Conv2d_0c_3x3/BatchNorm/batchnorm/add_1 | Add | [[1 96 28 28]] | 35.333 | 301056 | 0 | 45726208 | GPU_0_bfc | 0 | 0 | 0 | 0 | void Eigen::internal::EigenMetaKernel<Eigen::TensorEvaluator<Eigen::TensorAssignOp<Eigen::TensorMap<Eigen::Tensor<float, 2, 1, int>, 16, Eigen::MakePointer>, Eigen::TensorCwiseBinaryOp<Eigen::internal::scalar_sum_op<float, float>, Eigen::TensorMap<Eigen::Tensor<float const, 2, 1, int>, 16, Eigen::MakePointer> const, Eigen::TensorBroadcastingOp<Eigen::array<long, 2ul> const, Eigen::TensorMap<Eigen::Tensor<float const, 2, 1, int>, 16, Eigen::MakePointer> const> const> const> const, Eigen::GpuDevice>, int>(Eigen::TensorEvaluator<Eigen::TensorAssignOp<Eigen::TensorMap<Eigen::Tensor<float, 2, 1, int>, 16, Eigen::MakePointer>, Eigen::TensorCwiseBinaryOp<Eigen::internal::scalar_sum_op<float, float>, Eigen::TensorMap<Eigen::Tensor<float const, 2, 1, int>, 16, Eigen::MakePointer> const, Eigen::TensorBroadcastingOp<Eigen::array<long, 2ul> const, Eigen::TensorMap<Eigen::Tensor<float const, 2, 1, int>, 16, Eigen::MakePointer> const> const> const> const, Eigen::GpuDevice>, int) | 4.00 | 75264 | 0.00 | 0.00 | 0.00 | 0.00 | 18.82 | true | 75264;75264;75264;75264;75264 |
| 53 | InceptionV2/InceptionV2/Mixed_3c/Branch_0/Conv2d_0a_1x1/Relu | Relu | [[1 320 28 28]] | 30.333 | 1003520 | 0 | 45726208 | GPU_0_bfc | 0 | 0 | 0 | 0 | void Eigen::internal::EigenMetaKernel<Eigen::TensorEvaluator<Eigen::TensorAssignOp<Eigen::TensorMap<Eigen::Tensor<float, 1, 1, long>, 16, Eigen::MakePointer>, Eigen::TensorCwiseBinaryOp<Eigen::internal::scalar_max_op<float const, float const>, Eigen::TensorMap<Eigen::Tensor<float const, 1, 1, long>, 16, Eigen::MakePointer> const, Eigen::TensorCwiseNullaryOp<Eigen::internal::scalar_constant_op<float const>, Eigen::TensorMap<Eigen::Tensor<float const, 1, 1, long>, 16, Eigen::MakePointer> const> const> const> const, Eigen::GpuDevice>, long>(Eigen::TensorEvaluator<Eigen::TensorAssignOp<Eigen::TensorMap<Eigen::Tensor<float, 1, 1, long>, 16, Eigen::MakePointer>, Eigen::TensorCwiseBinaryOp<Eigen::internal::scalar_max_op<float const, float const>, Eigen::TensorMap<Eigen::Tensor<float const, 1, 1, long>, 16, Eigen::MakePointer> const, Eigen::TensorCwiseNullaryOp<Eigen::internal::scalar_constant_op<float const>, Eigen::TensorMap<Eigen::Tensor<float const, 1, 1, long>, 16, Eigen::MakePointer> const> const> const> const, Eigen::GpuDevice>, long) | 4.00 | 0 | 0.00 | 0.00 | 0.00 | 0.00 | 0.00 | true | 0;0;0;0;0 |
| 54 | InceptionV2/InceptionV2/Mixed_4a/Branch_2/MaxPool_1a_3x3/MaxPool | MaxPool | [[1 320 14 14]] | 56.333 | 250880 | 250880 | 45977088 | GPU_0_bfc | 0 | 0 | 0 | 0 | void cudnn::detail::pooling_fw_4d_kernel<float, float, cudnn::detail::maxpooling_func<float, (cudnnNanPropagation_t)0>, 0, false>(cudnnTensorStruct, float const*, cudnnTensorStruct, float*, cudnnPoolingStruct, float, float, int, cudnn::reduced_divisor, cudnn::reduced_divisor) | 9.00 | 62720 | 0.00 | 0.00 | 0.00 | 0.00 | 6.97 | true | 62720;62720;62720;62720;62720 |
| 55 | InceptionV2/InceptionV2/Mixed_4a/Branch_1/Conv2d_0a_1x1/BatchNorm/batchnorm/mul | Conv2D | [[1 64 28 28]] | 198.333 | 200704 | 287488 | 46177792 | GPU_0_bfc | 86784 | 0 | 0 | 0 | maxwell_scudnn_128x64_relu_interior_nn | 77.00 | 36814848 | 0.00 | 0.00 | 0.00 | 0.00 | 478.11 | true | 36814848;36814848;36814848;36814848;36814848 |
| 55 | InceptionV2/InceptionV2/Mixed_4a/Branch_1/Conv2d_0a_1x1/BatchNorm/batchnorm/mul | Conv2D | [[1 64 28 28]] | 198.333 | 200704 | 287488 | 46177792 | GPU_0_bfc | 86784 | 0 | 0 | 0 | void tensorflow::functor::ShuffleInTensor3Simple<float, 2, 1, 0, false>(int, float const*, tensorflow::functor::Dimension<3>, float*) | 5.00 | 0 | 0.00 | 0.00 | 0.00 | 0.00 | 0.00 | true | 0;0;0;0;0 |
| 55 | InceptionV2/InceptionV2/Mixed_4a/Branch_1/Conv2d_0a_1x1/BatchNorm/batchnorm/mul | Conv2D | [[1 64 28 28]] | 198.333 | 200704 | 287488 | 46177792 | GPU_0_bfc | 86784 | 0 | 0 | 0 | cudnn::maxwell::gemm::computeOffsetsKernel(cudnn::maxwell::gemm::ComputeOffsetsParams) | 3.00 | 0 | 0.00 | 0.00 | 0.00 | 0.00 | 0.00 | true | 0;0;0;0;0 |
| 56 | InceptionV2/InceptionV2/Mixed_4a/Branch_0/Conv2d_0a_1x1/BatchNorm/batchnorm/mul | Conv2D | [[1 128 28 28]] | 222.333 | 401408 | 570112 | 46579200 | GPU_0_bfc | 168704 | 0 | 0 | 0 | maxwell_scudnn_128x128_relu_interior_nn | 101.50 | 73629696 | 0.00 | 0.00 | 0.00 | 0.00 | 725.42 | true | 73629696;73629696;73629696;73629696;73629696 |
| 56 | InceptionV2/InceptionV2/Mixed_4a/Branch_0/Conv2d_0a_1x1/BatchNorm/batchnorm/mul | Conv2D | [[1 128 28 28]] | 222.333 | 401408 | 570112 | 46579200 | GPU_0_bfc | 168704 | 0 | 0 | 0 | void tensorflow::functor::ShuffleInTensor3Simple<float, 2, 1, 0, false>(int, float const*, tensorflow::functor::Dimension<3>, float*) | 5.00 | 0 | 0.00 | 0.00 | 0.00 | 0.00 | 0.00 | true | 0;0;0;0;0 |
| 56 | InceptionV2/InceptionV2/Mixed_4a/Branch_0/Conv2d_0a_1x1/BatchNorm/batchnorm/mul | Conv2D | [[1 128 28 28]] | 222.333 | 401408 | 570112 | 46579200 | GPU_0_bfc | 168704 | 0 | 0 | 0 | cudnn::maxwell::gemm::computeOffsetsKernel(cudnn::maxwell::gemm::ComputeOffsetsParams) | 2.67 | 0 | 0.00 | 0.00 | 0.00 | 0.00 | 0.00 | true | 0;0;0;0;0 |
| 57 | InceptionV2/InceptionV2/Mixed_4a/Branch_1/Conv2d_0a_1x1/BatchNorm/batchnorm/add_1 | Add | [[1 64 28 28]] | 37.333 | 200704 | 0 | 45575680 | GPU_0_bfc | 0 | 0 | 0 | 0 | void Eigen::internal::EigenMetaKernel<Eigen::TensorEvaluator<Eigen::TensorAssignOp<Eigen::TensorMap<Eigen::Tensor<float, 2, 1, int>, 16, Eigen::MakePointer>, Eigen::TensorCwiseBinaryOp<Eigen::internal::scalar_sum_op<float, float>, Eigen::TensorMap<Eigen::Tensor<float const, 2, 1, int>, 16, Eigen::MakePointer> const, Eigen::TensorBroadcastingOp<Eigen::array<long, 2ul> const, Eigen::TensorMap<Eigen::Tensor<float const, 2, 1, int>, 16, Eigen::MakePointer> const> const> const> const, Eigen::GpuDevice>, int>(Eigen::TensorEvaluator<Eigen::TensorAssignOp<Eigen::TensorMap<Eigen::Tensor<float, 2, 1, int>, 16, Eigen::MakePointer>, Eigen::TensorCwiseBinaryOp<Eigen::internal::scalar_sum_op<float, float>, Eigen::TensorMap<Eigen::Tensor<float const, 2, 1, int>, 16, Eigen::MakePointer> const, Eigen::TensorBroadcastingOp<Eigen::array<long, 2ul> const, Eigen::TensorMap<Eigen::Tensor<float const, 2, 1, int>, 16, Eigen::MakePointer> const> const> const> const, Eigen::GpuDevice>, int) | 4.00 | 50176 | 0.00 | 0.00 | 0.00 | 0.00 | 12.54 | true | 50176;50176;50176;50176;50176 |
| 58 | InceptionV2/InceptionV2/Mixed_4a/Branch_0/Conv2d_0a_1x1/BatchNorm/batchnorm/add_1 | Add | [[1 128 28 28]] | 28.333 | 401408 | 0 | 45575680 | GPU_0_bfc | 0 | 0 | 0 | 0 | void Eigen::internal::EigenMetaKernel<Eigen::TensorEvaluator<Eigen::TensorAssignOp<Eigen::TensorMap<Eigen::Tensor<float, 2, 1, int>, 16, Eigen::MakePointer>, Eigen::TensorCwiseBinaryOp<Eigen::internal::scalar_sum_op<float, float>, Eigen::TensorMap<Eigen::Tensor<float const, 2, 1, int>, 16, Eigen::MakePointer> const, Eigen::TensorBroadcastingOp<Eigen::array<long, 2ul> const, Eigen::TensorMap<Eigen::Tensor<float const, 2, 1, int>, 16, Eigen::MakePointer> const> const> const> const, Eigen::GpuDevice>, int>(Eigen::TensorEvaluator<Eigen::TensorAssignOp<Eigen::TensorMap<Eigen::Tensor<float, 2, 1, int>, 16, Eigen::MakePointer>, Eigen::TensorCwiseBinaryOp<Eigen::internal::scalar_sum_op<float, float>, Eigen::TensorMap<Eigen::Tensor<float const, 2, 1, int>, 16, Eigen::MakePointer> const, Eigen::TensorBroadcastingOp<Eigen::array<long, 2ul> const, Eigen::TensorMap<Eigen::Tensor<float const, 2, 1, int>, 16, Eigen::MakePointer> const> const> const> const, Eigen::GpuDevice>, int) | 4.00 | 100352 | 0.00 | 0.00 | 0.00 | 0.00 | 25.09 | true | 100352;100352;100352;100352;100352 |
| 59 | InceptionV2/InceptionV2/Mixed_4a/Branch_1/Conv2d_0a_1x1/Relu | Relu | [[1 64 28 28]] | 25.667 | 200704 | 0 | 45575680 | GPU_0_bfc | 0 | 0 | 0 | 0 | void Eigen::internal::EigenMetaKernel<Eigen::TensorEvaluator<Eigen::TensorAssignOp<Eigen::TensorMap<Eigen::Tensor<float, 1, 1, long>, 16, Eigen::MakePointer>, Eigen::TensorCwiseBinaryOp<Eigen::internal::scalar_max_op<float const, float const>, Eigen::TensorMap<Eigen::Tensor<float const, 1, 1, long>, 16, Eigen::MakePointer> const, Eigen::TensorCwiseNullaryOp<Eigen::internal::scalar_constant_op<float const>, Eigen::TensorMap<Eigen::Tensor<float const, 1, 1, long>, 16, Eigen::MakePointer> const> const> const> const, Eigen::GpuDevice>, long>(Eigen::TensorEvaluator<Eigen::TensorAssignOp<Eigen::TensorMap<Eigen::Tensor<float, 1, 1, long>, 16, Eigen::MakePointer>, Eigen::TensorCwiseBinaryOp<Eigen::internal::scalar_max_op<float const, float const>, Eigen::TensorMap<Eigen::Tensor<float const, 1, 1, long>, 16, Eigen::MakePointer> const, Eigen::TensorCwiseNullaryOp<Eigen::internal::scalar_constant_op<float const>, Eigen::TensorMap<Eigen::Tensor<float const, 1, 1, long>, 16, Eigen::MakePointer> const> const> const> const, Eigen::GpuDevice>, long) | 3.00 | 0 | 0.00 | 0.00 | 0.00 | 0.00 | 0.00 | true | 0;0;0;0;0 |
| 60 | InceptionV2/InceptionV2/Mixed_4a/Branch_0/Conv2d_0a_1x1/Relu | Relu | [[1 128 28 28]] | 25.333 | 401408 | 0 | 45575680 | GPU_0_bfc | 0 | 0 | 0 | 0 | void Eigen::internal::EigenMetaKernel<Eigen::TensorEvaluator<Eigen::TensorAssignOp<Eigen::TensorMap<Eigen::Tensor<float, 1, 1, long>, 16, Eigen::MakePointer>, Eigen::TensorCwiseBinaryOp<Eigen::internal::scalar_max_op<float const, float const>, Eigen::TensorMap<Eigen::Tensor<float const, 1, 1, long>, 16, Eigen::MakePointer> const, Eigen::TensorCwiseNullaryOp<Eigen::internal::scalar_constant_op<float const>, Eigen::TensorMap<Eigen::Tensor<float const, 1, 1, long>, 16, Eigen::MakePointer> const> const> const> const, Eigen::GpuDevice>, long>(Eigen::TensorEvaluator<Eigen::TensorAssignOp<Eigen::TensorMap<Eigen::Tensor<float, 1, 1, long>, 16, Eigen::MakePointer>, Eigen::TensorCwiseBinaryOp<Eigen::internal::scalar_max_op<float const, float const>, Eigen::TensorMap<Eigen::Tensor<float const, 1, 1, long>, 16, Eigen::MakePointer> const, Eigen::TensorCwiseNullaryOp<Eigen::internal::scalar_constant_op<float const>, Eigen::TensorMap<Eigen::Tensor<float const, 1, 1, long>, 16, Eigen::MakePointer> const> const> const> const, Eigen::GpuDevice>, long) | 3.00 | 0 | 0.00 | 0.00 | 0.00 | 0.00 | 0.00 | true | 0;0;0;0;0 |
| 61 | InceptionV2/InceptionV2/Mixed_4a/Branch_1/Conv2d_0b_3x3/BatchNorm/batchnorm/mul | Conv2D | [[1 96 28 28]] | 196.667 | 301056 | 1137152 | 45876736 | GPU_0_bfc | 836096 | 0 | 0 | 0 | maxwell_scudnn_winograd_128x128_ldg1_ldg4_tile148n_nt | 26.00 | 53231616 | 0.00 | 0.00 | 0.00 | 0.00 | 2047.37 | true | 53231616;53231616;53231616;53231616;53231616 |
| 61 | InceptionV2/InceptionV2/Mixed_4a/Branch_1/Conv2d_0b_3x3/BatchNorm/batchnorm/mul | Conv2D | [[1 96 28 28]] | 196.667 | 301056 | 1137152 | 45876736 | GPU_0_bfc | 836096 | 0 | 0 | 0 | void tensorflow::functor::ShuffleInTensor3Simple<float, 2, 1, 0, false>(int, float const*, tensorflow::functor::Dimension<3>, float*) | 5.33 | 0 | 0.00 | 0.00 | 0.00 | 0.00 | 0.00 | true | 0;0;0;0;0 |
| 61 | InceptionV2/InceptionV2/Mixed_4a/Branch_1/Conv2d_0b_3x3/BatchNorm/batchnorm/mul | Conv2D | [[1 96 28 28]] | 196.667 | 301056 | 1137152 | 45876736 | GPU_0_bfc | 836096 | 0 | 0 | 0 | void cudnn::winograd::generateWinogradTilesKernel<0, float, float>(cudnn::winograd::GenerateWinogradTilesParams<float, float>) | 4.00 | 356352 | 0.00 | 0.00 | 0.00 | 0.00 | 89.09 | true | 356352;356352;356352;356352;356352 |
| 62 | InceptionV2/InceptionV2/Mixed_4a/Branch_0/Conv2d_1a_3x3/BatchNorm/batchnorm/mul | Conv2D | [[1 160 14 14]] | 313.333 | 200704 | 2271744 | 45876736 | GPU_0_bfc | 2071040 | 0 | 0 | 0 | void cudnn::detail::explicit_convolve_sgemm<float, int, 128, 5, 5, 3, 3, 3, 0, true>(int, int, int, float const*, int, float const*, int, float*, kernel_conv_params, int, int, float, float, int, float*, float*) | 71.00 | 82606720 | 0.00 | 0.00 | 0.00 | 0.00 | 1163.47 | true | 82606720;82606720;82606720;82606720;82606720 |
| 62 | InceptionV2/InceptionV2/Mixed_4a/Branch_0/Conv2d_1a_3x3/BatchNorm/batchnorm/mul | Conv2D | [[1 160 14 14]] | 313.333 | 200704 | 2271744 | 45876736 | GPU_0_bfc | 2071040 | 0 | 0 | 0 | void im2col4d_kernel<float, int>(im2col4d_params, cudnnConvolutionStruct, cudnnTensor4dStruct, float const*, float*, int) | 63.67 | 0 | 0.00 | 0.00 | 0.00 | 0.00 | 0.00 | true | 0;0;0;0;0 |
| 62 | InceptionV2/InceptionV2/Mixed_4a/Branch_0/Conv2d_1a_3x3/BatchNorm/batchnorm/mul | Conv2D | [[1 160 14 14]] | 313.333 | 200704 | 2271744 | 45876736 | GPU_0_bfc | 2071040 | 0 | 0 | 0 | void tensorflow::functor::ShuffleInTensor3Simple<float, 2, 1, 0, false>(int, float const*, tensorflow::functor::Dimension<3>, float*) | 12.00 | 0 | 0.00 | 0.00 | 0.00 | 0.00 | 0.00 | true | 0;0;0;0;0 |
| 62 | InceptionV2/InceptionV2/Mixed_4a/Branch_0/Conv2d_1a_3x3/BatchNorm/batchnorm/mul | Conv2D | [[1 160 14 14]] | 313.333 | 200704 | 2271744 | 45876736 | GPU_0_bfc | 2071040 | 0 | 0 | 0 | void tensorflow::functor::PadInputCustomKernelNCHW<float, 4>(int, float const*, tensorflow::functor::Dimension<4>, float*, tensorflow::functor::Dimension<4>, tensorflow::functor::Dimension<(4)-(2)>) | 8.00 | 0 | 0.00 | 0.00 | 0.00 | 0.00 | 0.00 | true | 0;0;0;0;0 |
| 63 | InceptionV2/InceptionV2/Mixed_4a/Branch_1/Conv2d_0b_3x3/BatchNorm/batchnorm/add_1 | Add | [[1 96 28 28]] | 38 | 301056 | 0 | 45475328 | GPU_0_bfc | 0 | 0 | 0 | 0 | void Eigen::internal::EigenMetaKernel<Eigen::TensorEvaluator<Eigen::TensorAssignOp<Eigen::TensorMap<Eigen::Tensor<float, 2, 1, int>, 16, Eigen::MakePointer>, Eigen::TensorCwiseBinaryOp<Eigen::internal::scalar_sum_op<float, float>, Eigen::TensorMap<Eigen::Tensor<float const, 2, 1, int>, 16, Eigen::MakePointer> const, Eigen::TensorBroadcastingOp<Eigen::array<long, 2ul> const, Eigen::TensorMap<Eigen::Tensor<float const, 2, 1, int>, 16, Eigen::MakePointer> const> const> const> const, Eigen::GpuDevice>, int>(Eigen::TensorEvaluator<Eigen::TensorAssignOp<Eigen::TensorMap<Eigen::Tensor<float, 2, 1, int>, 16, Eigen::MakePointer>, Eigen::TensorCwiseBinaryOp<Eigen::internal::scalar_sum_op<float, float>, Eigen::TensorMap<Eigen::Tensor<float const, 2, 1, int>, 16, Eigen::MakePointer> const, Eigen::TensorBroadcastingOp<Eigen::array<long, 2ul> const, Eigen::TensorMap<Eigen::Tensor<float const, 2, 1, int>, 16, Eigen::MakePointer> const> const> const> const, Eigen::GpuDevice>, int) | 4.00 | 75264 | 0.00 | 0.00 | 0.00 | 0.00 | 18.82 | true | 75264;75264;75264;75264;75264 |
| 64 | InceptionV2/InceptionV2/Mixed_4a/Branch_0/Conv2d_1a_3x3/BatchNorm/batchnorm/add_1 | Add | [[1 160 14 14]] | 29.333 | 200704 | 0 | 45475328 | GPU_0_bfc | 0 | 0 | 0 | 0 | void Eigen::internal::EigenMetaKernel<Eigen::TensorEvaluator<Eigen::TensorAssignOp<Eigen::TensorMap<Eigen::Tensor<float, 2, 1, int>, 16, Eigen::MakePointer>, Eigen::TensorCwiseBinaryOp<Eigen::internal::scalar_sum_op<float, float>, Eigen::TensorMap<Eigen::Tensor<float const, 2, 1, int>, 16, Eigen::MakePointer> const, Eigen::TensorBroadcastingOp<Eigen::array<long, 2ul> const, Eigen::TensorMap<Eigen::Tensor<float const, 2, 1, int>, 16, Eigen::MakePointer> const> const> const> const, Eigen::GpuDevice>, int>(Eigen::TensorEvaluator<Eigen::TensorAssignOp<Eigen::TensorMap<Eigen::Tensor<float, 2, 1, int>, 16, Eigen::MakePointer>, Eigen::TensorCwiseBinaryOp<Eigen::internal::scalar_sum_op<float, float>, Eigen::TensorMap<Eigen::Tensor<float const, 2, 1, int>, 16, Eigen::MakePointer> const, Eigen::TensorBroadcastingOp<Eigen::array<long, 2ul> const, Eigen::TensorMap<Eigen::Tensor<float const, 2, 1, int>, 16, Eigen::MakePointer> const> const> const> const, Eigen::GpuDevice>, int) | 4.00 | 31360 | 0.00 | 0.00 | 0.00 | 0.00 | 7.84 | true | 31360;31360;31360;31360;31360 |
| 65 | InceptionV2/InceptionV2/Mixed_4a/Branch_1/Conv2d_0b_3x3/Relu | Relu | [[1 96 28 28]] | 27 | 301056 | 0 | 45475328 | GPU_0_bfc | 0 | 0 | 0 | 0 | void Eigen::internal::EigenMetaKernel<Eigen::TensorEvaluator<Eigen::TensorAssignOp<Eigen::TensorMap<Eigen::Tensor<float, 1, 1, long>, 16, Eigen::MakePointer>, Eigen::TensorCwiseBinaryOp<Eigen::internal::scalar_max_op<float const, float const>, Eigen::TensorMap<Eigen::Tensor<float const, 1, 1, long>, 16, Eigen::MakePointer> const, Eigen::TensorCwiseNullaryOp<Eigen::internal::scalar_constant_op<float const>, Eigen::TensorMap<Eigen::Tensor<float const, 1, 1, long>, 16, Eigen::MakePointer> const> const> const> const, Eigen::GpuDevice>, long>(Eigen::TensorEvaluator<Eigen::TensorAssignOp<Eigen::TensorMap<Eigen::Tensor<float, 1, 1, long>, 16, Eigen::MakePointer>, Eigen::TensorCwiseBinaryOp<Eigen::internal::scalar_max_op<float const, float const>, Eigen::TensorMap<Eigen::Tensor<float const, 1, 1, long>, 16, Eigen::MakePointer> const, Eigen::TensorCwiseNullaryOp<Eigen::internal::scalar_constant_op<float const>, Eigen::TensorMap<Eigen::Tensor<float const, 1, 1, long>, 16, Eigen::MakePointer> const> const> const> const, Eigen::GpuDevice>, long) | 3.00 | 0 | 0.00 | 0.00 | 0.00 | 0.00 | 0.00 | true | 0;0;0;0;0 |
| 66 | InceptionV2/InceptionV2/Mixed_4a/Branch_0/Conv2d_1a_3x3/Relu | Relu | [[1 160 14 14]] | 24.667 | 200704 | 0 | 45475328 | GPU_0_bfc | 0 | 0 | 0 | 0 | void Eigen::internal::EigenMetaKernel<Eigen::TensorEvaluator<Eigen::TensorAssignOp<Eigen::TensorMap<Eigen::Tensor<float, 1, 1, long>, 16, Eigen::MakePointer>, Eigen::TensorCwiseBinaryOp<Eigen::internal::scalar_max_op<float const, float const>, Eigen::TensorMap<Eigen::Tensor<float const, 1, 1, long>, 16, Eigen::MakePointer> const, Eigen::TensorCwiseNullaryOp<Eigen::internal::scalar_constant_op<float const>, Eigen::TensorMap<Eigen::Tensor<float const, 1, 1, long>, 16, Eigen::MakePointer> const> const> const> const, Eigen::GpuDevice>, long>(Eigen::TensorEvaluator<Eigen::TensorAssignOp<Eigen::TensorMap<Eigen::Tensor<float, 1, 1, long>, 16, Eigen::MakePointer>, Eigen::TensorCwiseBinaryOp<Eigen::internal::scalar_max_op<float const, float const>, Eigen::TensorMap<Eigen::Tensor<float const, 1, 1, long>, 16, Eigen::MakePointer> const, Eigen::TensorCwiseNullaryOp<Eigen::internal::scalar_constant_op<float const>, Eigen::TensorMap<Eigen::Tensor<float const, 1, 1, long>, 16, Eigen::MakePointer> const> const> const> const, Eigen::GpuDevice>, long) | 3.00 | 0 | 0.00 | 0.00 | 0.00 | 0.00 | 0.00 | true | 0;0;0;0;0 |
| 67 | InceptionV2/InceptionV2/Mixed_4a/Branch_1/Conv2d_1a_3x3/BatchNorm/batchnorm/mul | Conv2D | [[1 96 14 14]] | 280.333 | 75264 | 1410560 | 45550592 | GPU_0_bfc | 1335296 | 0 | 0 | 0 | void im2col4d_kernel<float, int>(im2col4d_params, cudnnConvolutionStruct, cudnnTensor4dStruct, float const*, float*, int) | 63.17 | 0 | 0.00 | 0.00 | 0.00 | 0.00 | 0.00 | true | 0;0;0;0;0 |
| 67 | InceptionV2/InceptionV2/Mixed_4a/Branch_1/Conv2d_1a_3x3/BatchNorm/batchnorm/mul | Conv2D | [[1 96 14 14]] | 280.333 | 75264 | 1410560 | 45550592 | GPU_0_bfc | 1335296 | 0 | 0 | 0 | void cudnn::detail::explicit_convolve_sgemm<float, int, 128, 5, 5, 3, 3, 3, 0, true>(int, int, int, float const*, int, float const*, int, float*, kernel_conv_params, int, int, float, float, int, float*, float*) | 53.33 | 37177728 | 0.00 | 0.00 | 0.00 | 0.00 | 697.09 | true | 37177728;37177728;37177728;37177728;37177728 |
| 67 | InceptionV2/InceptionV2/Mixed_4a/Branch_1/Conv2d_1a_3x3/BatchNorm/batchnorm/mul | Conv2D | [[1 96 14 14]] | 280.333 | 75264 | 1410560 | 45550592 | GPU_0_bfc | 1335296 | 0 | 0 | 0 | void tensorflow::functor::ShuffleInTensor3Simple<float, 2, 1, 0, false>(int, float const*, tensorflow::functor::Dimension<3>, float*) | 8.00 | 0 | 0.00 | 0.00 | 0.00 | 0.00 | 0.00 | true | 0;0;0;0;0 |
| 67 | InceptionV2/InceptionV2/Mixed_4a/Branch_1/Conv2d_1a_3x3/BatchNorm/batchnorm/mul | Conv2D | [[1 96 14 14]] | 280.333 | 75264 | 1410560 | 45550592 | GPU_0_bfc | 1335296 | 0 | 0 | 0 | void tensorflow::functor::PadInputCustomKernelNCHW<float, 4>(int, float const*, tensorflow::functor::Dimension<4>, float*, tensorflow::functor::Dimension<4>, tensorflow::functor::Dimension<(4)-(2)>) | 7.83 | 0 | 0.00 | 0.00 | 0.00 | 0.00 | 0.00 | true | 0;0;0;0;0 |
| 68 | InceptionV2/InceptionV2/Mixed_4a/Branch_1/Conv2d_1a_3x3/BatchNorm/batchnorm/add_1 | Add | [[1 96 14 14]] | 36.667 | 75264 | 0 | 45249536 | GPU_0_bfc | 0 | 0 | 0 | 0 | void Eigen::internal::EigenMetaKernel<Eigen::TensorEvaluator<Eigen::TensorAssignOp<Eigen::TensorMap<Eigen::Tensor<float, 2, 1, int>, 16, Eigen::MakePointer>, Eigen::TensorCwiseBinaryOp<Eigen::internal::scalar_sum_op<float, float>, Eigen::TensorMap<Eigen::Tensor<float const, 2, 1, int>, 16, Eigen::MakePointer> const, Eigen::TensorBroadcastingOp<Eigen::array<long, 2ul> const, Eigen::TensorMap<Eigen::Tensor<float const, 2, 1, int>, 16, Eigen::MakePointer> const> const> const> const, Eigen::GpuDevice>, int>(Eigen::TensorEvaluator<Eigen::TensorAssignOp<Eigen::TensorMap<Eigen::Tensor<float, 2, 1, int>, 16, Eigen::MakePointer>, Eigen::TensorCwiseBinaryOp<Eigen::internal::scalar_sum_op<float, float>, Eigen::TensorMap<Eigen::Tensor<float const, 2, 1, int>, 16, Eigen::MakePointer> const, Eigen::TensorBroadcastingOp<Eigen::array<long, 2ul> const, Eigen::TensorMap<Eigen::Tensor<float const, 2, 1, int>, 16, Eigen::MakePointer> const> const> const> const, Eigen::GpuDevice>, int) | 4.00 | 18816 | 0.00 | 0.00 | 0.00 | 0.00 | 4.70 | true | 18816;18816;18816;18816;18816 |
| 69 | InceptionV2/InceptionV2/Mixed_4a/Branch_1/Conv2d_1a_3x3/Relu | Relu | [[1 96 14 14]] | 26.667 | 75264 | 0 | 45249536 | GPU_0_bfc | 0 | 0 | 0 | 0 | void Eigen::internal::EigenMetaKernel<Eigen::TensorEvaluator<Eigen::TensorAssignOp<Eigen::TensorMap<Eigen::Tensor<float, 1, 1, long>, 16, Eigen::MakePointer>, Eigen::TensorCwiseBinaryOp<Eigen::internal::scalar_max_op<float const, float const>, Eigen::TensorMap<Eigen::Tensor<float const, 1, 1, long>, 16, Eigen::MakePointer> const, Eigen::TensorCwiseNullaryOp<Eigen::internal::scalar_constant_op<float const>, Eigen::TensorMap<Eigen::Tensor<float const, 1, 1, long>, 16, Eigen::MakePointer> const> const> const> const, Eigen::GpuDevice>, long>(Eigen::TensorEvaluator<Eigen::TensorAssignOp<Eigen::TensorMap<Eigen::Tensor<float, 1, 1, long>, 16, Eigen::MakePointer>, Eigen::TensorCwiseBinaryOp<Eigen::internal::scalar_max_op<float const, float const>, Eigen::TensorMap<Eigen::Tensor<float const, 1, 1, long>, 16, Eigen::MakePointer> const, Eigen::TensorCwiseNullaryOp<Eigen::internal::scalar_constant_op<float const>, Eigen::TensorMap<Eigen::Tensor<float const, 1, 1, long>, 16, Eigen::MakePointer> const> const> const> const, Eigen::GpuDevice>, long) | 3.00 | 0 | 0.00 | 0.00 | 0.00 | 0.00 | 0.00 | true | 0;0;0;0;0 |
| 71 | InceptionV2/InceptionV2/Mixed_4b/Branch_3/AvgPool_0a_3x3/AvgPool | AvgPool | [[1 576 14 14]] | 65.333 | 526848 | 526848 | 45701120 | GPU_0_bfc | 0 | 0 | 0 | 0 | void cudnn::detail::pooling_fw_4d_kernel<float, float, cudnn::detail::averpooling_func<float>, 2, false>(cudnnTensorStruct, float const*, cudnnTensorStruct, float*, cudnnPoolingStruct, float, float, int, cudnn::reduced_divisor, cudnn::reduced_divisor) | 16.67 | 2893038 | 0.00 | 0.00 | 0.00 | 0.00 | 173.58 | true | 2893038;2893038;2893038;2893038;2893038 |
| 72 | InceptionV2/InceptionV2/Mixed_4b/Branch_2/Conv2d_0a_1x1/BatchNorm/batchnorm/mul | Conv2D | [[1 96 14 14]] | 222.333 | 75264 | 297728 | 45776384 | GPU_0_bfc | 222464 | 0 | 0 | 0 | maxwell_scudnn_128x32_relu_interior_nn | 100.83 | 28360704 | 0.00 | 0.00 | 0.00 | 0.00 | 281.26 | true | 28360704;28360704;28360704;28360704;28360704 |
| 72 | InceptionV2/InceptionV2/Mixed_4b/Branch_2/Conv2d_0a_1x1/BatchNorm/batchnorm/mul | Conv2D | [[1 96 14 14]] | 222.333 | 75264 | 297728 | 45776384 | GPU_0_bfc | 222464 | 0 | 0 | 0 | void tensorflow::functor::ShuffleInTensor3Simple<float, 2, 1, 0, false>(int, float const*, tensorflow::functor::Dimension<3>, float*) | 5.33 | 0 | 0.00 | 0.00 | 0.00 | 0.00 | 0.00 | true | 0;0;0;0;0 |
| 72 | InceptionV2/InceptionV2/Mixed_4b/Branch_2/Conv2d_0a_1x1/BatchNorm/batchnorm/mul | Conv2D | [[1 96 14 14]] | 222.333 | 75264 | 297728 | 45776384 | GPU_0_bfc | 222464 | 0 | 0 | 0 | cudnn::maxwell::gemm::computeOffsetsKernel(cudnn::maxwell::gemm::ComputeOffsetsParams) | 2.00 | 0 | 0.00 | 0.00 | 0.00 | 0.00 | 0.00 | true | 0;0;0;0;0 |
| 73 | InceptionV2/InceptionV2/Mixed_4b/Branch_1/Conv2d_0a_1x1/BatchNorm/batchnorm/mul | Conv2D | [[1 64 14 14]] | 245.667 | 50176 | 198912 | 45826560 | GPU_0_bfc | 148736 | 0 | 0 | 0 | maxwell_scudnn_128x64_relu_interior_nn | 131.00 | 18907136 | 0.00 | 0.00 | 0.00 | 0.00 | 144.33 | true | 18907136;18907136;18907136;18907136;18907136 |
| 73 | InceptionV2/InceptionV2/Mixed_4b/Branch_1/Conv2d_0a_1x1/BatchNorm/batchnorm/mul | Conv2D | [[1 64 14 14]] | 245.667 | 50176 | 198912 | 45826560 | GPU_0_bfc | 148736 | 0 | 0 | 0 | void tensorflow::functor::ShuffleInTensor3Simple<float, 2, 1, 0, false>(int, float const*, tensorflow::functor::Dimension<3>, float*) | 5.00 | 0 | 0.00 | 0.00 | 0.00 | 0.00 | 0.00 | true | 0;0;0;0;0 |
| 73 | InceptionV2/InceptionV2/Mixed_4b/Branch_1/Conv2d_0a_1x1/BatchNorm/batchnorm/mul | Conv2D | [[1 64 14 14]] | 245.667 | 50176 | 198912 | 45826560 | GPU_0_bfc | 148736 | 0 | 0 | 0 | cudnn::maxwell::gemm::computeOffsetsKernel(cudnn::maxwell::gemm::ComputeOffsetsParams) | 2.00 | 0 | 0.00 | 0.00 | 0.00 | 0.00 | 0.00 | true | 0;0;0;0;0 |
| 74 | InceptionV2/InceptionV2/Mixed_4b/Branch_0/Conv2d_0a_1x1/BatchNorm/batchnorm/mul | Conv2D | [[1 224 14 14]] | 218.333 | 175616 | 691712 | 46002176 | GPU_0_bfc | 516096 | 0 | 0 | 0 | void cudnn::detail::implicit_convolve_sgemm<float, float, 128, 5, 5, 3, 3, 3, 1, true, false, true>(int, int, int, float const*, int, float*, float*, kernel_conv_params, int, float, float, int, float*, float*, int, int) | 97.67 | 57846656 | 0.00 | 0.00 | 0.00 | 0.00 | 592.28 | true | 57846656;57846656;57846656;57846656;57846656 |
| 74 | InceptionV2/InceptionV2/Mixed_4b/Branch_0/Conv2d_0a_1x1/BatchNorm/batchnorm/mul | Conv2D | [[1 224 14 14]] | 218.333 | 175616 | 691712 | 46002176 | GPU_0_bfc | 516096 | 0 | 0 | 0 | void tensorflow::functor::ShuffleInTensor3Simple<float, 2, 1, 0, false>(int, float const*, tensorflow::functor::Dimension<3>, float*) | 9.00 | 0 | 0.00 | 0.00 | 0.00 | 0.00 | 0.00 | true | 0;0;0;0;0 |
| 75 | InceptionV2/InceptionV2/Mixed_4b/Branch_3/Conv2d_0b_1x1/BatchNorm/batchnorm/mul | Conv2D | [[1 128 14 14]] | 186.667 | 100352 | 451584 | 45650944 | GPU_0_bfc | 351232 | 0 | 0 | 0 | void cudnn::detail::implicit_convolve_sgemm<float, float, 1024, 5, 5, 3, 3, 3, 1, true, false, true>(int, int, int, float const*, int, float*, float*, kernel_conv_params, int, float, float, int, float*, float*, int, int) | 75.50 | 33055232 | 0.00 | 0.00 | 0.00 | 0.00 | 437.82 | true | 33055232;33055232;33055232;33055232;33055232 |
| 75 | InceptionV2/InceptionV2/Mixed_4b/Branch_3/Conv2d_0b_1x1/BatchNorm/batchnorm/mul | Conv2D | [[1 128 14 14]] | 186.667 | 100352 | 451584 | 45650944 | GPU_0_bfc | 351232 | 0 | 0 | 0 | void tensorflow::functor::ShuffleInTensor3Simple<float, 2, 1, 0, false>(int, float const*, tensorflow::functor::Dimension<3>, float*) | 7.00 | 0 | 0.00 | 0.00 | 0.00 | 0.00 | 0.00 | true | 0;0;0;0;0 |
| 76 | InceptionV2/InceptionV2/Mixed_4b/Branch_2/Conv2d_0a_1x1/BatchNorm/batchnorm/add_1 | Add | [[1 96 14 14]] | 37 | 75264 | 0 | 45124096 | GPU_0_bfc | 0 | 0 | 0 | 0 | void Eigen::internal::EigenMetaKernel<Eigen::TensorEvaluator<Eigen::TensorAssignOp<Eigen::TensorMap<Eigen::Tensor<float, 2, 1, int>, 16, Eigen::MakePointer>, Eigen::TensorCwiseBinaryOp<Eigen::internal::scalar_sum_op<float, float>, Eigen::TensorMap<Eigen::Tensor<float const, 2, 1, int>, 16, Eigen::MakePointer> const, Eigen::TensorBroadcastingOp<Eigen::array<long, 2ul> const, Eigen::TensorMap<Eigen::Tensor<float const, 2, 1, int>, 16, Eigen::MakePointer> const> const> const> const, Eigen::GpuDevice>, int>(Eigen::TensorEvaluator<Eigen::TensorAssignOp<Eigen::TensorMap<Eigen::Tensor<float, 2, 1, int>, 16, Eigen::MakePointer>, Eigen::TensorCwiseBinaryOp<Eigen::internal::scalar_sum_op<float, float>, Eigen::TensorMap<Eigen::Tensor<float const, 2, 1, int>, 16, Eigen::MakePointer> const, Eigen::TensorBroadcastingOp<Eigen::array<long, 2ul> const, Eigen::TensorMap<Eigen::Tensor<float const, 2, 1, int>, 16, Eigen::MakePointer> const> const> const> const, Eigen::GpuDevice>, int) | 4.00 | 18816 | 0.00 | 0.00 | 0.00 | 0.00 | 4.70 | true | 18816;18816;18816;18816;18816 |
| 77 | InceptionV2/InceptionV2/Mixed_4b/Branch_1/Conv2d_0a_1x1/BatchNorm/batchnorm/add_1 | Add | [[1 64 14 14]] | 29.333 | 50176 | 0 | 45124096 | GPU_0_bfc | 0 | 0 | 0 | 0 | void Eigen::internal::EigenMetaKernel<Eigen::TensorEvaluator<Eigen::TensorAssignOp<Eigen::TensorMap<Eigen::Tensor<float, 2, 1, int>, 16, Eigen::MakePointer>, Eigen::TensorCwiseBinaryOp<Eigen::internal::scalar_sum_op<float, float>, Eigen::TensorMap<Eigen::Tensor<float const, 2, 1, int>, 16, Eigen::MakePointer> const, Eigen::TensorBroadcastingOp<Eigen::array<long, 2ul> const, Eigen::TensorMap<Eigen::Tensor<float const, 2, 1, int>, 16, Eigen::MakePointer> const> const> const> const, Eigen::GpuDevice>, int>(Eigen::TensorEvaluator<Eigen::TensorAssignOp<Eigen::TensorMap<Eigen::Tensor<float, 2, 1, int>, 16, Eigen::MakePointer>, Eigen::TensorCwiseBinaryOp<Eigen::internal::scalar_sum_op<float, float>, Eigen::TensorMap<Eigen::Tensor<float const, 2, 1, int>, 16, Eigen::MakePointer> const, Eigen::TensorBroadcastingOp<Eigen::array<long, 2ul> const, Eigen::TensorMap<Eigen::Tensor<float const, 2, 1, int>, 16, Eigen::MakePointer> const> const> const> const, Eigen::GpuDevice>, int) | 4.00 | 12544 | 0.00 | 0.00 | 0.00 | 0.00 | 3.14 | true | 12544;12544;12544;12544;12544 |
| 78 | InceptionV2/InceptionV2/Mixed_4b/Branch_0/Conv2d_0a_1x1/BatchNorm/batchnorm/add_1 | Add | [[1 224 14 14]] | 29.333 | 175616 | 0 | 45124096 | GPU_0_bfc | 0 | 0 | 0 | 0 | void Eigen::internal::EigenMetaKernel<Eigen::TensorEvaluator<Eigen::TensorAssignOp<Eigen::TensorMap<Eigen::Tensor<float, 2, 1, int>, 16, Eigen::MakePointer>, Eigen::TensorCwiseBinaryOp<Eigen::internal::scalar_sum_op<float, float>, Eigen::TensorMap<Eigen::Tensor<float const, 2, 1, int>, 16, Eigen::MakePointer> const, Eigen::TensorBroadcastingOp<Eigen::array<long, 2ul> const, Eigen::TensorMap<Eigen::Tensor<float const, 2, 1, int>, 16, Eigen::MakePointer> const> const> const> const, Eigen::GpuDevice>, int>(Eigen::TensorEvaluator<Eigen::TensorAssignOp<Eigen::TensorMap<Eigen::Tensor<float, 2, 1, int>, 16, Eigen::MakePointer>, Eigen::TensorCwiseBinaryOp<Eigen::internal::scalar_sum_op<float, float>, Eigen::TensorMap<Eigen::Tensor<float const, 2, 1, int>, 16, Eigen::MakePointer> const, Eigen::TensorBroadcastingOp<Eigen::array<long, 2ul> const, Eigen::TensorMap<Eigen::Tensor<float const, 2, 1, int>, 16, Eigen::MakePointer> const> const> const> const, Eigen::GpuDevice>, int) | 4.00 | 43904 | 0.00 | 0.00 | 0.00 | 0.00 | 10.98 | true | 43904;43904;43904;43904;43904 |
| 79 | InceptionV2/InceptionV2/Mixed_4b/Branch_3/Conv2d_0b_1x1/BatchNorm/batchnorm/add_1 | Add | [[1 128 14 14]] | 28 | 100352 | 0 | 45124096 | GPU_0_bfc | 0 | 0 | 0 | 0 | void Eigen::internal::EigenMetaKernel<Eigen::TensorEvaluator<Eigen::TensorAssignOp<Eigen::TensorMap<Eigen::Tensor<float, 2, 1, int>, 16, Eigen::MakePointer>, Eigen::TensorCwiseBinaryOp<Eigen::internal::scalar_sum_op<float, float>, Eigen::TensorMap<Eigen::Tensor<float const, 2, 1, int>, 16, Eigen::MakePointer> const, Eigen::TensorBroadcastingOp<Eigen::array<long, 2ul> const, Eigen::TensorMap<Eigen::Tensor<float const, 2, 1, int>, 16, Eigen::MakePointer> const> const> const> const, Eigen::GpuDevice>, int>(Eigen::TensorEvaluator<Eigen::TensorAssignOp<Eigen::TensorMap<Eigen::Tensor<float, 2, 1, int>, 16, Eigen::MakePointer>, Eigen::TensorCwiseBinaryOp<Eigen::internal::scalar_sum_op<float, float>, Eigen::TensorMap<Eigen::Tensor<float const, 2, 1, int>, 16, Eigen::MakePointer> const, Eigen::TensorBroadcastingOp<Eigen::array<long, 2ul> const, Eigen::TensorMap<Eigen::Tensor<float const, 2, 1, int>, 16, Eigen::MakePointer> const> const> const> const, Eigen::GpuDevice>, int) | 4.00 | 25088 | 0.00 | 0.00 | 0.00 | 0.00 | 6.27 | true | 25088;25088;25088;25088;25088 |
| 80 | InceptionV2/InceptionV2/Mixed_4b/Branch_2/Conv2d_0a_1x1/Relu | Relu | [[1 96 14 14]] | 27.333 | 75264 | 0 | 45124096 | GPU_0_bfc | 0 | 0 | 0 | 0 | void Eigen::internal::EigenMetaKernel<Eigen::TensorEvaluator<Eigen::TensorAssignOp<Eigen::TensorMap<Eigen::Tensor<float, 1, 1, long>, 16, Eigen::MakePointer>, Eigen::TensorCwiseBinaryOp<Eigen::internal::scalar_max_op<float const, float const>, Eigen::TensorMap<Eigen::Tensor<float const, 1, 1, long>, 16, Eigen::MakePointer> const, Eigen::TensorCwiseNullaryOp<Eigen::internal::scalar_constant_op<float const>, Eigen::TensorMap<Eigen::Tensor<float const, 1, 1, long>, 16, Eigen::MakePointer> const> const> const> const, Eigen::GpuDevice>, long>(Eigen::TensorEvaluator<Eigen::TensorAssignOp<Eigen::TensorMap<Eigen::Tensor<float, 1, 1, long>, 16, Eigen::MakePointer>, Eigen::TensorCwiseBinaryOp<Eigen::internal::scalar_max_op<float const, float const>, Eigen::TensorMap<Eigen::Tensor<float const, 1, 1, long>, 16, Eigen::MakePointer> const, Eigen::TensorCwiseNullaryOp<Eigen::internal::scalar_constant_op<float const>, Eigen::TensorMap<Eigen::Tensor<float const, 1, 1, long>, 16, Eigen::MakePointer> const> const> const> const, Eigen::GpuDevice>, long) | 3.00 | 0 | 0.00 | 0.00 | 0.00 | 0.00 | 0.00 | true | 0;0;0;0;0 |
| 81 | InceptionV2/InceptionV2/Mixed_4b/Branch_1/Conv2d_0a_1x1/Relu | Relu | [[1 64 14 14]] | 24.333 | 50176 | 0 | 45124096 | GPU_0_bfc | 0 | 0 | 0 | 0 | void Eigen::internal::EigenMetaKernel<Eigen::TensorEvaluator<Eigen::TensorAssignOp<Eigen::TensorMap<Eigen::Tensor<float, 1, 1, long>, 16, Eigen::MakePointer>, Eigen::TensorCwiseBinaryOp<Eigen::internal::scalar_max_op<float const, float const>, Eigen::TensorMap<Eigen::Tensor<float const, 1, 1, long>, 16, Eigen::MakePointer> const, Eigen::TensorCwiseNullaryOp<Eigen::internal::scalar_constant_op<float const>, Eigen::TensorMap<Eigen::Tensor<float const, 1, 1, long>, 16, Eigen::MakePointer> const> const> const> const, Eigen::GpuDevice>, long>(Eigen::TensorEvaluator<Eigen::TensorAssignOp<Eigen::TensorMap<Eigen::Tensor<float, 1, 1, long>, 16, Eigen::MakePointer>, Eigen::TensorCwiseBinaryOp<Eigen::internal::scalar_max_op<float const, float const>, Eigen::TensorMap<Eigen::Tensor<float const, 1, 1, long>, 16, Eigen::MakePointer> const, Eigen::TensorCwiseNullaryOp<Eigen::internal::scalar_constant_op<float const>, Eigen::TensorMap<Eigen::Tensor<float const, 1, 1, long>, 16, Eigen::MakePointer> const> const> const> const, Eigen::GpuDevice>, long) | 3.00 | 0 | 0.00 | 0.00 | 0.00 | 0.00 | 0.00 | true | 0;0;0;0;0 |
| 82 | InceptionV2/InceptionV2/Mixed_4b/Branch_2/Conv2d_0b_3x3/BatchNorm/batchnorm/mul | Conv2D | [[1 128 14 14]] | 213 | 100352 | 1856512 | 45224448 | GPU_0_bfc | 1756160 | 0 | 0 | 0 | maxwell_scudnn_winograd_128x128_ldg1_ldg4_tile148n_nt | 35.00 | 26460160 | 0.00 | 0.00 | 0.00 | 0.00 | 756.00 | true | 26460160;26460160;26460160;26460160;26460160 |
| 82 | InceptionV2/InceptionV2/Mixed_4b/Branch_2/Conv2d_0b_3x3/BatchNorm/batchnorm/mul | Conv2D | [[1 128 14 14]] | 213 | 100352 | 1856512 | 45224448 | GPU_0_bfc | 1756160 | 0 | 0 | 0 | void tensorflow::functor::ShuffleInTensor3Simple<float, 2, 1, 0, false>(int, float const*, tensorflow::functor::Dimension<3>, float*) | 7.67 | 0 | 0.00 | 0.00 | 0.00 | 0.00 | 0.00 | true | 0;0;0;0;0 |
| 82 | InceptionV2/InceptionV2/Mixed_4b/Branch_2/Conv2d_0b_3x3/BatchNorm/batchnorm/mul | Conv2D | [[1 128 14 14]] | 213 | 100352 | 1856512 | 45224448 | GPU_0_bfc | 1756160 | 0 | 0 | 0 | void cudnn::winograd::generateWinogradTilesKernel<0, float, float>(cudnn::winograd::GenerateWinogradTilesParams<float, float>) | 5.50 | 712704 | 0.00 | 0.00 | 0.00 | 0.00 | 129.58 | true | 712704;712704;712704;712704;712704 |
| 83 | InceptionV2/InceptionV2/Mixed_4b/Branch_1/Conv2d_0b_3x3/BatchNorm/batchnorm/mul | Conv2D | [[1 96 14 14]] | 196 | 75264 | 941056 | 45224448 | GPU_0_bfc | 865792 | 0 | 0 | 0 | maxwell_scudnn_winograd_128x128_ldg1_ldg4_tile148n_nt | 25.00 | 13307904 | 0.00 | 0.00 | 0.00 | 0.00 | 532.32 | true | 13307904;13307904;13307904;13307904;13307904 |
| 83 | InceptionV2/InceptionV2/Mixed_4b/Branch_1/Conv2d_0b_3x3/BatchNorm/batchnorm/mul | Conv2D | [[1 96 14 14]] | 196 | 75264 | 941056 | 45224448 | GPU_0_bfc | 865792 | 0 | 0 | 0 | void tensorflow::functor::ShuffleInTensor3Simple<float, 2, 1, 0, false>(int, float const*, tensorflow::functor::Dimension<3>, float*) | 5.83 | 0 | 0.00 | 0.00 | 0.00 | 0.00 | 0.00 | true | 0;0;0;0;0 |
| 83 | InceptionV2/InceptionV2/Mixed_4b/Branch_1/Conv2d_0b_3x3/BatchNorm/batchnorm/mul | Conv2D | [[1 96 14 14]] | 196 | 75264 | 941056 | 45224448 | GPU_0_bfc | 865792 | 0 | 0 | 0 | void cudnn::winograd::generateWinogradTilesKernel<0, float, float>(cudnn::winograd::GenerateWinogradTilesParams<float, float>) | 4.00 | 356352 | 0.00 | 0.00 | 0.00 | 0.00 | 89.09 | true | 356352;356352;356352;356352;356352 |
| 84 | InceptionV2/InceptionV2/Mixed_4b/Branch_2/Conv2d_0b_3x3/BatchNorm/batchnorm/add_1 | Add | [[1 128 14 14]] | 37 | 100352 | 0 | 45174272 | GPU_0_bfc | 0 | 0 | 0 | 0 | void Eigen::internal::EigenMetaKernel<Eigen::TensorEvaluator<Eigen::TensorAssignOp<Eigen::TensorMap<Eigen::Tensor<float, 2, 1, int>, 16, Eigen::MakePointer>, Eigen::TensorCwiseBinaryOp<Eigen::internal::scalar_sum_op<float, float>, Eigen::TensorMap<Eigen::Tensor<float const, 2, 1, int>, 16, Eigen::MakePointer> const, Eigen::TensorBroadcastingOp<Eigen::array<long, 2ul> const, Eigen::TensorMap<Eigen::Tensor<float const, 2, 1, int>, 16, Eigen::MakePointer> const> const> const> const, Eigen::GpuDevice>, int>(Eigen::TensorEvaluator<Eigen::TensorAssignOp<Eigen::TensorMap<Eigen::Tensor<float, 2, 1, int>, 16, Eigen::MakePointer>, Eigen::TensorCwiseBinaryOp<Eigen::internal::scalar_sum_op<float, float>, Eigen::TensorMap<Eigen::Tensor<float const, 2, 1, int>, 16, Eigen::MakePointer> const, Eigen::TensorBroadcastingOp<Eigen::array<long, 2ul> const, Eigen::TensorMap<Eigen::Tensor<float const, 2, 1, int>, 16, Eigen::MakePointer> const> const> const> const, Eigen::GpuDevice>, int) | 4.00 | 25088 | 0.00 | 0.00 | 0.00 | 0.00 | 6.27 | true | 25088;25088;25088;25088;25088 |
| 85 | InceptionV2/InceptionV2/Mixed_4b/Branch_1/Conv2d_0b_3x3/BatchNorm/batchnorm/add_1 | Add | [[1 96 14 14]] | 28.667 | 75264 | 0 | 45174272 | GPU_0_bfc | 0 | 0 | 0 | 0 | void Eigen::internal::EigenMetaKernel<Eigen::TensorEvaluator<Eigen::TensorAssignOp<Eigen::TensorMap<Eigen::Tensor<float, 2, 1, int>, 16, Eigen::MakePointer>, Eigen::TensorCwiseBinaryOp<Eigen::internal::scalar_sum_op<float, float>, Eigen::TensorMap<Eigen::Tensor<float const, 2, 1, int>, 16, Eigen::MakePointer> const, Eigen::TensorBroadcastingOp<Eigen::array<long, 2ul> const, Eigen::TensorMap<Eigen::Tensor<float const, 2, 1, int>, 16, Eigen::MakePointer> const> const> const> const, Eigen::GpuDevice>, int>(Eigen::TensorEvaluator<Eigen::TensorAssignOp<Eigen::TensorMap<Eigen::Tensor<float, 2, 1, int>, 16, Eigen::MakePointer>, Eigen::TensorCwiseBinaryOp<Eigen::internal::scalar_sum_op<float, float>, Eigen::TensorMap<Eigen::Tensor<float const, 2, 1, int>, 16, Eigen::MakePointer> const, Eigen::TensorBroadcastingOp<Eigen::array<long, 2ul> const, Eigen::TensorMap<Eigen::Tensor<float const, 2, 1, int>, 16, Eigen::MakePointer> const> const> const> const, Eigen::GpuDevice>, int) | 4.00 | 18816 | 0.00 | 0.00 | 0.00 | 0.00 | 4.70 | true | 18816;18816;18816;18816;18816 |
| 86 | InceptionV2/InceptionV2/Mixed_4b/Branch_2/Conv2d_0b_3x3/Relu | Relu | [[1 128 14 14]] | 25.667 | 100352 | 0 | 45174272 | GPU_0_bfc | 0 | 0 | 0 | 0 | void Eigen::internal::EigenMetaKernel<Eigen::TensorEvaluator<Eigen::TensorAssignOp<Eigen::TensorMap<Eigen::Tensor<float, 1, 1, long>, 16, Eigen::MakePointer>, Eigen::TensorCwiseBinaryOp<Eigen::internal::scalar_max_op<float const, float const>, Eigen::TensorMap<Eigen::Tensor<float const, 1, 1, long>, 16, Eigen::MakePointer> const, Eigen::TensorCwiseNullaryOp<Eigen::internal::scalar_constant_op<float const>, Eigen::TensorMap<Eigen::Tensor<float const, 1, 1, long>, 16, Eigen::MakePointer> const> const> const> const, Eigen::GpuDevice>, long>(Eigen::TensorEvaluator<Eigen::TensorAssignOp<Eigen::TensorMap<Eigen::Tensor<float, 1, 1, long>, 16, Eigen::MakePointer>, Eigen::TensorCwiseBinaryOp<Eigen::internal::scalar_max_op<float const, float const>, Eigen::TensorMap<Eigen::Tensor<float const, 1, 1, long>, 16, Eigen::MakePointer> const, Eigen::TensorCwiseNullaryOp<Eigen::internal::scalar_constant_op<float const>, Eigen::TensorMap<Eigen::Tensor<float const, 1, 1, long>, 16, Eigen::MakePointer> const> const> const> const, Eigen::GpuDevice>, long) | 3.00 | 0 | 0.00 | 0.00 | 0.00 | 0.00 | 0.00 | true | 0;0;0;0;0 |
| 87 | InceptionV2/InceptionV2/Mixed_4b/Branch_2/Conv2d_0c_3x3/BatchNorm/batchnorm/mul | Conv2D | [[1 128 14 14]] | 217.667 | 100352 | 2329088 | 45274624 | GPU_0_bfc | 2228736 | 0 | 0 | 0 | maxwell_scudnn_winograd_128x128_ldg1_ldg4_tile148n_nt | 42.50 | 35176448 | 0.00 | 0.00 | 0.00 | 0.00 | 827.68 | true | 35176448;35176448;35176448;35176448;35176448 |
| 87 | InceptionV2/InceptionV2/Mixed_4b/Branch_2/Conv2d_0c_3x3/BatchNorm/batchnorm/mul | Conv2D | [[1 128 14 14]] | 217.667 | 100352 | 2329088 | 45274624 | GPU_0_bfc | 2228736 | 0 | 0 | 0 | void tensorflow::functor::ShuffleInTensor3Simple<float, 2, 1, 0, false>(int, float const*, tensorflow::functor::Dimension<3>, float*) | 9.50 | 0 | 0.00 | 0.00 | 0.00 | 0.00 | 0.00 | true | 0;0;0;0;0 |
| 87 | InceptionV2/InceptionV2/Mixed_4b/Branch_2/Conv2d_0c_3x3/BatchNorm/batchnorm/mul | Conv2D | [[1 128 14 14]] | 217.667 | 100352 | 2329088 | 45274624 | GPU_0_bfc | 2228736 | 0 | 0 | 0 | void cudnn::winograd::generateWinogradTilesKernel<0, float, float>(cudnn::winograd::GenerateWinogradTilesParams<float, float>) | 6.00 | 950272 | 0.00 | 0.00 | 0.00 | 0.00 | 158.38 | true | 950272;950272;950272;950272;950272 |
| 88 | InceptionV2/InceptionV2/Mixed_4b/Branch_2/Conv2d_0c_3x3/BatchNorm/batchnorm/add_1 | Add | [[1 128 14 14]] | 34.333 | 100352 | 0 | 45174272 | GPU_0_bfc | 0 | 0 | 0 | 0 | void Eigen::internal::EigenMetaKernel<Eigen::TensorEvaluator<Eigen::TensorAssignOp<Eigen::TensorMap<Eigen::Tensor<float, 2, 1, int>, 16, Eigen::MakePointer>, Eigen::TensorCwiseBinaryOp<Eigen::internal::scalar_sum_op<float, float>, Eigen::TensorMap<Eigen::Tensor<float const, 2, 1, int>, 16, Eigen::MakePointer> const, Eigen::TensorBroadcastingOp<Eigen::array<long, 2ul> const, Eigen::TensorMap<Eigen::Tensor<float const, 2, 1, int>, 16, Eigen::MakePointer> const> const> const> const, Eigen::GpuDevice>, int>(Eigen::TensorEvaluator<Eigen::TensorAssignOp<Eigen::TensorMap<Eigen::Tensor<float, 2, 1, int>, 16, Eigen::MakePointer>, Eigen::TensorCwiseBinaryOp<Eigen::internal::scalar_sum_op<float, float>, Eigen::TensorMap<Eigen::Tensor<float const, 2, 1, int>, 16, Eigen::MakePointer> const, Eigen::TensorBroadcastingOp<Eigen::array<long, 2ul> const, Eigen::TensorMap<Eigen::Tensor<float const, 2, 1, int>, 16, Eigen::MakePointer> const> const> const> const, Eigen::GpuDevice>, int) | 4.00 | 25088 | 0.00 | 0.00 | 0.00 | 0.00 | 6.27 | true | 25088;25088;25088;25088;25088 |
| 90 | InceptionV2/InceptionV2/Mixed_4b/Branch_0/Conv2d_0a_1x1/Relu | Relu | [[1 576 14 14]] | 29.667 | 526848 | 0 | 45249536 | GPU_0_bfc | 0 | 0 | 0 | 0 | void Eigen::internal::EigenMetaKernel<Eigen::TensorEvaluator<Eigen::TensorAssignOp<Eigen::TensorMap<Eigen::Tensor<float, 1, 1, long>, 16, Eigen::MakePointer>, Eigen::TensorCwiseBinaryOp<Eigen::internal::scalar_max_op<float const, float const>, Eigen::TensorMap<Eigen::Tensor<float const, 1, 1, long>, 16, Eigen::MakePointer> const, Eigen::TensorCwiseNullaryOp<Eigen::internal::scalar_constant_op<float const>, Eigen::TensorMap<Eigen::Tensor<float const, 1, 1, long>, 16, Eigen::MakePointer> const> const> const> const, Eigen::GpuDevice>, long>(Eigen::TensorEvaluator<Eigen::TensorAssignOp<Eigen::TensorMap<Eigen::Tensor<float, 1, 1, long>, 16, Eigen::MakePointer>, Eigen::TensorCwiseBinaryOp<Eigen::internal::scalar_max_op<float const, float const>, Eigen::TensorMap<Eigen::Tensor<float const, 1, 1, long>, 16, Eigen::MakePointer> const, Eigen::TensorCwiseNullaryOp<Eigen::internal::scalar_constant_op<float const>, Eigen::TensorMap<Eigen::Tensor<float const, 1, 1, long>, 16, Eigen::MakePointer> const> const> const> const, Eigen::GpuDevice>, long) | 3.00 | 0 | 0.00 | 0.00 | 0.00 | 0.00 | 0.00 | true | 0;0;0;0;0 |
| 91 | InceptionV2/InceptionV2/Mixed_4c/Branch_3/AvgPool_0a_3x3/AvgPool | AvgPool | [[1 576 14 14]] | 66 | 451584 | 451584 | 45701120 | GPU_0_bfc | 0 | 0 | 0 | 0 | void cudnn::detail::pooling_fw_4d_kernel<float, float, cudnn::detail::averpooling_func<float>, 2, false>(cudnnTensorStruct, float const*, cudnnTensorStruct, float*, cudnnPoolingStruct, float, float, int, cudnn::reduced_divisor, cudnn::reduced_divisor) | 18.00 | 2514363 | 0.00 | 0.00 | 0.00 | 0.00 | 139.69 | true | 2514363;2514363;2514363;2514363;2514363 |
| 92 | InceptionV2/InceptionV2/Mixed_4c/Branch_2/Conv2d_0a_1x1/BatchNorm/batchnorm/mul | Conv2D | [[1 96 14 14]] | 217.333 | 75264 | 297728 | 45776384 | GPU_0_bfc | 222464 | 0 | 0 | 0 | maxwell_scudnn_128x32_relu_interior_nn | 100.00 | 28360704 | 0.00 | 0.00 | 0.00 | 0.00 | 283.61 | true | 28360704;28360704;28360704;28360704;28360704 |
| 92 | InceptionV2/InceptionV2/Mixed_4c/Branch_2/Conv2d_0a_1x1/BatchNorm/batchnorm/mul | Conv2D | [[1 96 14 14]] | 217.333 | 75264 | 297728 | 45776384 | GPU_0_bfc | 222464 | 0 | 0 | 0 | void tensorflow::functor::ShuffleInTensor3Simple<float, 2, 1, 0, false>(int, float const*, tensorflow::functor::Dimension<3>, float*) | 5.00 | 0 | 0.00 | 0.00 | 0.00 | 0.00 | 0.00 | true | 0;0;0;0;0 |
| 92 | InceptionV2/InceptionV2/Mixed_4c/Branch_2/Conv2d_0a_1x1/BatchNorm/batchnorm/mul | Conv2D | [[1 96 14 14]] | 217.333 | 75264 | 297728 | 45776384 | GPU_0_bfc | 222464 | 0 | 0 | 0 | cudnn::maxwell::gemm::computeOffsetsKernel(cudnn::maxwell::gemm::ComputeOffsetsParams) | 2.00 | 0 | 0.00 | 0.00 | 0.00 | 0.00 | 0.00 | true | 0;0;0;0;0 |
| 93 | InceptionV2/InceptionV2/Mixed_4c/Branch_1/Conv2d_0a_1x1/BatchNorm/batchnorm/mul | Conv2D | [[1 96 14 14]] | 216 | 75264 | 297728 | 45851648 | GPU_0_bfc | 222464 | 0 | 0 | 0 | maxwell_scudnn_128x32_relu_interior_nn | 99.00 | 28360704 | 0.00 | 0.00 | 0.00 | 0.00 | 286.47 | true | 28360704;28360704;28360704;28360704;28360704 |
| 93 | InceptionV2/InceptionV2/Mixed_4c/Branch_1/Conv2d_0a_1x1/BatchNorm/batchnorm/mul | Conv2D | [[1 96 14 14]] | 216 | 75264 | 297728 | 45851648 | GPU_0_bfc | 222464 | 0 | 0 | 0 | void tensorflow::functor::ShuffleInTensor3Simple<float, 2, 1, 0, false>(int, float const*, tensorflow::functor::Dimension<3>, float*) | 5.50 | 0 | 0.00 | 0.00 | 0.00 | 0.00 | 0.00 | true | 0;0;0;0;0 |
| 93 | InceptionV2/InceptionV2/Mixed_4c/Branch_1/Conv2d_0a_1x1/BatchNorm/batchnorm/mul | Conv2D | [[1 96 14 14]] | 216 | 75264 | 297728 | 45851648 | GPU_0_bfc | 222464 | 0 | 0 | 0 | cudnn::maxwell::gemm::computeOffsetsKernel(cudnn::maxwell::gemm::ComputeOffsetsParams) | 2.00 | 0 | 0.00 | 0.00 | 0.00 | 0.00 | 0.00 | true | 0;0;0;0;0 |
| 94 | InceptionV2/InceptionV2/Mixed_4c/Branch_0/Conv2d_0a_1x1/BatchNorm/batchnorm/mul | Conv2D | [[1 192 14 14]] | 249.667 | 150528 | 594176 | 46002176 | GPU_0_bfc | 443648 | 0 | 0 | 0 | maxwell_scudnn_128x64_relu_interior_nn | 131.00 | 56721408 | 0.00 | 0.00 | 0.00 | 0.00 | 432.99 | true | 56721408;56721408;56721408;56721408;56721408 |
| 94 | InceptionV2/InceptionV2/Mixed_4c/Branch_0/Conv2d_0a_1x1/BatchNorm/batchnorm/mul | Conv2D | [[1 192 14 14]] | 249.667 | 150528 | 594176 | 46002176 | GPU_0_bfc | 443648 | 0 | 0 | 0 | void tensorflow::functor::ShuffleInTensor3Simple<float, 2, 1, 0, false>(int, float const*, tensorflow::functor::Dimension<3>, float*) | 8.00 | 0 | 0.00 | 0.00 | 0.00 | 0.00 | 0.00 | true | 0;0;0;0;0 |
| 94 | InceptionV2/InceptionV2/Mixed_4c/Branch_0/Conv2d_0a_1x1/BatchNorm/batchnorm/mul | Conv2D | [[1 192 14 14]] | 249.667 | 150528 | 594176 | 46002176 | GPU_0_bfc | 443648 | 0 | 0 | 0 | cudnn::maxwell::gemm::computeOffsetsKernel(cudnn::maxwell::gemm::ComputeOffsetsParams) | 2.00 | 0 | 0.00 | 0.00 | 0.00 | 0.00 | 0.00 | true | 0;0;0;0;0 |
| 95 | InceptionV2/InceptionV2/Mixed_4c/Branch_3/Conv2d_0b_1x1/BatchNorm/batchnorm/mul | Conv2D | [[1 128 14 14]] | 181 | 100352 | 526848 | 45575680 | GPU_0_bfc | 426496 | 0 | 0 | 0 | void cudnn::detail::implicit_convolve_sgemm<float, float, 1024, 5, 5, 3, 3, 3, 1, true, false, true>(int, int, int, float const*, int, float*, float*, kernel_conv_params, int, float, float, int, float*, float*, int, int) | 68.17 | 33055232 | 0.00 | 0.00 | 0.00 | 0.00 | 484.92 | true | 33055232;33055232;33055232;33055232;33055232 |
| 95 | InceptionV2/InceptionV2/Mixed_4c/Branch_3/Conv2d_0b_1x1/BatchNorm/batchnorm/mul | Conv2D | [[1 128 14 14]] | 181 | 100352 | 526848 | 45575680 | GPU_0_bfc | 426496 | 0 | 0 | 0 | void tensorflow::functor::ShuffleInTensor3Simple<float, 2, 1, 0, false>(int, float const*, tensorflow::functor::Dimension<3>, float*) | 7.00 | 0 | 0.00 | 0.00 | 0.00 | 0.00 | 0.00 | true | 0;0;0;0;0 |
| 96 | InceptionV2/InceptionV2/Mixed_4c/Branch_2/Conv2d_0a_1x1/BatchNorm/batchnorm/add_1 | Add | [[1 96 14 14]] | 38.333 | 75264 | 0 | 45124096 | GPU_0_bfc | 0 | 0 | 0 | 0 | void Eigen::internal::EigenMetaKernel<Eigen::TensorEvaluator<Eigen::TensorAssignOp<Eigen::TensorMap<Eigen::Tensor<float, 2, 1, int>, 16, Eigen::MakePointer>, Eigen::TensorCwiseBinaryOp<Eigen::internal::scalar_sum_op<float, float>, Eigen::TensorMap<Eigen::Tensor<float const, 2, 1, int>, 16, Eigen::MakePointer> const, Eigen::TensorBroadcastingOp<Eigen::array<long, 2ul> const, Eigen::TensorMap<Eigen::Tensor<float const, 2, 1, int>, 16, Eigen::MakePointer> const> const> const> const, Eigen::GpuDevice>, int>(Eigen::TensorEvaluator<Eigen::TensorAssignOp<Eigen::TensorMap<Eigen::Tensor<float, 2, 1, int>, 16, Eigen::MakePointer>, Eigen::TensorCwiseBinaryOp<Eigen::internal::scalar_sum_op<float, float>, Eigen::TensorMap<Eigen::Tensor<float const, 2, 1, int>, 16, Eigen::MakePointer> const, Eigen::TensorBroadcastingOp<Eigen::array<long, 2ul> const, Eigen::TensorMap<Eigen::Tensor<float const, 2, 1, int>, 16, Eigen::MakePointer> const> const> const> const, Eigen::GpuDevice>, int) | 4.00 | 18816 | 0.00 | 0.00 | 0.00 | 0.00 | 4.70 | true | 18816;18816;18816;18816;18816 |
| 97 | InceptionV2/InceptionV2/Mixed_4c/Branch_1/Conv2d_0a_1x1/BatchNorm/batchnorm/add_1 | Add | [[1 96 14 14]] | 29.667 | 75264 | 0 | 45124096 | GPU_0_bfc | 0 | 0 | 0 | 0 | void Eigen::internal::EigenMetaKernel<Eigen::TensorEvaluator<Eigen::TensorAssignOp<Eigen::TensorMap<Eigen::Tensor<float, 2, 1, int>, 16, Eigen::MakePointer>, Eigen::TensorCwiseBinaryOp<Eigen::internal::scalar_sum_op<float, float>, Eigen::TensorMap<Eigen::Tensor<float const, 2, 1, int>, 16, Eigen::MakePointer> const, Eigen::TensorBroadcastingOp<Eigen::array<long, 2ul> const, Eigen::TensorMap<Eigen::Tensor<float const, 2, 1, int>, 16, Eigen::MakePointer> const> const> const> const, Eigen::GpuDevice>, int>(Eigen::TensorEvaluator<Eigen::TensorAssignOp<Eigen::TensorMap<Eigen::Tensor<float, 2, 1, int>, 16, Eigen::MakePointer>, Eigen::TensorCwiseBinaryOp<Eigen::internal::scalar_sum_op<float, float>, Eigen::TensorMap<Eigen::Tensor<float const, 2, 1, int>, 16, Eigen::MakePointer> const, Eigen::TensorBroadcastingOp<Eigen::array<long, 2ul> const, Eigen::TensorMap<Eigen::Tensor<float const, 2, 1, int>, 16, Eigen::MakePointer> const> const> const> const, Eigen::GpuDevice>, int) | 4.00 | 18816 | 0.00 | 0.00 | 0.00 | 0.00 | 4.70 | true | 18816;18816;18816;18816;18816 |
| 98 | InceptionV2/InceptionV2/Mixed_4c/Branch_0/Conv2d_0a_1x1/BatchNorm/batchnorm/add_1 | Add | [[1 192 14 14]] | 29 | 150528 | 0 | 45124096 | GPU_0_bfc | 0 | 0 | 0 | 0 | void Eigen::internal::EigenMetaKernel<Eigen::TensorEvaluator<Eigen::TensorAssignOp<Eigen::TensorMap<Eigen::Tensor<float, 2, 1, int>, 16, Eigen::MakePointer>, Eigen::TensorCwiseBinaryOp<Eigen::internal::scalar_sum_op<float, float>, Eigen::TensorMap<Eigen::Tensor<float const, 2, 1, int>, 16, Eigen::MakePointer> const, Eigen::TensorBroadcastingOp<Eigen::array<long, 2ul> const, Eigen::TensorMap<Eigen::Tensor<float const, 2, 1, int>, 16, Eigen::MakePointer> const> const> const> const, Eigen::GpuDevice>, int>(Eigen::TensorEvaluator<Eigen::TensorAssignOp<Eigen::TensorMap<Eigen::Tensor<float, 2, 1, int>, 16, Eigen::MakePointer>, Eigen::TensorCwiseBinaryOp<Eigen::internal::scalar_sum_op<float, float>, Eigen::TensorMap<Eigen::Tensor<float const, 2, 1, int>, 16, Eigen::MakePointer> const, Eigen::TensorBroadcastingOp<Eigen::array<long, 2ul> const, Eigen::TensorMap<Eigen::Tensor<float const, 2, 1, int>, 16, Eigen::MakePointer> const> const> const> const, Eigen::GpuDevice>, int) | 4.00 | 37632 | 0.00 | 0.00 | 0.00 | 0.00 | 9.41 | true | 37632;37632;37632;37632;37632 |
| 99 | InceptionV2/InceptionV2/Mixed_4c/Branch_3/Conv2d_0b_1x1/BatchNorm/batchnorm/add_1 | Add | [[1 128 14 14]] | 27.667 | 100352 | 0 | 45124096 | GPU_0_bfc | 0 | 0 | 0 | 0 | void Eigen::internal::EigenMetaKernel<Eigen::TensorEvaluator<Eigen::TensorAssignOp<Eigen::TensorMap<Eigen::Tensor<float, 2, 1, int>, 16, Eigen::MakePointer>, Eigen::TensorCwiseBinaryOp<Eigen::internal::scalar_sum_op<float, float>, Eigen::TensorMap<Eigen::Tensor<float const, 2, 1, int>, 16, Eigen::MakePointer> const, Eigen::TensorBroadcastingOp<Eigen::array<long, 2ul> const, Eigen::TensorMap<Eigen::Tensor<float const, 2, 1, int>, 16, Eigen::MakePointer> const> const> const> const, Eigen::GpuDevice>, int>(Eigen::TensorEvaluator<Eigen::TensorAssignOp<Eigen::TensorMap<Eigen::Tensor<float, 2, 1, int>, 16, Eigen::MakePointer>, Eigen::TensorCwiseBinaryOp<Eigen::internal::scalar_sum_op<float, float>, Eigen::TensorMap<Eigen::Tensor<float const, 2, 1, int>, 16, Eigen::MakePointer> const, Eigen::TensorBroadcastingOp<Eigen::array<long, 2ul> const, Eigen::TensorMap<Eigen::Tensor<float const, 2, 1, int>, 16, Eigen::MakePointer> const> const> const> const, Eigen::GpuDevice>, int) | 4.00 | 25088 | 0.00 | 0.00 | 0.00 | 0.00 | 6.27 | true | 25088;25088;25088;25088;25088 |
| 100 | InceptionV2/InceptionV2/Mixed_4c/Branch_2/Conv2d_0a_1x1/Relu | Relu | [[1 96 14 14]] | 26.667 | 75264 | 0 | 45124096 | GPU_0_bfc | 0 | 0 | 0 | 0 | void Eigen::internal::EigenMetaKernel<Eigen::TensorEvaluator<Eigen::TensorAssignOp<Eigen::TensorMap<Eigen::Tensor<float, 1, 1, long>, 16, Eigen::MakePointer>, Eigen::TensorCwiseBinaryOp<Eigen::internal::scalar_max_op<float const, float const>, Eigen::TensorMap<Eigen::Tensor<float const, 1, 1, long>, 16, Eigen::MakePointer> const, Eigen::TensorCwiseNullaryOp<Eigen::internal::scalar_constant_op<float const>, Eigen::TensorMap<Eigen::Tensor<float const, 1, 1, long>, 16, Eigen::MakePointer> const> const> const> const, Eigen::GpuDevice>, long>(Eigen::TensorEvaluator<Eigen::TensorAssignOp<Eigen::TensorMap<Eigen::Tensor<float, 1, 1, long>, 16, Eigen::MakePointer>, Eigen::TensorCwiseBinaryOp<Eigen::internal::scalar_max_op<float const, float const>, Eigen::TensorMap<Eigen::Tensor<float const, 1, 1, long>, 16, Eigen::MakePointer> const, Eigen::TensorCwiseNullaryOp<Eigen::internal::scalar_constant_op<float const>, Eigen::TensorMap<Eigen::Tensor<float const, 1, 1, long>, 16, Eigen::MakePointer> const> const> const> const, Eigen::GpuDevice>, long) | 3.00 | 0 | 0.00 | 0.00 | 0.00 | 0.00 | 0.00 | true | 0;0;0;0;0 |
| 101 | InceptionV2/InceptionV2/Mixed_4c/Branch_1/Conv2d_0a_1x1/Relu | Relu | [[1 96 14 14]] | 25 | 75264 | 0 | 45124096 | GPU_0_bfc | 0 | 0 | 0 | 0 | void Eigen::internal::EigenMetaKernel<Eigen::TensorEvaluator<Eigen::TensorAssignOp<Eigen::TensorMap<Eigen::Tensor<float, 1, 1, long>, 16, Eigen::MakePointer>, Eigen::TensorCwiseBinaryOp<Eigen::internal::scalar_max_op<float const, float const>, Eigen::TensorMap<Eigen::Tensor<float const, 1, 1, long>, 16, Eigen::MakePointer> const, Eigen::TensorCwiseNullaryOp<Eigen::internal::scalar_constant_op<float const>, Eigen::TensorMap<Eigen::Tensor<float const, 1, 1, long>, 16, Eigen::MakePointer> const> const> const> const, Eigen::GpuDevice>, long>(Eigen::TensorEvaluator<Eigen::TensorAssignOp<Eigen::TensorMap<Eigen::Tensor<float, 1, 1, long>, 16, Eigen::MakePointer>, Eigen::TensorCwiseBinaryOp<Eigen::internal::scalar_max_op<float const, float const>, Eigen::TensorMap<Eigen::Tensor<float const, 1, 1, long>, 16, Eigen::MakePointer> const, Eigen::TensorCwiseNullaryOp<Eigen::internal::scalar_constant_op<float const>, Eigen::TensorMap<Eigen::Tensor<float const, 1, 1, long>, 16, Eigen::MakePointer> const> const> const> const, Eigen::GpuDevice>, long) | 3.00 | 0 | 0.00 | 0.00 | 0.00 | 0.00 | 0.00 | true | 0;0;0;0;0 |
| 102 | InceptionV2/InceptionV2/Mixed_4c/Branch_2/Conv2d_0b_3x3/BatchNorm/batchnorm/mul | Conv2D | [[1 128 14 14]] | 213.333 | 100352 | 2107392 | 45224448 | GPU_0_bfc | 2007040 | 0 | 0 | 0 | maxwell_scudnn_winograd_128x128_ldg1_ldg4_tile148n_nt | 34.00 | 26460160 | 0.00 | 0.00 | 0.00 | 0.00 | 778.24 | true | 26460160;26460160;26460160;26460160;26460160 |
| 102 | InceptionV2/InceptionV2/Mixed_4c/Branch_2/Conv2d_0b_3x3/BatchNorm/batchnorm/mul | Conv2D | [[1 128 14 14]] | 213.333 | 100352 | 2107392 | 45224448 | GPU_0_bfc | 2007040 | 0 | 0 | 0 | void tensorflow::functor::ShuffleInTensor3Simple<float, 2, 1, 0, false>(int, float const*, tensorflow::functor::Dimension<3>, float*) | 7.67 | 0 | 0.00 | 0.00 | 0.00 | 0.00 | 0.00 | true | 0;0;0;0;0 |
| 102 | InceptionV2/InceptionV2/Mixed_4c/Branch_2/Conv2d_0b_3x3/BatchNorm/batchnorm/mul | Conv2D | [[1 128 14 14]] | 213.333 | 100352 | 2107392 | 45224448 | GPU_0_bfc | 2007040 | 0 | 0 | 0 | void cudnn::winograd::generateWinogradTilesKernel<0, float, float>(cudnn::winograd::GenerateWinogradTilesParams<float, float>) | 5.00 | 712704 | 0.00 | 0.00 | 0.00 | 0.00 | 142.54 | true | 712704;712704;712704;712704;712704 |
| 103 | InceptionV2/InceptionV2/Mixed_4c/Branch_1/Conv2d_0b_3x3/BatchNorm/batchnorm/mul | Conv2D | [[1 128 14 14]] | 205.667 | 100352 | 2082304 | 45249536 | GPU_0_bfc | 1981952 | 0 | 0 | 0 | maxwell_scudnn_winograd_128x128_ldg1_ldg4_tile148n_nt | 34.00 | 26460160 | 0.00 | 0.00 | 0.00 | 0.00 | 778.24 | true | 26460160;26460160;26460160;26460160;26460160 |
| 103 | InceptionV2/InceptionV2/Mixed_4c/Branch_1/Conv2d_0b_3x3/BatchNorm/batchnorm/mul | Conv2D | [[1 128 14 14]] | 205.667 | 100352 | 2082304 | 45249536 | GPU_0_bfc | 1981952 | 0 | 0 | 0 | void tensorflow::functor::ShuffleInTensor3Simple<float, 2, 1, 0, false>(int, float const*, tensorflow::functor::Dimension<3>, float*) | 7.50 | 0 | 0.00 | 0.00 | 0.00 | 0.00 | 0.00 | true | 0;0;0;0;0 |
| 103 | InceptionV2/InceptionV2/Mixed_4c/Branch_1/Conv2d_0b_3x3/BatchNorm/batchnorm/mul | Conv2D | [[1 128 14 14]] | 205.667 | 100352 | 2082304 | 45249536 | GPU_0_bfc | 1981952 | 0 | 0 | 0 | void cudnn::winograd::generateWinogradTilesKernel<0, float, float>(cudnn::winograd::GenerateWinogradTilesParams<float, float>) | 5.00 | 712704 | 0.00 | 0.00 | 0.00 | 0.00 | 142.54 | true | 712704;712704;712704;712704;712704 |
| 104 | InceptionV2/InceptionV2/Mixed_4c/Branch_2/Conv2d_0b_3x3/BatchNorm/batchnorm/add_1 | Add | [[1 128 14 14]] | 37 | 100352 | 0 | 45174272 | GPU_0_bfc | 0 | 0 | 0 | 0 | void Eigen::internal::EigenMetaKernel<Eigen::TensorEvaluator<Eigen::TensorAssignOp<Eigen::TensorMap<Eigen::Tensor<float, 2, 1, int>, 16, Eigen::MakePointer>, Eigen::TensorCwiseBinaryOp<Eigen::internal::scalar_sum_op<float, float>, Eigen::TensorMap<Eigen::Tensor<float const, 2, 1, int>, 16, Eigen::MakePointer> const, Eigen::TensorBroadcastingOp<Eigen::array<long, 2ul> const, Eigen::TensorMap<Eigen::Tensor<float const, 2, 1, int>, 16, Eigen::MakePointer> const> const> const> const, Eigen::GpuDevice>, int>(Eigen::TensorEvaluator<Eigen::TensorAssignOp<Eigen::TensorMap<Eigen::Tensor<float, 2, 1, int>, 16, Eigen::MakePointer>, Eigen::TensorCwiseBinaryOp<Eigen::internal::scalar_sum_op<float, float>, Eigen::TensorMap<Eigen::Tensor<float const, 2, 1, int>, 16, Eigen::MakePointer> const, Eigen::TensorBroadcastingOp<Eigen::array<long, 2ul> const, Eigen::TensorMap<Eigen::Tensor<float const, 2, 1, int>, 16, Eigen::MakePointer> const> const> const> const, Eigen::GpuDevice>, int) | 4.00 | 25088 | 0.00 | 0.00 | 0.00 | 0.00 | 6.27 | true | 25088;25088;25088;25088;25088 |
| 105 | InceptionV2/InceptionV2/Mixed_4c/Branch_1/Conv2d_0b_3x3/BatchNorm/batchnorm/add_1 | Add | [[1 128 14 14]] | 27.667 | 100352 | 0 | 45174272 | GPU_0_bfc | 0 | 0 | 0 | 0 | void Eigen::internal::EigenMetaKernel<Eigen::TensorEvaluator<Eigen::TensorAssignOp<Eigen::TensorMap<Eigen::Tensor<float, 2, 1, int>, 16, Eigen::MakePointer>, Eigen::TensorCwiseBinaryOp<Eigen::internal::scalar_sum_op<float, float>, Eigen::TensorMap<Eigen::Tensor<float const, 2, 1, int>, 16, Eigen::MakePointer> const, Eigen::TensorBroadcastingOp<Eigen::array<long, 2ul> const, Eigen::TensorMap<Eigen::Tensor<float const, 2, 1, int>, 16, Eigen::MakePointer> const> const> const> const, Eigen::GpuDevice>, int>(Eigen::TensorEvaluator<Eigen::TensorAssignOp<Eigen::TensorMap<Eigen::Tensor<float, 2, 1, int>, 16, Eigen::MakePointer>, Eigen::TensorCwiseBinaryOp<Eigen::internal::scalar_sum_op<float, float>, Eigen::TensorMap<Eigen::Tensor<float const, 2, 1, int>, 16, Eigen::MakePointer> const, Eigen::TensorBroadcastingOp<Eigen::array<long, 2ul> const, Eigen::TensorMap<Eigen::Tensor<float const, 2, 1, int>, 16, Eigen::MakePointer> const> const> const> const, Eigen::GpuDevice>, int) | 4.00 | 25088 | 0.00 | 0.00 | 0.00 | 0.00 | 6.27 | true | 25088;25088;25088;25088;25088 |
| 106 | InceptionV2/InceptionV2/Mixed_4c/Branch_2/Conv2d_0b_3x3/Relu | Relu | [[1 128 14 14]] | 24.667 | 100352 | 0 | 45174272 | GPU_0_bfc | 0 | 0 | 0 | 0 | void Eigen::internal::EigenMetaKernel<Eigen::TensorEvaluator<Eigen::TensorAssignOp<Eigen::TensorMap<Eigen::Tensor<float, 1, 1, long>, 16, Eigen::MakePointer>, Eigen::TensorCwiseBinaryOp<Eigen::internal::scalar_max_op<float const, float const>, Eigen::TensorMap<Eigen::Tensor<float const, 1, 1, long>, 16, Eigen::MakePointer> const, Eigen::TensorCwiseNullaryOp<Eigen::internal::scalar_constant_op<float const>, Eigen::TensorMap<Eigen::Tensor<float const, 1, 1, long>, 16, Eigen::MakePointer> const> const> const> const, Eigen::GpuDevice>, long>(Eigen::TensorEvaluator<Eigen::TensorAssignOp<Eigen::TensorMap<Eigen::Tensor<float, 1, 1, long>, 16, Eigen::MakePointer>, Eigen::TensorCwiseBinaryOp<Eigen::internal::scalar_max_op<float const, float const>, Eigen::TensorMap<Eigen::Tensor<float const, 1, 1, long>, 16, Eigen::MakePointer> const, Eigen::TensorCwiseNullaryOp<Eigen::internal::scalar_constant_op<float const>, Eigen::TensorMap<Eigen::Tensor<float const, 1, 1, long>, 16, Eigen::MakePointer> const> const> const> const, Eigen::GpuDevice>, long) | 3.00 | 0 | 0.00 | 0.00 | 0.00 | 0.00 | 0.00 | true | 0;0;0;0;0 |
| 107 | InceptionV2/InceptionV2/Mixed_4c/Branch_2/Conv2d_0c_3x3/BatchNorm/batchnorm/mul | Conv2D | [[1 128 14 14]] | 208.667 | 100352 | 2466816 | 45274624 | GPU_0_bfc | 2366464 | 0 | 0 | 0 | maxwell_scudnn_winograd_128x128_ldg1_ldg4_tile148n_nt | 42.50 | 35176448 | 0.00 | 0.00 | 0.00 | 0.00 | 827.68 | true | 35176448;35176448;35176448;35176448;35176448 |
| 107 | InceptionV2/InceptionV2/Mixed_4c/Branch_2/Conv2d_0c_3x3/BatchNorm/batchnorm/mul | Conv2D | [[1 128 14 14]] | 208.667 | 100352 | 2466816 | 45274624 | GPU_0_bfc | 2366464 | 0 | 0 | 0 | void tensorflow::functor::ShuffleInTensor3Simple<float, 2, 1, 0, false>(int, float const*, tensorflow::functor::Dimension<3>, float*) | 9.17 | 0 | 0.00 | 0.00 | 0.00 | 0.00 | 0.00 | true | 0;0;0;0;0 |
| 107 | InceptionV2/InceptionV2/Mixed_4c/Branch_2/Conv2d_0c_3x3/BatchNorm/batchnorm/mul | Conv2D | [[1 128 14 14]] | 208.667 | 100352 | 2466816 | 45274624 | GPU_0_bfc | 2366464 | 0 | 0 | 0 | void cudnn::winograd::generateWinogradTilesKernel<0, float, float>(cudnn::winograd::GenerateWinogradTilesParams<float, float>) | 6.00 | 950272 | 0.00 | 0.00 | 0.00 | 0.00 | 158.38 | true | 950272;950272;950272;950272;950272 |
| 108 | InceptionV2/InceptionV2/Mixed_4c/Branch_2/Conv2d_0c_3x3/BatchNorm/batchnorm/add_1 | Add | [[1 128 14 14]] | 33.333 | 100352 | 0 | 45174272 | GPU_0_bfc | 0 | 0 | 0 | 0 | void Eigen::internal::EigenMetaKernel<Eigen::TensorEvaluator<Eigen::TensorAssignOp<Eigen::TensorMap<Eigen::Tensor<float, 2, 1, int>, 16, Eigen::MakePointer>, Eigen::TensorCwiseBinaryOp<Eigen::internal::scalar_sum_op<float, float>, Eigen::TensorMap<Eigen::Tensor<float const, 2, 1, int>, 16, Eigen::MakePointer> const, Eigen::TensorBroadcastingOp<Eigen::array<long, 2ul> const, Eigen::TensorMap<Eigen::Tensor<float const, 2, 1, int>, 16, Eigen::MakePointer> const> const> const> const, Eigen::GpuDevice>, int>(Eigen::TensorEvaluator<Eigen::TensorAssignOp<Eigen::TensorMap<Eigen::Tensor<float, 2, 1, int>, 16, Eigen::MakePointer>, Eigen::TensorCwiseBinaryOp<Eigen::internal::scalar_sum_op<float, float>, Eigen::TensorMap<Eigen::Tensor<float const, 2, 1, int>, 16, Eigen::MakePointer> const, Eigen::TensorBroadcastingOp<Eigen::array<long, 2ul> const, Eigen::TensorMap<Eigen::Tensor<float const, 2, 1, int>, 16, Eigen::MakePointer> const> const> const> const, Eigen::GpuDevice>, int) | 4.00 | 25088 | 0.00 | 0.00 | 0.00 | 0.00 | 6.27 | true | 25088;25088;25088;25088;25088 |
| 110 | InceptionV2/InceptionV2/Mixed_4c/Branch_0/Conv2d_0a_1x1/Relu | Relu | [[1 576 14 14]] | 28.667 | 727552 | 0 | 45450240 | GPU_0_bfc | 0 | 0 | 0 | 0 | void Eigen::internal::EigenMetaKernel<Eigen::TensorEvaluator<Eigen::TensorAssignOp<Eigen::TensorMap<Eigen::Tensor<float, 1, 1, long>, 16, Eigen::MakePointer>, Eigen::TensorCwiseBinaryOp<Eigen::internal::scalar_max_op<float const, float const>, Eigen::TensorMap<Eigen::Tensor<float const, 1, 1, long>, 16, Eigen::MakePointer> const, Eigen::TensorCwiseNullaryOp<Eigen::internal::scalar_constant_op<float const>, Eigen::TensorMap<Eigen::Tensor<float const, 1, 1, long>, 16, Eigen::MakePointer> const> const> const> const, Eigen::GpuDevice>, long>(Eigen::TensorEvaluator<Eigen::TensorAssignOp<Eigen::TensorMap<Eigen::Tensor<float, 1, 1, long>, 16, Eigen::MakePointer>, Eigen::TensorCwiseBinaryOp<Eigen::internal::scalar_max_op<float const, float const>, Eigen::TensorMap<Eigen::Tensor<float const, 1, 1, long>, 16, Eigen::MakePointer> const, Eigen::TensorCwiseNullaryOp<Eigen::internal::scalar_constant_op<float const>, Eigen::TensorMap<Eigen::Tensor<float const, 1, 1, long>, 16, Eigen::MakePointer> const> const> const> const, Eigen::GpuDevice>, long) | 3.00 | 0 | 0.00 | 0.00 | 0.00 | 0.00 | 0.00 | true | 0;0;0;0;0 |
| 111 | InceptionV2/InceptionV2/Mixed_4d/Branch_3/AvgPool_0a_3x3/AvgPool | AvgPool | [[1 576 14 14]] | 64.667 | 451584 | 451584 | 45901824 | GPU_0_bfc | 0 | 0 | 0 | 0 | void cudnn::detail::pooling_fw_4d_kernel<float, float, cudnn::detail::averpooling_func<float>, 2, false>(cudnnTensorStruct, float const*, cudnnTensorStruct, float*, cudnnPoolingStruct, float, float, int, cudnn::reduced_divisor, cudnn::reduced_divisor) | 18.00 | 2461563 | 0.00 | 0.00 | 0.00 | 0.00 | 136.75 | true | 2461563;2461563;2461563;2461563;2461563 |
| 112 | InceptionV2/InceptionV2/Mixed_4d/Branch_2/Conv2d_0a_1x1/BatchNorm/batchnorm/mul | Conv2D | [[1 128 14 14]] | 169.333 | 100352 | 401408 | 46002176 | GPU_0_bfc | 301056 | 0 | 0 | 0 | void cudnn::detail::implicit_convolve_sgemm<float, float, 1024, 5, 5, 3, 3, 3, 1, true, false, true>(int, int, int, float const*, int, float*, float*, kernel_conv_params, int, float, float, int, float*, float*, int, int) | 68.17 | 33055232 | 0.00 | 0.00 | 0.00 | 0.00 | 484.92 | true | 33055232;33055232;33055232;33055232;33055232 |
| 112 | InceptionV2/InceptionV2/Mixed_4d/Branch_2/Conv2d_0a_1x1/BatchNorm/batchnorm/mul | Conv2D | [[1 128 14 14]] | 169.333 | 100352 | 401408 | 46002176 | GPU_0_bfc | 301056 | 0 | 0 | 0 | void tensorflow::functor::ShuffleInTensor3Simple<float, 2, 1, 0, false>(int, float const*, tensorflow::functor::Dimension<3>, float*) | 7.00 | 0 | 0.00 | 0.00 | 0.00 | 0.00 | 0.00 | true | 0;0;0;0;0 |
| 113 | InceptionV2/InceptionV2/Mixed_4d/Branch_1/Conv2d_0a_1x1/BatchNorm/batchnorm/mul | Conv2D | [[1 128 14 14]] | 179.667 | 100352 | 395264 | 46102528 | GPU_0_bfc | 294912 | 0 | 0 | 0 | void cudnn::detail::implicit_convolve_sgemm<float, float, 1024, 5, 5, 3, 3, 3, 1, true, false, true>(int, int, int, float const*, int, float*, float*, kernel_conv_params, int, float, float, int, float*, float*, int, int) | 68.17 | 33055232 | 0.00 | 0.00 | 0.00 | 0.00 | 484.92 | true | 33055232;33055232;33055232;33055232;33055232 |
| 113 | InceptionV2/InceptionV2/Mixed_4d/Branch_1/Conv2d_0a_1x1/BatchNorm/batchnorm/mul | Conv2D | [[1 128 14 14]] | 179.667 | 100352 | 395264 | 46102528 | GPU_0_bfc | 294912 | 0 | 0 | 0 | void tensorflow::functor::ShuffleInTensor3Simple<float, 2, 1, 0, false>(int, float const*, tensorflow::functor::Dimension<3>, float*) | 7.00 | 0 | 0.00 | 0.00 | 0.00 | 0.00 | 0.00 | true | 0;0;0;0;0 |
| 114 | InceptionV2/InceptionV2/Mixed_4d/Branch_0/Conv2d_0a_1x1/BatchNorm/batchnorm/mul | Conv2D | [[1 160 14 14]] | 249 | 200704 | 570624 | 46303232 | GPU_0_bfc | 369920 | 0 | 0 | 0 | maxwell_scudnn_128x64_relu_interior_nn | 130.33 | 56721408 | 0.00 | 0.00 | 0.00 | 0.00 | 435.20 | true | 56721408;56721408;56721408;56721408;56721408 |
| 114 | InceptionV2/InceptionV2/Mixed_4d/Branch_0/Conv2d_0a_1x1/BatchNorm/batchnorm/mul | Conv2D | [[1 160 14 14]] | 249 | 200704 | 570624 | 46303232 | GPU_0_bfc | 369920 | 0 | 0 | 0 | void tensorflow::functor::ShuffleInTensor3Simple<float, 2, 1, 0, false>(int, float const*, tensorflow::functor::Dimension<3>, float*) | 8.00 | 0 | 0.00 | 0.00 | 0.00 | 0.00 | 0.00 | true | 0;0;0;0;0 |
| 114 | InceptionV2/InceptionV2/Mixed_4d/Branch_0/Conv2d_0a_1x1/BatchNorm/batchnorm/mul | Conv2D | [[1 160 14 14]] | 249 | 200704 | 570624 | 46303232 | GPU_0_bfc | 369920 | 0 | 0 | 0 | cudnn::maxwell::gemm::computeOffsetsKernel(cudnn::maxwell::gemm::ComputeOffsetsParams) | 2.00 | 0 | 0.00 | 0.00 | 0.00 | 0.00 | 0.00 | true | 0;0;0;0;0 |
| 115 | InceptionV2/InceptionV2/Mixed_4d/Branch_3/Conv2d_0b_1x1/BatchNorm/batchnorm/mul | Conv2D | [[1 96 14 14]] | 218 | 75264 | 297728 | 45650944 | GPU_0_bfc | 222464 | 0 | 0 | 0 | maxwell_scudnn_128x32_relu_interior_nn | 99.00 | 28360704 | 0.00 | 0.00 | 0.00 | 0.00 | 286.47 | true | 28360704;28360704;28360704;28360704;28360704 |
| 115 | InceptionV2/InceptionV2/Mixed_4d/Branch_3/Conv2d_0b_1x1/BatchNorm/batchnorm/mul | Conv2D | [[1 96 14 14]] | 218 | 75264 | 297728 | 45650944 | GPU_0_bfc | 222464 | 0 | 0 | 0 | void tensorflow::functor::ShuffleInTensor3Simple<float, 2, 1, 0, false>(int, float const*, tensorflow::functor::Dimension<3>, float*) | 5.00 | 0 | 0.00 | 0.00 | 0.00 | 0.00 | 0.00 | true | 0;0;0;0;0 |
| 115 | InceptionV2/InceptionV2/Mixed_4d/Branch_3/Conv2d_0b_1x1/BatchNorm/batchnorm/mul | Conv2D | [[1 96 14 14]] | 218 | 75264 | 297728 | 45650944 | GPU_0_bfc | 222464 | 0 | 0 | 0 | cudnn::maxwell::gemm::computeOffsetsKernel(cudnn::maxwell::gemm::ComputeOffsetsParams) | 2.00 | 0 | 0.00 | 0.00 | 0.00 | 0.00 | 0.00 | true | 0;0;0;0;0 |
| 116 | InceptionV2/InceptionV2/Mixed_4d/Branch_2/Conv2d_0a_1x1/BatchNorm/batchnorm/add_1 | Add | [[1 128 14 14]] | 35.333 | 100352 | 0 | 45199360 | GPU_0_bfc | 0 | 0 | 0 | 0 | void Eigen::internal::EigenMetaKernel<Eigen::TensorEvaluator<Eigen::TensorAssignOp<Eigen::TensorMap<Eigen::Tensor<float, 2, 1, int>, 16, Eigen::MakePointer>, Eigen::TensorCwiseBinaryOp<Eigen::internal::scalar_sum_op<float, float>, Eigen::TensorMap<Eigen::Tensor<float const, 2, 1, int>, 16, Eigen::MakePointer> const, Eigen::TensorBroadcastingOp<Eigen::array<long, 2ul> const, Eigen::TensorMap<Eigen::Tensor<float const, 2, 1, int>, 16, Eigen::MakePointer> const> const> const> const, Eigen::GpuDevice>, int>(Eigen::TensorEvaluator<Eigen::TensorAssignOp<Eigen::TensorMap<Eigen::Tensor<float, 2, 1, int>, 16, Eigen::MakePointer>, Eigen::TensorCwiseBinaryOp<Eigen::internal::scalar_sum_op<float, float>, Eigen::TensorMap<Eigen::Tensor<float const, 2, 1, int>, 16, Eigen::MakePointer> const, Eigen::TensorBroadcastingOp<Eigen::array<long, 2ul> const, Eigen::TensorMap<Eigen::Tensor<float const, 2, 1, int>, 16, Eigen::MakePointer> const> const> const> const, Eigen::GpuDevice>, int) | 4.00 | 25088 | 0.00 | 0.00 | 0.00 | 0.00 | 6.27 | true | 25088;25088;25088;25088;25088 |
| 117 | InceptionV2/InceptionV2/Mixed_4d/Branch_1/Conv2d_0a_1x1/BatchNorm/batchnorm/add_1 | Add | [[1 128 14 14]] | 28.333 | 100352 | 0 | 45199360 | GPU_0_bfc | 0 | 0 | 0 | 0 | void Eigen::internal::EigenMetaKernel<Eigen::TensorEvaluator<Eigen::TensorAssignOp<Eigen::TensorMap<Eigen::Tensor<float, 2, 1, int>, 16, Eigen::MakePointer>, Eigen::TensorCwiseBinaryOp<Eigen::internal::scalar_sum_op<float, float>, Eigen::TensorMap<Eigen::Tensor<float const, 2, 1, int>, 16, Eigen::MakePointer> const, Eigen::TensorBroadcastingOp<Eigen::array<long, 2ul> const, Eigen::TensorMap<Eigen::Tensor<float const, 2, 1, int>, 16, Eigen::MakePointer> const> const> const> const, Eigen::GpuDevice>, int>(Eigen::TensorEvaluator<Eigen::TensorAssignOp<Eigen::TensorMap<Eigen::Tensor<float, 2, 1, int>, 16, Eigen::MakePointer>, Eigen::TensorCwiseBinaryOp<Eigen::internal::scalar_sum_op<float, float>, Eigen::TensorMap<Eigen::Tensor<float const, 2, 1, int>, 16, Eigen::MakePointer> const, Eigen::TensorBroadcastingOp<Eigen::array<long, 2ul> const, Eigen::TensorMap<Eigen::Tensor<float const, 2, 1, int>, 16, Eigen::MakePointer> const> const> const> const, Eigen::GpuDevice>, int) | 4.00 | 25088 | 0.00 | 0.00 | 0.00 | 0.00 | 6.27 | true | 25088;25088;25088;25088;25088 |
| 118 | InceptionV2/InceptionV2/Mixed_4d/Branch_0/Conv2d_0a_1x1/BatchNorm/batchnorm/add_1 | Add | [[1 160 14 14]] | 28 | 200704 | 0 | 45199360 | GPU_0_bfc | 0 | 0 | 0 | 0 | void Eigen::internal::EigenMetaKernel<Eigen::TensorEvaluator<Eigen::TensorAssignOp<Eigen::TensorMap<Eigen::Tensor<float, 2, 1, int>, 16, Eigen::MakePointer>, Eigen::TensorCwiseBinaryOp<Eigen::internal::scalar_sum_op<float, float>, Eigen::TensorMap<Eigen::Tensor<float const, 2, 1, int>, 16, Eigen::MakePointer> const, Eigen::TensorBroadcastingOp<Eigen::array<long, 2ul> const, Eigen::TensorMap<Eigen::Tensor<float const, 2, 1, int>, 16, Eigen::MakePointer> const> const> const> const, Eigen::GpuDevice>, int>(Eigen::TensorEvaluator<Eigen::TensorAssignOp<Eigen::TensorMap<Eigen::Tensor<float, 2, 1, int>, 16, Eigen::MakePointer>, Eigen::TensorCwiseBinaryOp<Eigen::internal::scalar_sum_op<float, float>, Eigen::TensorMap<Eigen::Tensor<float const, 2, 1, int>, 16, Eigen::MakePointer> const, Eigen::TensorBroadcastingOp<Eigen::array<long, 2ul> const, Eigen::TensorMap<Eigen::Tensor<float const, 2, 1, int>, 16, Eigen::MakePointer> const> const> const> const, Eigen::GpuDevice>, int) | 4.00 | 31360 | 0.00 | 0.00 | 0.00 | 0.00 | 7.84 | true | 31360;31360;31360;31360;31360 |
| 119 | InceptionV2/InceptionV2/Mixed_4d/Branch_3/Conv2d_0b_1x1/BatchNorm/batchnorm/add_1 | Add | [[1 96 14 14]] | 27.333 | 75264 | 0 | 45199360 | GPU_0_bfc | 0 | 0 | 0 | 0 | void Eigen::internal::EigenMetaKernel<Eigen::TensorEvaluator<Eigen::TensorAssignOp<Eigen::TensorMap<Eigen::Tensor<float, 2, 1, int>, 16, Eigen::MakePointer>, Eigen::TensorCwiseBinaryOp<Eigen::internal::scalar_sum_op<float, float>, Eigen::TensorMap<Eigen::Tensor<float const, 2, 1, int>, 16, Eigen::MakePointer> const, Eigen::TensorBroadcastingOp<Eigen::array<long, 2ul> const, Eigen::TensorMap<Eigen::Tensor<float const, 2, 1, int>, 16, Eigen::MakePointer> const> const> const> const, Eigen::GpuDevice>, int>(Eigen::TensorEvaluator<Eigen::TensorAssignOp<Eigen::TensorMap<Eigen::Tensor<float, 2, 1, int>, 16, Eigen::MakePointer>, Eigen::TensorCwiseBinaryOp<Eigen::internal::scalar_sum_op<float, float>, Eigen::TensorMap<Eigen::Tensor<float const, 2, 1, int>, 16, Eigen::MakePointer> const, Eigen::TensorBroadcastingOp<Eigen::array<long, 2ul> const, Eigen::TensorMap<Eigen::Tensor<float const, 2, 1, int>, 16, Eigen::MakePointer> const> const> const> const, Eigen::GpuDevice>, int) | 4.00 | 18816 | 0.00 | 0.00 | 0.00 | 0.00 | 4.70 | true | 18816;18816;18816;18816;18816 |
| 120 | InceptionV2/InceptionV2/Mixed_4d/Branch_2/Conv2d_0a_1x1/Relu | Relu | [[1 128 14 14]] | 24.667 | 100352 | 0 | 45199360 | GPU_0_bfc | 0 | 0 | 0 | 0 | void Eigen::internal::EigenMetaKernel<Eigen::TensorEvaluator<Eigen::TensorAssignOp<Eigen::TensorMap<Eigen::Tensor<float, 1, 1, long>, 16, Eigen::MakePointer>, Eigen::TensorCwiseBinaryOp<Eigen::internal::scalar_max_op<float const, float const>, Eigen::TensorMap<Eigen::Tensor<float const, 1, 1, long>, 16, Eigen::MakePointer> const, Eigen::TensorCwiseNullaryOp<Eigen::internal::scalar_constant_op<float const>, Eigen::TensorMap<Eigen::Tensor<float const, 1, 1, long>, 16, Eigen::MakePointer> const> const> const> const, Eigen::GpuDevice>, long>(Eigen::TensorEvaluator<Eigen::TensorAssignOp<Eigen::TensorMap<Eigen::Tensor<float, 1, 1, long>, 16, Eigen::MakePointer>, Eigen::TensorCwiseBinaryOp<Eigen::internal::scalar_max_op<float const, float const>, Eigen::TensorMap<Eigen::Tensor<float const, 1, 1, long>, 16, Eigen::MakePointer> const, Eigen::TensorCwiseNullaryOp<Eigen::internal::scalar_constant_op<float const>, Eigen::TensorMap<Eigen::Tensor<float const, 1, 1, long>, 16, Eigen::MakePointer> const> const> const> const, Eigen::GpuDevice>, long) | 3.00 | 0 | 0.00 | 0.00 | 0.00 | 0.00 | 0.00 | true | 0;0;0;0;0 |
| 121 | InceptionV2/InceptionV2/Mixed_4d/Branch_1/Conv2d_0a_1x1/Relu | Relu | [[1 128 14 14]] | 23.333 | 100352 | 0 | 45199360 | GPU_0_bfc | 0 | 0 | 0 | 0 | void Eigen::internal::EigenMetaKernel<Eigen::TensorEvaluator<Eigen::TensorAssignOp<Eigen::TensorMap<Eigen::Tensor<float, 1, 1, long>, 16, Eigen::MakePointer>, Eigen::TensorCwiseBinaryOp<Eigen::internal::scalar_max_op<float const, float const>, Eigen::TensorMap<Eigen::Tensor<float const, 1, 1, long>, 16, Eigen::MakePointer> const, Eigen::TensorCwiseNullaryOp<Eigen::internal::scalar_constant_op<float const>, Eigen::TensorMap<Eigen::Tensor<float const, 1, 1, long>, 16, Eigen::MakePointer> const> const> const> const, Eigen::GpuDevice>, long>(Eigen::TensorEvaluator<Eigen::TensorAssignOp<Eigen::TensorMap<Eigen::Tensor<float, 1, 1, long>, 16, Eigen::MakePointer>, Eigen::TensorCwiseBinaryOp<Eigen::internal::scalar_max_op<float const, float const>, Eigen::TensorMap<Eigen::Tensor<float const, 1, 1, long>, 16, Eigen::MakePointer> const, Eigen::TensorCwiseNullaryOp<Eigen::internal::scalar_constant_op<float const>, Eigen::TensorMap<Eigen::Tensor<float const, 1, 1, long>, 16, Eigen::MakePointer> const> const> const> const, Eigen::GpuDevice>, long) | 3.00 | 0 | 0.00 | 0.00 | 0.00 | 0.00 | 0.00 | true | 0;0;0;0;0 |
| 122 | InceptionV2/InceptionV2/Mixed_4d/Branch_2/Conv2d_0b_3x3/BatchNorm/batchnorm/mul | Conv2D | [[1 160 14 14]] | 201 | 125440 | 2911488 | 45324800 | GPU_0_bfc | 2786048 | 0 | 0 | 0 | maxwell_scudnn_winograd_128x128_ldg1_ldg4_tile148n_nt | 43.00 | 43970560 | 0.00 | 0.00 | 0.00 | 0.00 | 1022.57 | true | 43970560;43970560;43970560;43970560;43970560 |
| 122 | InceptionV2/InceptionV2/Mixed_4d/Branch_2/Conv2d_0b_3x3/BatchNorm/batchnorm/mul | Conv2D | [[1 160 14 14]] | 201 | 125440 | 2911488 | 45324800 | GPU_0_bfc | 2786048 | 0 | 0 | 0 | void tensorflow::functor::ShuffleInTensor3Simple<float, 2, 1, 0, false>(int, float const*, tensorflow::functor::Dimension<3>, float*) | 12.00 | 0 | 0.00 | 0.00 | 0.00 | 0.00 | 0.00 | true | 0;0;0;0;0 |
| 122 | InceptionV2/InceptionV2/Mixed_4d/Branch_2/Conv2d_0b_3x3/BatchNorm/batchnorm/mul | Conv2D | [[1 160 14 14]] | 201 | 125440 | 2911488 | 45324800 | GPU_0_bfc | 2786048 | 0 | 0 | 0 | void cudnn::winograd::generateWinogradTilesKernel<0, float, float>(cudnn::winograd::GenerateWinogradTilesParams<float, float>) | 6.00 | 1187840 | 0.00 | 0.00 | 0.00 | 0.00 | 197.97 | true | 1187840;1187840;1187840;1187840;1187840 |
| 123 | InceptionV2/InceptionV2/Mixed_4d/Branch_1/Conv2d_0b_3x3/BatchNorm/batchnorm/mul | Conv2D | [[1 160 14 14]] | 191 | 125440 | 2911488 | 45349888 | GPU_0_bfc | 2786048 | 0 | 0 | 0 | maxwell_scudnn_winograd_128x128_ldg1_ldg4_tile148n_nt | 43.00 | 43970560 | 0.00 | 0.00 | 0.00 | 0.00 | 1022.57 | true | 43970560;43970560;43970560;43970560;43970560 |
| 123 | InceptionV2/InceptionV2/Mixed_4d/Branch_1/Conv2d_0b_3x3/BatchNorm/batchnorm/mul | Conv2D | [[1 160 14 14]] | 191 | 125440 | 2911488 | 45349888 | GPU_0_bfc | 2786048 | 0 | 0 | 0 | void tensorflow::functor::ShuffleInTensor3Simple<float, 2, 1, 0, false>(int, float const*, tensorflow::functor::Dimension<3>, float*) | 12.00 | 0 | 0.00 | 0.00 | 0.00 | 0.00 | 0.00 | true | 0;0;0;0;0 |
| 123 | InceptionV2/InceptionV2/Mixed_4d/Branch_1/Conv2d_0b_3x3/BatchNorm/batchnorm/mul | Conv2D | [[1 160 14 14]] | 191 | 125440 | 2911488 | 45349888 | GPU_0_bfc | 2786048 | 0 | 0 | 0 | void cudnn::winograd::generateWinogradTilesKernel<0, float, float>(cudnn::winograd::GenerateWinogradTilesParams<float, float>) | 6.00 | 1187840 | 0.00 | 0.00 | 0.00 | 0.00 | 197.97 | true | 1187840;1187840;1187840;1187840;1187840 |
| 124 | InceptionV2/InceptionV2/Mixed_4d/Branch_2/Conv2d_0b_3x3/BatchNorm/batchnorm/add_1 | Add | [[1 160 14 14]] | 29.667 | 125440 | 0 | 45249536 | GPU_0_bfc | 0 | 0 | 0 | 0 | void Eigen::internal::EigenMetaKernel<Eigen::TensorEvaluator<Eigen::TensorAssignOp<Eigen::TensorMap<Eigen::Tensor<float, 2, 1, int>, 16, Eigen::MakePointer>, Eigen::TensorCwiseBinaryOp<Eigen::internal::scalar_sum_op<float, float>, Eigen::TensorMap<Eigen::Tensor<float const, 2, 1, int>, 16, Eigen::MakePointer> const, Eigen::TensorBroadcastingOp<Eigen::array<long, 2ul> const, Eigen::TensorMap<Eigen::Tensor<float const, 2, 1, int>, 16, Eigen::MakePointer> const> const> const> const, Eigen::GpuDevice>, int>(Eigen::TensorEvaluator<Eigen::TensorAssignOp<Eigen::TensorMap<Eigen::Tensor<float, 2, 1, int>, 16, Eigen::MakePointer>, Eigen::TensorCwiseBinaryOp<Eigen::internal::scalar_sum_op<float, float>, Eigen::TensorMap<Eigen::Tensor<float const, 2, 1, int>, 16, Eigen::MakePointer> const, Eigen::TensorBroadcastingOp<Eigen::array<long, 2ul> const, Eigen::TensorMap<Eigen::Tensor<float const, 2, 1, int>, 16, Eigen::MakePointer> const> const> const> const, Eigen::GpuDevice>, int) | 4.00 | 31360 | 0.00 | 0.00 | 0.00 | 0.00 | 7.84 | true | 31360;31360;31360;31360;31360 |
| 125 | InceptionV2/InceptionV2/Mixed_4d/Branch_1/Conv2d_0b_3x3/BatchNorm/batchnorm/add_1 | Add | [[1 160 14 14]] | 23.667 | 125440 | 0 | 45249536 | GPU_0_bfc | 0 | 0 | 0 | 0 | void Eigen::internal::EigenMetaKernel<Eigen::TensorEvaluator<Eigen::TensorAssignOp<Eigen::TensorMap<Eigen::Tensor<float, 2, 1, int>, 16, Eigen::MakePointer>, Eigen::TensorCwiseBinaryOp<Eigen::internal::scalar_sum_op<float, float>, Eigen::TensorMap<Eigen::Tensor<float const, 2, 1, int>, 16, Eigen::MakePointer> const, Eigen::TensorBroadcastingOp<Eigen::array<long, 2ul> const, Eigen::TensorMap<Eigen::Tensor<float const, 2, 1, int>, 16, Eigen::MakePointer> const> const> const> const, Eigen::GpuDevice>, int>(Eigen::TensorEvaluator<Eigen::TensorAssignOp<Eigen::TensorMap<Eigen::Tensor<float, 2, 1, int>, 16, Eigen::MakePointer>, Eigen::TensorCwiseBinaryOp<Eigen::internal::scalar_sum_op<float, float>, Eigen::TensorMap<Eigen::Tensor<float const, 2, 1, int>, 16, Eigen::MakePointer> const, Eigen::TensorBroadcastingOp<Eigen::array<long, 2ul> const, Eigen::TensorMap<Eigen::Tensor<float const, 2, 1, int>, 16, Eigen::MakePointer> const> const> const> const, Eigen::GpuDevice>, int) | 4.00 | 31360 | 0.00 | 0.00 | 0.00 | 0.00 | 7.84 | true | 31360;31360;31360;31360;31360 |
| 126 | InceptionV2/InceptionV2/Mixed_4d/Branch_2/Conv2d_0b_3x3/Relu | Relu | [[1 160 14 14]] | 22.667 | 125440 | 0 | 45249536 | GPU_0_bfc | 0 | 0 | 0 | 0 | void Eigen::internal::EigenMetaKernel<Eigen::TensorEvaluator<Eigen::TensorAssignOp<Eigen::TensorMap<Eigen::Tensor<float, 1, 1, long>, 16, Eigen::MakePointer>, Eigen::TensorCwiseBinaryOp<Eigen::internal::scalar_max_op<float const, float const>, Eigen::TensorMap<Eigen::Tensor<float const, 1, 1, long>, 16, Eigen::MakePointer> const, Eigen::TensorCwiseNullaryOp<Eigen::internal::scalar_constant_op<float const>, Eigen::TensorMap<Eigen::Tensor<float const, 1, 1, long>, 16, Eigen::MakePointer> const> const> const> const, Eigen::GpuDevice>, long>(Eigen::TensorEvaluator<Eigen::TensorAssignOp<Eigen::TensorMap<Eigen::Tensor<float, 1, 1, long>, 16, Eigen::MakePointer>, Eigen::TensorCwiseBinaryOp<Eigen::internal::scalar_max_op<float const, float const>, Eigen::TensorMap<Eigen::Tensor<float const, 1, 1, long>, 16, Eigen::MakePointer> const, Eigen::TensorCwiseNullaryOp<Eigen::internal::scalar_constant_op<float const>, Eigen::TensorMap<Eigen::Tensor<float const, 1, 1, long>, 16, Eigen::MakePointer> const> const> const> const, Eigen::GpuDevice>, long) | 3.00 | 0 | 0.00 | 0.00 | 0.00 | 0.00 | 0.00 | true | 0;0;0;0;0 |
| 127 | InceptionV2/InceptionV2/Mixed_4d/Branch_2/Conv2d_0c_3x3/BatchNorm/batchnorm/mul | Conv2D | [[1 160 14 14]] | 202.333 | 200704 | 3683072 | 45450240 | GPU_0_bfc | 3482368 | 0 | 0 | 0 | maxwell_scudnn_winograd_128x128_ldg1_ldg4_tile148n_nt | 51.00 | 54865920 | 0.00 | 0.00 | 0.00 | 0.00 | 1075.80 | true | 54865920;54865920;54865920;54865920;54865920 |
| 127 | InceptionV2/InceptionV2/Mixed_4d/Branch_2/Conv2d_0c_3x3/BatchNorm/batchnorm/mul | Conv2D | [[1 160 14 14]] | 202.333 | 200704 | 3683072 | 45450240 | GPU_0_bfc | 3482368 | 0 | 0 | 0 | void tensorflow::functor::ShuffleInTensor3Simple<float, 2, 1, 0, false>(int, float const*, tensorflow::functor::Dimension<3>, float*) | 13.00 | 0 | 0.00 | 0.00 | 0.00 | 0.00 | 0.00 | true | 0;0;0;0;0 |
| 127 | InceptionV2/InceptionV2/Mixed_4d/Branch_2/Conv2d_0c_3x3/BatchNorm/batchnorm/mul | Conv2D | [[1 160 14 14]] | 202.333 | 200704 | 3683072 | 45450240 | GPU_0_bfc | 3482368 | 0 | 0 | 0 | void cudnn::winograd::generateWinogradTilesKernel<0, float, float>(cudnn::winograd::GenerateWinogradTilesParams<float, float>) | 7.00 | 1484800 | 0.00 | 0.00 | 0.00 | 0.00 | 212.11 | true | 1484800;1484800;1484800;1484800;1484800 |
| 128 | InceptionV2/InceptionV2/Mixed_4d/Branch_2/Conv2d_0c_3x3/BatchNorm/batchnorm/add_1 | Add | [[1 160 14 14]] | 29.667 | 200704 | 0 | 45324800 | GPU_0_bfc | 0 | 0 | 0 | 0 | void Eigen::internal::EigenMetaKernel<Eigen::TensorEvaluator<Eigen::TensorAssignOp<Eigen::TensorMap<Eigen::Tensor<float, 2, 1, int>, 16, Eigen::MakePointer>, Eigen::TensorCwiseBinaryOp<Eigen::internal::scalar_sum_op<float, float>, Eigen::TensorMap<Eigen::Tensor<float const, 2, 1, int>, 16, Eigen::MakePointer> const, Eigen::TensorBroadcastingOp<Eigen::array<long, 2ul> const, Eigen::TensorMap<Eigen::Tensor<float const, 2, 1, int>, 16, Eigen::MakePointer> const> const> const> const, Eigen::GpuDevice>, int>(Eigen::TensorEvaluator<Eigen::TensorAssignOp<Eigen::TensorMap<Eigen::Tensor<float, 2, 1, int>, 16, Eigen::MakePointer>, Eigen::TensorCwiseBinaryOp<Eigen::internal::scalar_sum_op<float, float>, Eigen::TensorMap<Eigen::Tensor<float const, 2, 1, int>, 16, Eigen::MakePointer> const, Eigen::TensorBroadcastingOp<Eigen::array<long, 2ul> const, Eigen::TensorMap<Eigen::Tensor<float const, 2, 1, int>, 16, Eigen::MakePointer> const> const> const> const, Eigen::GpuDevice>, int) | 4.00 | 31360 | 0.00 | 0.00 | 0.00 | 0.00 | 7.84 | true | 31360;31360;31360;31360;31360 |
| 130 | InceptionV2/InceptionV2/Mixed_4d/Branch_0/Conv2d_0a_1x1/Relu | Relu | [[1 576 14 14]] | 27.333 | 451584 | 0 | 45174272 | GPU_0_bfc | 0 | 0 | 0 | 0 | void Eigen::internal::EigenMetaKernel<Eigen::TensorEvaluator<Eigen::TensorAssignOp<Eigen::TensorMap<Eigen::Tensor<float, 1, 1, long>, 16, Eigen::MakePointer>, Eigen::TensorCwiseBinaryOp<Eigen::internal::scalar_max_op<float const, float const>, Eigen::TensorMap<Eigen::Tensor<float const, 1, 1, long>, 16, Eigen::MakePointer> const, Eigen::TensorCwiseNullaryOp<Eigen::internal::scalar_constant_op<float const>, Eigen::TensorMap<Eigen::Tensor<float const, 1, 1, long>, 16, Eigen::MakePointer> const> const> const> const, Eigen::GpuDevice>, long>(Eigen::TensorEvaluator<Eigen::TensorAssignOp<Eigen::TensorMap<Eigen::Tensor<float, 1, 1, long>, 16, Eigen::MakePointer>, Eigen::TensorCwiseBinaryOp<Eigen::internal::scalar_max_op<float const, float const>, Eigen::TensorMap<Eigen::Tensor<float const, 1, 1, long>, 16, Eigen::MakePointer> const, Eigen::TensorCwiseNullaryOp<Eigen::internal::scalar_constant_op<float const>, Eigen::TensorMap<Eigen::Tensor<float const, 1, 1, long>, 16, Eigen::MakePointer> const> const> const> const, Eigen::GpuDevice>, long) | 3.00 | 0 | 0.00 | 0.00 | 0.00 | 0.00 | 0.00 | true | 0;0;0;0;0 |
| 131 | InceptionV2/InceptionV2/Mixed_4e/Branch_3/AvgPool_0a_3x3/AvgPool | AvgPool | [[1 576 14 14]] | 57 | 727552 | 727552 | 45901824 | GPU_0_bfc | 0 | 0 | 0 | 0 | void cudnn::detail::pooling_fw_4d_kernel<float, float, cudnn::detail::averpooling_func<float>, 2, false>(cudnnTensorStruct, float const*, cudnnTensorStruct, float*, cudnnPoolingStruct, float, float, int, cudnn::reduced_divisor, cudnn::reduced_divisor) | 18.50 | 2417538 | 0.00 | 0.00 | 0.00 | 0.00 | 130.68 | true | 2417538;2417538;2417538;2417538;2417538 |
| 132 | InceptionV2/InceptionV2/Mixed_4e/Branch_2/Conv2d_0a_1x1/BatchNorm/batchnorm/mul | Conv2D | [[1 160 14 14]] | 242.667 | 125440 | 495360 | 46027264 | GPU_0_bfc | 369920 | 0 | 0 | 0 | maxwell_scudnn_128x64_relu_interior_nn | 130.50 | 56721408 | 0.00 | 0.00 | 0.00 | 0.00 | 434.65 | true | 56721408;56721408;56721408;56721408;56721408 |
| 132 | InceptionV2/InceptionV2/Mixed_4e/Branch_2/Conv2d_0a_1x1/BatchNorm/batchnorm/mul | Conv2D | [[1 160 14 14]] | 242.667 | 125440 | 495360 | 46027264 | GPU_0_bfc | 369920 | 0 | 0 | 0 | void tensorflow::functor::ShuffleInTensor3Simple<float, 2, 1, 0, false>(int, float const*, tensorflow::functor::Dimension<3>, float*) | 8.00 | 0 | 0.00 | 0.00 | 0.00 | 0.00 | 0.00 | true | 0;0;0;0;0 |
| 132 | InceptionV2/InceptionV2/Mixed_4e/Branch_2/Conv2d_0a_1x1/BatchNorm/batchnorm/mul | Conv2D | [[1 160 14 14]] | 242.667 | 125440 | 495360 | 46027264 | GPU_0_bfc | 369920 | 0 | 0 | 0 | cudnn::maxwell::gemm::computeOffsetsKernel(cudnn::maxwell::gemm::ComputeOffsetsParams) | 2.00 | 0 | 0.00 | 0.00 | 0.00 | 0.00 | 0.00 | true | 0;0;0;0;0 |
| 133 | InceptionV2/InceptionV2/Mixed_4e/Branch_1/Conv2d_0a_1x1/BatchNorm/batchnorm/mul | Conv2D | [[1 128 14 14]] | 140.333 | 100352 | 395264 | 46127616 | GPU_0_bfc | 294912 | 0 | 0 | 0 | void cudnn::detail::implicit_convolve_sgemm<float, float, 1024, 5, 5, 3, 3, 3, 1, true, false, true>(int, int, int, float const*, int, float*, float*, kernel_conv_params, int, float, float, int, float*, float*, int, int) | 68.33 | 33055232 | 0.00 | 0.00 | 0.00 | 0.00 | 483.74 | true | 33055232;33055232;33055232;33055232;33055232 |
| 133 | InceptionV2/InceptionV2/Mixed_4e/Branch_1/Conv2d_0a_1x1/BatchNorm/batchnorm/mul | Conv2D | [[1 128 14 14]] | 140.333 | 100352 | 395264 | 46127616 | GPU_0_bfc | 294912 | 0 | 0 | 0 | void tensorflow::functor::ShuffleInTensor3Simple<float, 2, 1, 0, false>(int, float const*, tensorflow::functor::Dimension<3>, float*) | 7.00 | 0 | 0.00 | 0.00 | 0.00 | 0.00 | 0.00 | true | 0;0;0;0;0 |
| 134 | InceptionV2/InceptionV2/Mixed_4e/Branch_0/Conv2d_0a_1x1/BatchNorm/batchnorm/mul | Conv2D | [[1 96 14 14]] | 179.667 | 75264 | 297728 | 46202880 | GPU_0_bfc | 222464 | 0 | 0 | 0 | maxwell_scudnn_128x32_relu_interior_nn | 99.00 | 28360704 | 0.00 | 0.00 | 0.00 | 0.00 | 286.47 | true | 28360704;28360704;28360704;28360704;28360704 |
| 134 | InceptionV2/InceptionV2/Mixed_4e/Branch_0/Conv2d_0a_1x1/BatchNorm/batchnorm/mul | Conv2D | [[1 96 14 14]] | 179.667 | 75264 | 297728 | 46202880 | GPU_0_bfc | 222464 | 0 | 0 | 0 | void tensorflow::functor::ShuffleInTensor3Simple<float, 2, 1, 0, false>(int, float const*, tensorflow::functor::Dimension<3>, float*) | 5.33 | 0 | 0.00 | 0.00 | 0.00 | 0.00 | 0.00 | true | 0;0;0;0;0 |
| 134 | InceptionV2/InceptionV2/Mixed_4e/Branch_0/Conv2d_0a_1x1/BatchNorm/batchnorm/mul | Conv2D | [[1 96 14 14]] | 179.667 | 75264 | 297728 | 46202880 | GPU_0_bfc | 222464 | 0 | 0 | 0 | cudnn::maxwell::gemm::computeOffsetsKernel(cudnn::maxwell::gemm::ComputeOffsetsParams) | 2.00 | 0 | 0.00 | 0.00 | 0.00 | 0.00 | 0.00 | true | 0;0;0;0;0 |
| 135 | InceptionV2/InceptionV2/Mixed_4e/Branch_3/Conv2d_0b_1x1/BatchNorm/batchnorm/mul | Conv2D | [[1 96 14 14]] | 172.667 | 75264 | 452864 | 45826560 | GPU_0_bfc | 377600 | 0 | 0 | 0 | maxwell_scudnn_128x32_relu_interior_nn | 99.00 | 28360704 | 0.00 | 0.00 | 0.00 | 0.00 | 286.47 | true | 28360704;28360704;28360704;28360704;28360704 |
| 135 | InceptionV2/InceptionV2/Mixed_4e/Branch_3/Conv2d_0b_1x1/BatchNorm/batchnorm/mul | Conv2D | [[1 96 14 14]] | 172.667 | 75264 | 452864 | 45826560 | GPU_0_bfc | 377600 | 0 | 0 | 0 | void tensorflow::functor::ShuffleInTensor3Simple<float, 2, 1, 0, false>(int, float const*, tensorflow::functor::Dimension<3>, float*) | 5.17 | 0 | 0.00 | 0.00 | 0.00 | 0.00 | 0.00 | true | 0;0;0;0;0 |
| 135 | InceptionV2/InceptionV2/Mixed_4e/Branch_3/Conv2d_0b_1x1/BatchNorm/batchnorm/mul | Conv2D | [[1 96 14 14]] | 172.667 | 75264 | 452864 | 45826560 | GPU_0_bfc | 377600 | 0 | 0 | 0 | cudnn::maxwell::gemm::computeOffsetsKernel(cudnn::maxwell::gemm::ComputeOffsetsParams) | 2.00 | 0 | 0.00 | 0.00 | 0.00 | 0.00 | 0.00 | true | 0;0;0;0;0 |
| 136 | InceptionV2/InceptionV2/Mixed_4e/Branch_2/Conv2d_0a_1x1/BatchNorm/batchnorm/add_1 | Add | [[1 160 14 14]] | 24 | 125440 | 0 | 45099008 | GPU_0_bfc | 0 | 0 | 0 | 0 | void Eigen::internal::EigenMetaKernel<Eigen::TensorEvaluator<Eigen::TensorAssignOp<Eigen::TensorMap<Eigen::Tensor<float, 2, 1, int>, 16, Eigen::MakePointer>, Eigen::TensorCwiseBinaryOp<Eigen::internal::scalar_sum_op<float, float>, Eigen::TensorMap<Eigen::Tensor<float const, 2, 1, int>, 16, Eigen::MakePointer> const, Eigen::TensorBroadcastingOp<Eigen::array<long, 2ul> const, Eigen::TensorMap<Eigen::Tensor<float const, 2, 1, int>, 16, Eigen::MakePointer> const> const> const> const, Eigen::GpuDevice>, int>(Eigen::TensorEvaluator<Eigen::TensorAssignOp<Eigen::TensorMap<Eigen::Tensor<float, 2, 1, int>, 16, Eigen::MakePointer>, Eigen::TensorCwiseBinaryOp<Eigen::internal::scalar_sum_op<float, float>, Eigen::TensorMap<Eigen::Tensor<float const, 2, 1, int>, 16, Eigen::MakePointer> const, Eigen::TensorBroadcastingOp<Eigen::array<long, 2ul> const, Eigen::TensorMap<Eigen::Tensor<float const, 2, 1, int>, 16, Eigen::MakePointer> const> const> const> const, Eigen::GpuDevice>, int) | 4.00 | 31360 | 0.00 | 0.00 | 0.00 | 0.00 | 7.84 | true | 31360;31360;31360;31360;31360 |
| 137 | InceptionV2/InceptionV2/Mixed_4e/Branch_1/Conv2d_0a_1x1/BatchNorm/batchnorm/add_1 | Add | [[1 128 14 14]] | 20 | 100352 | 0 | 45099008 | GPU_0_bfc | 0 | 0 | 0 | 0 | void Eigen::internal::EigenMetaKernel<Eigen::TensorEvaluator<Eigen::TensorAssignOp<Eigen::TensorMap<Eigen::Tensor<float, 2, 1, int>, 16, Eigen::MakePointer>, Eigen::TensorCwiseBinaryOp<Eigen::internal::scalar_sum_op<float, float>, Eigen::TensorMap<Eigen::Tensor<float const, 2, 1, int>, 16, Eigen::MakePointer> const, Eigen::TensorBroadcastingOp<Eigen::array<long, 2ul> const, Eigen::TensorMap<Eigen::Tensor<float const, 2, 1, int>, 16, Eigen::MakePointer> const> const> const> const, Eigen::GpuDevice>, int>(Eigen::TensorEvaluator<Eigen::TensorAssignOp<Eigen::TensorMap<Eigen::Tensor<float, 2, 1, int>, 16, Eigen::MakePointer>, Eigen::TensorCwiseBinaryOp<Eigen::internal::scalar_sum_op<float, float>, Eigen::TensorMap<Eigen::Tensor<float const, 2, 1, int>, 16, Eigen::MakePointer> const, Eigen::TensorBroadcastingOp<Eigen::array<long, 2ul> const, Eigen::TensorMap<Eigen::Tensor<float const, 2, 1, int>, 16, Eigen::MakePointer> const> const> const> const, Eigen::GpuDevice>, int) | 4.00 | 25088 | 0.00 | 0.00 | 0.00 | 0.00 | 6.27 | true | 25088;25088;25088;25088;25088 |
| 138 | InceptionV2/InceptionV2/Mixed_4e/Branch_0/Conv2d_0a_1x1/BatchNorm/batchnorm/add_1 | Add | [[1 96 14 14]] | 19.667 | 75264 | 0 | 45099008 | GPU_0_bfc | 0 | 0 | 0 | 0 | void Eigen::internal::EigenMetaKernel<Eigen::TensorEvaluator<Eigen::TensorAssignOp<Eigen::TensorMap<Eigen::Tensor<float, 2, 1, int>, 16, Eigen::MakePointer>, Eigen::TensorCwiseBinaryOp<Eigen::internal::scalar_sum_op<float, float>, Eigen::TensorMap<Eigen::Tensor<float const, 2, 1, int>, 16, Eigen::MakePointer> const, Eigen::TensorBroadcastingOp<Eigen::array<long, 2ul> const, Eigen::TensorMap<Eigen::Tensor<float const, 2, 1, int>, 16, Eigen::MakePointer> const> const> const> const, Eigen::GpuDevice>, int>(Eigen::TensorEvaluator<Eigen::TensorAssignOp<Eigen::TensorMap<Eigen::Tensor<float, 2, 1, int>, 16, Eigen::MakePointer>, Eigen::TensorCwiseBinaryOp<Eigen::internal::scalar_sum_op<float, float>, Eigen::TensorMap<Eigen::Tensor<float const, 2, 1, int>, 16, Eigen::MakePointer> const, Eigen::TensorBroadcastingOp<Eigen::array<long, 2ul> const, Eigen::TensorMap<Eigen::Tensor<float const, 2, 1, int>, 16, Eigen::MakePointer> const> const> const> const, Eigen::GpuDevice>, int) | 4.00 | 18816 | 0.00 | 0.00 | 0.00 | 0.00 | 4.70 | true | 18816;18816;18816;18816;18816 |
| 139 | InceptionV2/InceptionV2/Mixed_4e/Branch_3/Conv2d_0b_1x1/BatchNorm/batchnorm/add_1 | Add | [[1 96 14 14]] | 18.667 | 75264 | 0 | 45099008 | GPU_0_bfc | 0 | 0 | 0 | 0 | void Eigen::internal::EigenMetaKernel<Eigen::TensorEvaluator<Eigen::TensorAssignOp<Eigen::TensorMap<Eigen::Tensor<float, 2, 1, int>, 16, Eigen::MakePointer>, Eigen::TensorCwiseBinaryOp<Eigen::internal::scalar_sum_op<float, float>, Eigen::TensorMap<Eigen::Tensor<float const, 2, 1, int>, 16, Eigen::MakePointer> const, Eigen::TensorBroadcastingOp<Eigen::array<long, 2ul> const, Eigen::TensorMap<Eigen::Tensor<float const, 2, 1, int>, 16, Eigen::MakePointer> const> const> const> const, Eigen::GpuDevice>, int>(Eigen::TensorEvaluator<Eigen::TensorAssignOp<Eigen::TensorMap<Eigen::Tensor<float, 2, 1, int>, 16, Eigen::MakePointer>, Eigen::TensorCwiseBinaryOp<Eigen::internal::scalar_sum_op<float, float>, Eigen::TensorMap<Eigen::Tensor<float const, 2, 1, int>, 16, Eigen::MakePointer> const, Eigen::TensorBroadcastingOp<Eigen::array<long, 2ul> const, Eigen::TensorMap<Eigen::Tensor<float const, 2, 1, int>, 16, Eigen::MakePointer> const> const> const> const, Eigen::GpuDevice>, int) | 4.00 | 18816 | 0.00 | 0.00 | 0.00 | 0.00 | 4.70 | true | 18816;18816;18816;18816;18816 |
| 140 | InceptionV2/InceptionV2/Mixed_4e/Branch_2/Conv2d_0a_1x1/Relu | Relu | [[1 160 14 14]] | 17.667 | 125440 | 0 | 45099008 | GPU_0_bfc | 0 | 0 | 0 | 0 | void Eigen::internal::EigenMetaKernel<Eigen::TensorEvaluator<Eigen::TensorAssignOp<Eigen::TensorMap<Eigen::Tensor<float, 1, 1, long>, 16, Eigen::MakePointer>, Eigen::TensorCwiseBinaryOp<Eigen::internal::scalar_max_op<float const, float const>, Eigen::TensorMap<Eigen::Tensor<float const, 1, 1, long>, 16, Eigen::MakePointer> const, Eigen::TensorCwiseNullaryOp<Eigen::internal::scalar_constant_op<float const>, Eigen::TensorMap<Eigen::Tensor<float const, 1, 1, long>, 16, Eigen::MakePointer> const> const> const> const, Eigen::GpuDevice>, long>(Eigen::TensorEvaluator<Eigen::TensorAssignOp<Eigen::TensorMap<Eigen::Tensor<float, 1, 1, long>, 16, Eigen::MakePointer>, Eigen::TensorCwiseBinaryOp<Eigen::internal::scalar_max_op<float const, float const>, Eigen::TensorMap<Eigen::Tensor<float const, 1, 1, long>, 16, Eigen::MakePointer> const, Eigen::TensorCwiseNullaryOp<Eigen::internal::scalar_constant_op<float const>, Eigen::TensorMap<Eigen::Tensor<float const, 1, 1, long>, 16, Eigen::MakePointer> const> const> const> const, Eigen::GpuDevice>, long) | 3.00 | 0 | 0.00 | 0.00 | 0.00 | 0.00 | 0.00 | true | 0;0;0;0;0 |
| 141 | InceptionV2/InceptionV2/Mixed_4e/Branch_1/Conv2d_0a_1x1/Relu | Relu | [[1 128 14 14]] | 16.667 | 100352 | 0 | 45099008 | GPU_0_bfc | 0 | 0 | 0 | 0 | void Eigen::internal::EigenMetaKernel<Eigen::TensorEvaluator<Eigen::TensorAssignOp<Eigen::TensorMap<Eigen::Tensor<float, 1, 1, long>, 16, Eigen::MakePointer>, Eigen::TensorCwiseBinaryOp<Eigen::internal::scalar_max_op<float const, float const>, Eigen::TensorMap<Eigen::Tensor<float const, 1, 1, long>, 16, Eigen::MakePointer> const, Eigen::TensorCwiseNullaryOp<Eigen::internal::scalar_constant_op<float const>, Eigen::TensorMap<Eigen::Tensor<float const, 1, 1, long>, 16, Eigen::MakePointer> const> const> const> const, Eigen::GpuDevice>, long>(Eigen::TensorEvaluator<Eigen::TensorAssignOp<Eigen::TensorMap<Eigen::Tensor<float, 1, 1, long>, 16, Eigen::MakePointer>, Eigen::TensorCwiseBinaryOp<Eigen::internal::scalar_max_op<float const, float const>, Eigen::TensorMap<Eigen::Tensor<float const, 1, 1, long>, 16, Eigen::MakePointer> const, Eigen::TensorCwiseNullaryOp<Eigen::internal::scalar_constant_op<float const>, Eigen::TensorMap<Eigen::Tensor<float const, 1, 1, long>, 16, Eigen::MakePointer> const> const> const> const, Eigen::GpuDevice>, long) | 3.00 | 0 | 0.00 | 0.00 | 0.00 | 0.00 | 0.00 | true | 0;0;0;0;0 |
| 142 | InceptionV2/InceptionV2/Mixed_4e/Branch_2/Conv2d_0b_3x3/BatchNorm/batchnorm/mul | Conv2D | [[1 192 14 14]] | 173.667 | 150528 | 4329216 | 45249536 | GPU_0_bfc | 4178688 | 0 | 0 | 0 | maxwell_scudnn_winograd_128x128_ldg1_ldg4_tile148n_nt | 52.00 | 65839104 | 0.00 | 0.00 | 0.00 | 0.00 | 1266.14 | true | 65839104;65839104;65839104;65839104;65839104 |
| 142 | InceptionV2/InceptionV2/Mixed_4e/Branch_2/Conv2d_0b_3x3/BatchNorm/batchnorm/mul | Conv2D | [[1 192 14 14]] | 173.667 | 150528 | 4329216 | 45249536 | GPU_0_bfc | 4178688 | 0 | 0 | 0 | void tensorflow::functor::ShuffleInTensor3Simple<float, 2, 1, 0, false>(int, float const*, tensorflow::functor::Dimension<3>, float*) | 15.00 | 0 | 0.00 | 0.00 | 0.00 | 0.00 | 0.00 | true | 0;0;0;0;0 |
| 142 | InceptionV2/InceptionV2/Mixed_4e/Branch_2/Conv2d_0b_3x3/BatchNorm/batchnorm/mul | Conv2D | [[1 192 14 14]] | 173.667 | 150528 | 4329216 | 45249536 | GPU_0_bfc | 4178688 | 0 | 0 | 0 | void cudnn::winograd::generateWinogradTilesKernel<0, float, float>(cudnn::winograd::GenerateWinogradTilesParams<float, float>) | 8.00 | 1781760 | 0.00 | 0.00 | 0.00 | 0.00 | 222.72 | true | 1781760;1781760;1781760;1781760;1781760 |
| 143 | InceptionV2/InceptionV2/Mixed_4e/Branch_1/Conv2d_0b_3x3/BatchNorm/batchnorm/mul | Conv2D | [[1 192 14 14]] | 155.667 | 150528 | 3493632 | 45274624 | GPU_0_bfc | 3343104 | 0 | 0 | 0 | maxwell_scudnn_winograd_128x128_ldg1_ldg4_tile148n_nt | 43.33 | 52764672 | 0.00 | 0.00 | 0.00 | 0.00 | 1217.66 | true | 52764672;52764672;52764672;52764672;52764672 |
| 143 | InceptionV2/InceptionV2/Mixed_4e/Branch_1/Conv2d_0b_3x3/BatchNorm/batchnorm/mul | Conv2D | [[1 192 14 14]] | 155.667 | 150528 | 3493632 | 45274624 | GPU_0_bfc | 3343104 | 0 | 0 | 0 | void tensorflow::functor::ShuffleInTensor3Simple<float, 2, 1, 0, false>(int, float const*, tensorflow::functor::Dimension<3>, float*) | 13.00 | 0 | 0.00 | 0.00 | 0.00 | 0.00 | 0.00 | true | 0;0;0;0;0 |
| 143 | InceptionV2/InceptionV2/Mixed_4e/Branch_1/Conv2d_0b_3x3/BatchNorm/batchnorm/mul | Conv2D | [[1 192 14 14]] | 155.667 | 150528 | 3493632 | 45274624 | GPU_0_bfc | 3343104 | 0 | 0 | 0 | void cudnn::winograd::generateWinogradTilesKernel<0, float, float>(cudnn::winograd::GenerateWinogradTilesParams<float, float>) | 7.00 | 1425408 | 0.00 | 0.00 | 0.00 | 0.00 | 203.63 | true | 1425408;1425408;1425408;1425408;1425408 |
| 144 | InceptionV2/InceptionV2/Mixed_4e/Branch_2/Conv2d_0b_3x3/BatchNorm/batchnorm/add_1 | Add | [[1 192 14 14]] | 26.333 | 150528 | 0 | 45174272 | GPU_0_bfc | 0 | 0 | 0 | 0 | void Eigen::internal::EigenMetaKernel<Eigen::TensorEvaluator<Eigen::TensorAssignOp<Eigen::TensorMap<Eigen::Tensor<float, 2, 1, int>, 16, Eigen::MakePointer>, Eigen::TensorCwiseBinaryOp<Eigen::internal::scalar_sum_op<float, float>, Eigen::TensorMap<Eigen::Tensor<float const, 2, 1, int>, 16, Eigen::MakePointer> const, Eigen::TensorBroadcastingOp<Eigen::array<long, 2ul> const, Eigen::TensorMap<Eigen::Tensor<float const, 2, 1, int>, 16, Eigen::MakePointer> const> const> const> const, Eigen::GpuDevice>, int>(Eigen::TensorEvaluator<Eigen::TensorAssignOp<Eigen::TensorMap<Eigen::Tensor<float, 2, 1, int>, 16, Eigen::MakePointer>, Eigen::TensorCwiseBinaryOp<Eigen::internal::scalar_sum_op<float, float>, Eigen::TensorMap<Eigen::Tensor<float const, 2, 1, int>, 16, Eigen::MakePointer> const, Eigen::TensorBroadcastingOp<Eigen::array<long, 2ul> const, Eigen::TensorMap<Eigen::Tensor<float const, 2, 1, int>, 16, Eigen::MakePointer> const> const> const> const, Eigen::GpuDevice>, int) | 4.00 | 37632 | 0.00 | 0.00 | 0.00 | 0.00 | 9.41 | true | 37632;37632;37632;37632;37632 |
| 145 | InceptionV2/InceptionV2/Mixed_4e/Branch_1/Conv2d_0b_3x3/BatchNorm/batchnorm/add_1 | Add | [[1 192 14 14]] | 19 | 150528 | 0 | 45174272 | GPU_0_bfc | 0 | 0 | 0 | 0 | void Eigen::internal::EigenMetaKernel<Eigen::TensorEvaluator<Eigen::TensorAssignOp<Eigen::TensorMap<Eigen::Tensor<float, 2, 1, int>, 16, Eigen::MakePointer>, Eigen::TensorCwiseBinaryOp<Eigen::internal::scalar_sum_op<float, float>, Eigen::TensorMap<Eigen::Tensor<float const, 2, 1, int>, 16, Eigen::MakePointer> const, Eigen::TensorBroadcastingOp<Eigen::array<long, 2ul> const, Eigen::TensorMap<Eigen::Tensor<float const, 2, 1, int>, 16, Eigen::MakePointer> const> const> const> const, Eigen::GpuDevice>, int>(Eigen::TensorEvaluator<Eigen::TensorAssignOp<Eigen::TensorMap<Eigen::Tensor<float, 2, 1, int>, 16, Eigen::MakePointer>, Eigen::TensorCwiseBinaryOp<Eigen::internal::scalar_sum_op<float, float>, Eigen::TensorMap<Eigen::Tensor<float const, 2, 1, int>, 16, Eigen::MakePointer> const, Eigen::TensorBroadcastingOp<Eigen::array<long, 2ul> const, Eigen::TensorMap<Eigen::Tensor<float const, 2, 1, int>, 16, Eigen::MakePointer> const> const> const> const, Eigen::GpuDevice>, int) | 4.00 | 37632 | 0.00 | 0.00 | 0.00 | 0.00 | 9.41 | true | 37632;37632;37632;37632;37632 |
| 146 | InceptionV2/InceptionV2/Mixed_4e/Branch_2/Conv2d_0b_3x3/Relu | Relu | [[1 192 14 14]] | 17.333 | 150528 | 0 | 45174272 | GPU_0_bfc | 0 | 0 | 0 | 0 | void Eigen::internal::EigenMetaKernel<Eigen::TensorEvaluator<Eigen::TensorAssignOp<Eigen::TensorMap<Eigen::Tensor<float, 1, 1, long>, 16, Eigen::MakePointer>, Eigen::TensorCwiseBinaryOp<Eigen::internal::scalar_max_op<float const, float const>, Eigen::TensorMap<Eigen::Tensor<float const, 1, 1, long>, 16, Eigen::MakePointer> const, Eigen::TensorCwiseNullaryOp<Eigen::internal::scalar_constant_op<float const>, Eigen::TensorMap<Eigen::Tensor<float const, 1, 1, long>, 16, Eigen::MakePointer> const> const> const> const, Eigen::GpuDevice>, long>(Eigen::TensorEvaluator<Eigen::TensorAssignOp<Eigen::TensorMap<Eigen::Tensor<float, 1, 1, long>, 16, Eigen::MakePointer>, Eigen::TensorCwiseBinaryOp<Eigen::internal::scalar_max_op<float const, float const>, Eigen::TensorMap<Eigen::Tensor<float const, 1, 1, long>, 16, Eigen::MakePointer> const, Eigen::TensorCwiseNullaryOp<Eigen::internal::scalar_constant_op<float const>, Eigen::TensorMap<Eigen::Tensor<float const, 1, 1, long>, 16, Eigen::MakePointer> const> const> const> const, Eigen::GpuDevice>, long) | 3.00 | 0 | 0.00 | 0.00 | 0.00 | 0.00 | 0.00 | true | 0;0;0;0;0 |
| 147 | InceptionV2/InceptionV2/Mixed_4e/Branch_2/Conv2d_0c_3x3/BatchNorm/batchnorm/mul | Conv2D | [[1 192 14 14]] | 181.333 | 150528 | 5164800 | 45324800 | GPU_0_bfc | 5014272 | 0 | 0 | 0 | maxwell_scudnn_winograd_128x128_ldg1_ldg4_tile148n_nt | 60.50 | 78913536 | 0.00 | 0.00 | 0.00 | 0.00 | 1304.36 | true | 78913536;78913536;78913536;78913536;78913536 |
| 147 | InceptionV2/InceptionV2/Mixed_4e/Branch_2/Conv2d_0c_3x3/BatchNorm/batchnorm/mul | Conv2D | [[1 192 14 14]] | 181.333 | 150528 | 5164800 | 45324800 | GPU_0_bfc | 5014272 | 0 | 0 | 0 | void tensorflow::functor::ShuffleInTensor3Simple<float, 2, 1, 0, false>(int, float const*, tensorflow::functor::Dimension<3>, float*) | 17.00 | 0 | 0.00 | 0.00 | 0.00 | 0.00 | 0.00 | true | 0;0;0;0;0 |
| 147 | InceptionV2/InceptionV2/Mixed_4e/Branch_2/Conv2d_0c_3x3/BatchNorm/batchnorm/mul | Conv2D | [[1 192 14 14]] | 181.333 | 150528 | 5164800 | 45324800 | GPU_0_bfc | 5014272 | 0 | 0 | 0 | void cudnn::winograd::generateWinogradTilesKernel<0, float, float>(cudnn::winograd::GenerateWinogradTilesParams<float, float>) | 9.00 | 2138112 | 0.00 | 0.00 | 0.00 | 0.00 | 237.57 | true | 2138112;2138112;2138112;2138112;2138112 |
| 148 | InceptionV2/InceptionV2/Mixed_4e/Branch_2/Conv2d_0c_3x3/BatchNorm/batchnorm/add_1 | Add | [[1 192 14 14]] | 23.667 | 150528 | 0 | 45174272 | GPU_0_bfc | 0 | 0 | 0 | 0 | void Eigen::internal::EigenMetaKernel<Eigen::TensorEvaluator<Eigen::TensorAssignOp<Eigen::TensorMap<Eigen::Tensor<float, 2, 1, int>, 16, Eigen::MakePointer>, Eigen::TensorCwiseBinaryOp<Eigen::internal::scalar_sum_op<float, float>, Eigen::TensorMap<Eigen::Tensor<float const, 2, 1, int>, 16, Eigen::MakePointer> const, Eigen::TensorBroadcastingOp<Eigen::array<long, 2ul> const, Eigen::TensorMap<Eigen::Tensor<float const, 2, 1, int>, 16, Eigen::MakePointer> const> const> const> const, Eigen::GpuDevice>, int>(Eigen::TensorEvaluator<Eigen::TensorAssignOp<Eigen::TensorMap<Eigen::Tensor<float, 2, 1, int>, 16, Eigen::MakePointer>, Eigen::TensorCwiseBinaryOp<Eigen::internal::scalar_sum_op<float, float>, Eigen::TensorMap<Eigen::Tensor<float const, 2, 1, int>, 16, Eigen::MakePointer> const, Eigen::TensorBroadcastingOp<Eigen::array<long, 2ul> const, Eigen::TensorMap<Eigen::Tensor<float const, 2, 1, int>, 16, Eigen::MakePointer> const> const> const> const, Eigen::GpuDevice>, int) | 4.00 | 37632 | 0.00 | 0.00 | 0.00 | 0.00 | 9.41 | true | 37632;37632;37632;37632;37632 |
| 150 | InceptionV2/InceptionV2/Mixed_4e/Branch_0/Conv2d_0a_1x1/Relu | Relu | [[1 576 14 14]] | 20.667 | 727552 | 0 | 45450240 | GPU_0_bfc | 0 | 0 | 0 | 0 | void Eigen::internal::EigenMetaKernel<Eigen::TensorEvaluator<Eigen::TensorAssignOp<Eigen::TensorMap<Eigen::Tensor<float, 1, 1, long>, 16, Eigen::MakePointer>, Eigen::TensorCwiseBinaryOp<Eigen::internal::scalar_max_op<float const, float const>, Eigen::TensorMap<Eigen::Tensor<float const, 1, 1, long>, 16, Eigen::MakePointer> const, Eigen::TensorCwiseNullaryOp<Eigen::internal::scalar_constant_op<float const>, Eigen::TensorMap<Eigen::Tensor<float const, 1, 1, long>, 16, Eigen::MakePointer> const> const> const> const, Eigen::GpuDevice>, long>(Eigen::TensorEvaluator<Eigen::TensorAssignOp<Eigen::TensorMap<Eigen::Tensor<float, 1, 1, long>, 16, Eigen::MakePointer>, Eigen::TensorCwiseBinaryOp<Eigen::internal::scalar_max_op<float const, float const>, Eigen::TensorMap<Eigen::Tensor<float const, 1, 1, long>, 16, Eigen::MakePointer> const, Eigen::TensorCwiseNullaryOp<Eigen::internal::scalar_constant_op<float const>, Eigen::TensorMap<Eigen::Tensor<float const, 1, 1, long>, 16, Eigen::MakePointer> const> const> const> const, Eigen::GpuDevice>, long) | 3.00 | 0 | 0.00 | 0.00 | 0.00 | 0.00 | 0.00 | true | 0;0;0;0;0 |
| 151 | InceptionV2/InceptionV2/Mixed_5a/Branch_2/MaxPool_1a_3x3/MaxPool | MaxPool | [[1 576 7 7]] | 37 | 112896 | 112896 | 45563136 | GPU_0_bfc | 0 | 0 | 0 | 0 | void cudnn::detail::pooling_fw_4d_kernel<float, float, cudnn::detail::maxpooling_func<float, (cudnnNanPropagation_t)0>, 0, false>(cudnnTensorStruct, float const*, cudnnTensorStruct, float*, cudnnPoolingStruct, float, float, int, cudnn::reduced_divisor, cudnn::reduced_divisor) | 6.67 | 28224 | 0.00 | 0.00 | 0.00 | 0.00 | 4.23 | true | 28224;28224;28224;28224;28224 |
| 152 | InceptionV2/InceptionV2/Mixed_5a/Branch_1/Conv2d_0a_1x1/BatchNorm/batchnorm/mul | Conv2D | [[1 192 14 14]] | 214 | 150528 | 594176 | 45713664 | GPU_0_bfc | 443648 | 0 | 0 | 0 | maxwell_scudnn_128x64_relu_interior_nn | 130.83 | 56721408 | 0.00 | 0.00 | 0.00 | 0.00 | 433.54 | true | 56721408;56721408;56721408;56721408;56721408 |
| 152 | InceptionV2/InceptionV2/Mixed_5a/Branch_1/Conv2d_0a_1x1/BatchNorm/batchnorm/mul | Conv2D | [[1 192 14 14]] | 214 | 150528 | 594176 | 45713664 | GPU_0_bfc | 443648 | 0 | 0 | 0 | void tensorflow::functor::ShuffleInTensor3Simple<float, 2, 1, 0, false>(int, float const*, tensorflow::functor::Dimension<3>, float*) | 7.67 | 0 | 0.00 | 0.00 | 0.00 | 0.00 | 0.00 | true | 0;0;0;0;0 |
| 152 | InceptionV2/InceptionV2/Mixed_5a/Branch_1/Conv2d_0a_1x1/BatchNorm/batchnorm/mul | Conv2D | [[1 192 14 14]] | 214 | 150528 | 594176 | 45713664 | GPU_0_bfc | 443648 | 0 | 0 | 0 | cudnn::maxwell::gemm::computeOffsetsKernel(cudnn::maxwell::gemm::ComputeOffsetsParams) | 2.00 | 0 | 0.00 | 0.00 | 0.00 | 0.00 | 0.00 | true | 0;0;0;0;0 |
| 153 | InceptionV2/InceptionV2/Mixed_5a/Branch_0/Conv2d_0a_1x1/BatchNorm/batchnorm/mul | Conv2D | [[1 128 14 14]] | 141 | 100352 | 395264 | 45814016 | GPU_0_bfc | 294912 | 0 | 0 | 0 | void cudnn::detail::implicit_convolve_sgemm<float, float, 1024, 5, 5, 3, 3, 3, 1, true, false, true>(int, int, int, float const*, int, float*, float*, kernel_conv_params, int, float, float, int, float*, float*, int, int) | 68.33 | 33055232 | 0.00 | 0.00 | 0.00 | 0.00 | 483.74 | true | 33055232;33055232;33055232;33055232;33055232 |
| 153 | InceptionV2/InceptionV2/Mixed_5a/Branch_0/Conv2d_0a_1x1/BatchNorm/batchnorm/mul | Conv2D | [[1 128 14 14]] | 141 | 100352 | 395264 | 45814016 | GPU_0_bfc | 294912 | 0 | 0 | 0 | void tensorflow::functor::ShuffleInTensor3Simple<float, 2, 1, 0, false>(int, float const*, tensorflow::functor::Dimension<3>, float*) | 7.00 | 0 | 0.00 | 0.00 | 0.00 | 0.00 | 0.00 | true | 0;0;0;0;0 |
| 154 | InceptionV2/InceptionV2/Mixed_5a/Branch_1/Conv2d_0a_1x1/BatchNorm/batchnorm/add_1 | Add | [[1 192 14 14]] | 25 | 150528 | 0 | 45086464 | GPU_0_bfc | 0 | 0 | 0 | 0 | void Eigen::internal::EigenMetaKernel<Eigen::TensorEvaluator<Eigen::TensorAssignOp<Eigen::TensorMap<Eigen::Tensor<float, 2, 1, int>, 16, Eigen::MakePointer>, Eigen::TensorCwiseBinaryOp<Eigen::internal::scalar_sum_op<float, float>, Eigen::TensorMap<Eigen::Tensor<float const, 2, 1, int>, 16, Eigen::MakePointer> const, Eigen::TensorBroadcastingOp<Eigen::array<long, 2ul> const, Eigen::TensorMap<Eigen::Tensor<float const, 2, 1, int>, 16, Eigen::MakePointer> const> const> const> const, Eigen::GpuDevice>, int>(Eigen::TensorEvaluator<Eigen::TensorAssignOp<Eigen::TensorMap<Eigen::Tensor<float, 2, 1, int>, 16, Eigen::MakePointer>, Eigen::TensorCwiseBinaryOp<Eigen::internal::scalar_sum_op<float, float>, Eigen::TensorMap<Eigen::Tensor<float const, 2, 1, int>, 16, Eigen::MakePointer> const, Eigen::TensorBroadcastingOp<Eigen::array<long, 2ul> const, Eigen::TensorMap<Eigen::Tensor<float const, 2, 1, int>, 16, Eigen::MakePointer> const> const> const> const, Eigen::GpuDevice>, int) | 4.00 | 37632 | 0.00 | 0.00 | 0.00 | 0.00 | 9.41 | true | 37632;37632;37632;37632;37632 |
| 155 | InceptionV2/InceptionV2/Mixed_5a/Branch_0/Conv2d_0a_1x1/BatchNorm/batchnorm/add_1 | Add | [[1 128 14 14]] | 19 | 100352 | 0 | 45086464 | GPU_0_bfc | 0 | 0 | 0 | 0 | void Eigen::internal::EigenMetaKernel<Eigen::TensorEvaluator<Eigen::TensorAssignOp<Eigen::TensorMap<Eigen::Tensor<float, 2, 1, int>, 16, Eigen::MakePointer>, Eigen::TensorCwiseBinaryOp<Eigen::internal::scalar_sum_op<float, float>, Eigen::TensorMap<Eigen::Tensor<float const, 2, 1, int>, 16, Eigen::MakePointer> const, Eigen::TensorBroadcastingOp<Eigen::array<long, 2ul> const, Eigen::TensorMap<Eigen::Tensor<float const, 2, 1, int>, 16, Eigen::MakePointer> const> const> const> const, Eigen::GpuDevice>, int>(Eigen::TensorEvaluator<Eigen::TensorAssignOp<Eigen::TensorMap<Eigen::Tensor<float, 2, 1, int>, 16, Eigen::MakePointer>, Eigen::TensorCwiseBinaryOp<Eigen::internal::scalar_sum_op<float, float>, Eigen::TensorMap<Eigen::Tensor<float const, 2, 1, int>, 16, Eigen::MakePointer> const, Eigen::TensorBroadcastingOp<Eigen::array<long, 2ul> const, Eigen::TensorMap<Eigen::Tensor<float const, 2, 1, int>, 16, Eigen::MakePointer> const> const> const> const, Eigen::GpuDevice>, int) | 4.00 | 25088 | 0.00 | 0.00 | 0.00 | 0.00 | 6.27 | true | 25088;25088;25088;25088;25088 |
| 156 | InceptionV2/InceptionV2/Mixed_5a/Branch_1/Conv2d_0a_1x1/Relu | Relu | [[1 192 14 14]] | 20.667 | 150528 | 0 | 45086464 | GPU_0_bfc | 0 | 0 | 0 | 0 | void Eigen::internal::EigenMetaKernel<Eigen::TensorEvaluator<Eigen::TensorAssignOp<Eigen::TensorMap<Eigen::Tensor<float, 1, 1, long>, 16, Eigen::MakePointer>, Eigen::TensorCwiseBinaryOp<Eigen::internal::scalar_max_op<float const, float const>, Eigen::TensorMap<Eigen::Tensor<float const, 1, 1, long>, 16, Eigen::MakePointer> const, Eigen::TensorCwiseNullaryOp<Eigen::internal::scalar_constant_op<float const>, Eigen::TensorMap<Eigen::Tensor<float const, 1, 1, long>, 16, Eigen::MakePointer> const> const> const> const, Eigen::GpuDevice>, long>(Eigen::TensorEvaluator<Eigen::TensorAssignOp<Eigen::TensorMap<Eigen::Tensor<float, 1, 1, long>, 16, Eigen::MakePointer>, Eigen::TensorCwiseBinaryOp<Eigen::internal::scalar_max_op<float const, float const>, Eigen::TensorMap<Eigen::Tensor<float const, 1, 1, long>, 16, Eigen::MakePointer> const, Eigen::TensorCwiseNullaryOp<Eigen::internal::scalar_constant_op<float const>, Eigen::TensorMap<Eigen::Tensor<float const, 1, 1, long>, 16, Eigen::MakePointer> const> const> const> const, Eigen::GpuDevice>, long) | 3.00 | 0 | 0.00 | 0.00 | 0.00 | 0.00 | 0.00 | true | 0;0;0;0;0 |
| 157 | InceptionV2/InceptionV2/Mixed_5a/Branch_0/Conv2d_0a_1x1/Relu | Relu | [[1 128 14 14]] | 16.667 | 100352 | 0 | 45086464 | GPU_0_bfc | 0 | 0 | 0 | 0 | void Eigen::internal::EigenMetaKernel<Eigen::TensorEvaluator<Eigen::TensorAssignOp<Eigen::TensorMap<Eigen::Tensor<float, 1, 1, long>, 16, Eigen::MakePointer>, Eigen::TensorCwiseBinaryOp<Eigen::internal::scalar_max_op<float const, float const>, Eigen::TensorMap<Eigen::Tensor<float const, 1, 1, long>, 16, Eigen::MakePointer> const, Eigen::TensorCwiseNullaryOp<Eigen::internal::scalar_constant_op<float const>, Eigen::TensorMap<Eigen::Tensor<float const, 1, 1, long>, 16, Eigen::MakePointer> const> const> const> const, Eigen::GpuDevice>, long>(Eigen::TensorEvaluator<Eigen::TensorAssignOp<Eigen::TensorMap<Eigen::Tensor<float, 1, 1, long>, 16, Eigen::MakePointer>, Eigen::TensorCwiseBinaryOp<Eigen::internal::scalar_max_op<float const, float const>, Eigen::TensorMap<Eigen::Tensor<float const, 1, 1, long>, 16, Eigen::MakePointer> const, Eigen::TensorCwiseNullaryOp<Eigen::internal::scalar_constant_op<float const>, Eigen::TensorMap<Eigen::Tensor<float const, 1, 1, long>, 16, Eigen::MakePointer> const> const> const> const, Eigen::GpuDevice>, long) | 3.00 | 0 | 0.00 | 0.00 | 0.00 | 0.00 | 0.00 | true | 0;0;0;0;0 |
| 158 | InceptionV2/InceptionV2/Mixed_5a/Branch_1/Conv2d_0b_3x3/BatchNorm/batchnorm/mul | Conv2D | [[1 256 14 14]] | 193.333 | 200704 | 6886400 | 45287168 | GPU_0_bfc | 6685696 | 0 | 0 | 0 | maxwell_scudnn_winograd_128x128_ldg1_ldg4_tile148n_nt | 60.83 | 105218048 | 0.00 | 0.00 | 0.00 | 0.00 | 1729.62 | true | 105218048;105218048;105218048;105218048;105218048 |
| 158 | InceptionV2/InceptionV2/Mixed_5a/Branch_1/Conv2d_0b_3x3/BatchNorm/batchnorm/mul | Conv2D | [[1 256 14 14]] | 193.333 | 200704 | 6886400 | 45287168 | GPU_0_bfc | 6685696 | 0 | 0 | 0 | void tensorflow::functor::ShuffleInTensor3Simple<float, 2, 1, 0, false>(int, float const*, tensorflow::functor::Dimension<3>, float*) | 22.00 | 0 | 0.00 | 0.00 | 0.00 | 0.00 | 0.00 | true | 0;0;0;0;0 |
| 158 | InceptionV2/InceptionV2/Mixed_5a/Branch_1/Conv2d_0b_3x3/BatchNorm/batchnorm/mul | Conv2D | [[1 256 14 14]] | 193.333 | 200704 | 6886400 | 45287168 | GPU_0_bfc | 6685696 | 0 | 0 | 0 | void cudnn::winograd::generateWinogradTilesKernel<0, float, float>(cudnn::winograd::GenerateWinogradTilesParams<float, float>) | 11.00 | 2850816 | 0.00 | 0.00 | 0.00 | 0.00 | 259.17 | true | 2850816;2850816;2850816;2850816;2850816 |
| 159 | InceptionV2/InceptionV2/Mixed_5a/Branch_0/Conv2d_1a_3x3/BatchNorm/batchnorm/mul | Conv2D | [[1 192 7 7]] | 243.667 | 37632 | 1449216 | 45174272 | GPU_0_bfc | 1411584 | 0 | 0 | 0 | void cudnn::detail::explicit_convolve_sgemm<float, int, 128, 5, 5, 3, 3, 3, 0, true>(int, int, int, float const*, int, float const*, int, float*, kernel_conv_params, int, int, float, float, int, float*, float*) | 70.83 | 28320960 | 0.00 | 0.00 | 0.00 | 0.00 | 399.83 | true | 28320960;28320960;28320960;28320960;28320960 |
| 159 | InceptionV2/InceptionV2/Mixed_5a/Branch_0/Conv2d_1a_3x3/BatchNorm/batchnorm/mul | Conv2D | [[1 192 7 7]] | 243.667 | 37632 | 1449216 | 45174272 | GPU_0_bfc | 1411584 | 0 | 0 | 0 | void im2col4d_kernel<float, int>(im2col4d_params, cudnnConvolutionStruct, cudnnTensor4dStruct, float const*, float*, int) | 61.50 | 0 | 0.00 | 0.00 | 0.00 | 0.00 | 0.00 | true | 0;0;0;0;0 |
| 159 | InceptionV2/InceptionV2/Mixed_5a/Branch_0/Conv2d_1a_3x3/BatchNorm/batchnorm/mul | Conv2D | [[1 192 7 7]] | 243.667 | 37632 | 1449216 | 45174272 | GPU_0_bfc | 1411584 | 0 | 0 | 0 | void tensorflow::functor::ShuffleInTensor3Simple<float, 2, 1, 0, false>(int, float const*, tensorflow::functor::Dimension<3>, float*) | 13.00 | 0 | 0.00 | 0.00 | 0.00 | 0.00 | 0.00 | true | 0;0;0;0;0 |
| 159 | InceptionV2/InceptionV2/Mixed_5a/Branch_0/Conv2d_1a_3x3/BatchNorm/batchnorm/mul | Conv2D | [[1 192 7 7]] | 243.667 | 37632 | 1449216 | 45174272 | GPU_0_bfc | 1411584 | 0 | 0 | 0 | void tensorflow::functor::PadInputCustomKernelNCHW<float, 4>(int, float const*, tensorflow::functor::Dimension<4>, float*, tensorflow::functor::Dimension<4>, tensorflow::functor::Dimension<(4)-(2)>) | 5.17 | 0 | 0.00 | 0.00 | 0.00 | 0.00 | 0.00 | true | 0;0;0;0;0 |
| 160 | InceptionV2/InceptionV2/Mixed_5a/Branch_1/Conv2d_0b_3x3/BatchNorm/batchnorm/add_1 | Add | [[1 256 14 14]] | 25 | 200704 | 0 | 45073920 | GPU_0_bfc | 0 | 0 | 0 | 0 | void Eigen::internal::EigenMetaKernel<Eigen::TensorEvaluator<Eigen::TensorAssignOp<Eigen::TensorMap<Eigen::Tensor<float, 2, 1, int>, 16, Eigen::MakePointer>, Eigen::TensorCwiseBinaryOp<Eigen::internal::scalar_sum_op<float, float>, Eigen::TensorMap<Eigen::Tensor<float const, 2, 1, int>, 16, Eigen::MakePointer> const, Eigen::TensorBroadcastingOp<Eigen::array<long, 2ul> const, Eigen::TensorMap<Eigen::Tensor<float const, 2, 1, int>, 16, Eigen::MakePointer> const> const> const> const, Eigen::GpuDevice>, int>(Eigen::TensorEvaluator<Eigen::TensorAssignOp<Eigen::TensorMap<Eigen::Tensor<float, 2, 1, int>, 16, Eigen::MakePointer>, Eigen::TensorCwiseBinaryOp<Eigen::internal::scalar_sum_op<float, float>, Eigen::TensorMap<Eigen::Tensor<float const, 2, 1, int>, 16, Eigen::MakePointer> const, Eigen::TensorBroadcastingOp<Eigen::array<long, 2ul> const, Eigen::TensorMap<Eigen::Tensor<float const, 2, 1, int>, 16, Eigen::MakePointer> const> const> const> const, Eigen::GpuDevice>, int) | 4.00 | 50176 | 0.00 | 0.00 | 0.00 | 0.00 | 12.54 | true | 50176;50176;50176;50176;50176 |
| 161 | InceptionV2/InceptionV2/Mixed_5a/Branch_0/Conv2d_1a_3x3/BatchNorm/batchnorm/add_1 | Add | [[1 192 7 7]] | 21 | 37632 | 0 | 45073920 | GPU_0_bfc | 0 | 0 | 0 | 0 | void Eigen::internal::EigenMetaKernel<Eigen::TensorEvaluator<Eigen::TensorAssignOp<Eigen::TensorMap<Eigen::Tensor<float, 2, 1, int>, 16, Eigen::MakePointer>, Eigen::TensorCwiseBinaryOp<Eigen::internal::scalar_sum_op<float, float>, Eigen::TensorMap<Eigen::Tensor<float const, 2, 1, int>, 16, Eigen::MakePointer> const, Eigen::TensorBroadcastingOp<Eigen::array<long, 2ul> const, Eigen::TensorMap<Eigen::Tensor<float const, 2, 1, int>, 16, Eigen::MakePointer> const> const> const> const, Eigen::GpuDevice>, int>(Eigen::TensorEvaluator<Eigen::TensorAssignOp<Eigen::TensorMap<Eigen::Tensor<float, 2, 1, int>, 16, Eigen::MakePointer>, Eigen::TensorCwiseBinaryOp<Eigen::internal::scalar_sum_op<float, float>, Eigen::TensorMap<Eigen::Tensor<float const, 2, 1, int>, 16, Eigen::MakePointer> const, Eigen::TensorBroadcastingOp<Eigen::array<long, 2ul> const, Eigen::TensorMap<Eigen::Tensor<float const, 2, 1, int>, 16, Eigen::MakePointer> const> const> const> const, Eigen::GpuDevice>, int) | 5.00 | 9408 | 0.00 | 0.00 | 0.00 | 0.00 | 1.88 | true | 9408;9408;9408;9408;9408 |
| 162 | InceptionV2/InceptionV2/Mixed_5a/Branch_1/Conv2d_0b_3x3/Relu | Relu | [[1 256 14 14]] | 17.667 | 200704 | 0 | 45073920 | GPU_0_bfc | 0 | 0 | 0 | 0 | void Eigen::internal::EigenMetaKernel<Eigen::TensorEvaluator<Eigen::TensorAssignOp<Eigen::TensorMap<Eigen::Tensor<float, 1, 1, long>, 16, Eigen::MakePointer>, Eigen::TensorCwiseBinaryOp<Eigen::internal::scalar_max_op<float const, float const>, Eigen::TensorMap<Eigen::Tensor<float const, 1, 1, long>, 16, Eigen::MakePointer> const, Eigen::TensorCwiseNullaryOp<Eigen::internal::scalar_constant_op<float const>, Eigen::TensorMap<Eigen::Tensor<float const, 1, 1, long>, 16, Eigen::MakePointer> const> const> const> const, Eigen::GpuDevice>, long>(Eigen::TensorEvaluator<Eigen::TensorAssignOp<Eigen::TensorMap<Eigen::Tensor<float, 1, 1, long>, 16, Eigen::MakePointer>, Eigen::TensorCwiseBinaryOp<Eigen::internal::scalar_max_op<float const, float const>, Eigen::TensorMap<Eigen::Tensor<float const, 1, 1, long>, 16, Eigen::MakePointer> const, Eigen::TensorCwiseNullaryOp<Eigen::internal::scalar_constant_op<float const>, Eigen::TensorMap<Eigen::Tensor<float const, 1, 1, long>, 16, Eigen::MakePointer> const> const> const> const, Eigen::GpuDevice>, long) | 3.00 | 0 | 0.00 | 0.00 | 0.00 | 0.00 | 0.00 | true | 0;0;0;0;0 |
| 163 | InceptionV2/InceptionV2/Mixed_5a/Branch_0/Conv2d_1a_3x3/Relu | Relu | [[1 192 7 7]] | 18.333 | 37632 | 0 | 45073920 | GPU_0_bfc | 0 | 0 | 0 | 0 | void Eigen::internal::EigenMetaKernel<Eigen::TensorEvaluator<Eigen::TensorAssignOp<Eigen::TensorMap<Eigen::Tensor<float, 1, 1, long>, 16, Eigen::MakePointer>, Eigen::TensorCwiseBinaryOp<Eigen::internal::scalar_max_op<float const, float const>, Eigen::TensorMap<Eigen::Tensor<float const, 1, 1, long>, 16, Eigen::MakePointer> const, Eigen::TensorCwiseNullaryOp<Eigen::internal::scalar_constant_op<float const>, Eigen::TensorMap<Eigen::Tensor<float const, 1, 1, long>, 16, Eigen::MakePointer> const> const> const> const, Eigen::GpuDevice>, long>(Eigen::TensorEvaluator<Eigen::TensorAssignOp<Eigen::TensorMap<Eigen::Tensor<float, 1, 1, long>, 16, Eigen::MakePointer>, Eigen::TensorCwiseBinaryOp<Eigen::internal::scalar_max_op<float const, float const>, Eigen::TensorMap<Eigen::Tensor<float const, 1, 1, long>, 16, Eigen::MakePointer> const, Eigen::TensorCwiseNullaryOp<Eigen::internal::scalar_constant_op<float const>, Eigen::TensorMap<Eigen::Tensor<float const, 1, 1, long>, 16, Eigen::MakePointer> const> const> const> const, Eigen::GpuDevice>, long) | 3.00 | 0 | 0.00 | 0.00 | 0.00 | 0.00 | 0.00 | true | 0;0;0;0;0 |
| 164 | InceptionV2/InceptionV2/Mixed_5a/Branch_1/Conv2d_1a_3x3/BatchNorm/batchnorm/mul | Conv2D | [[1 256 7 7]] | 404.667 | 50176 | 3091456 | 45124096 | GPU_0_bfc | 3041280 | 0 | 0 | 0 | void cudnn::detail::explicit_convolve_sgemm<float, int, 1024, 5, 5, 3, 3, 3, 0, true>(int, int, int, float const*, int, float const*, int, float*, kernel_conv_params, int, int, float, float, int, float*, float*) | 192.00 | 75510016 | 0.00 | 0.00 | 0.00 | 0.00 | 393.28 | true | 75510016;75510016;75510016;75510016;75510016 |
| 164 | InceptionV2/InceptionV2/Mixed_5a/Branch_1/Conv2d_1a_3x3/BatchNorm/batchnorm/mul | Conv2D | [[1 256 7 7]] | 404.667 | 50176 | 3091456 | 45124096 | GPU_0_bfc | 3041280 | 0 | 0 | 0 | void im2col4d_kernel<float, int>(im2col4d_params, cudnnConvolutionStruct, cudnnTensor4dStruct, float const*, float*, int) | 87.67 | 0 | 0.00 | 0.00 | 0.00 | 0.00 | 0.00 | true | 0;0;0;0;0 |
| 164 | InceptionV2/InceptionV2/Mixed_5a/Branch_1/Conv2d_1a_3x3/BatchNorm/batchnorm/mul | Conv2D | [[1 256 7 7]] | 404.667 | 50176 | 3091456 | 45124096 | GPU_0_bfc | 3041280 | 0 | 0 | 0 | void tensorflow::functor::ShuffleInTensor3Simple<float, 2, 1, 0, false>(int, float const*, tensorflow::functor::Dimension<3>, float*) | 28.00 | 0 | 0.00 | 0.00 | 0.00 | 0.00 | 0.00 | true | 0;0;0;0;0 |
| 164 | InceptionV2/InceptionV2/Mixed_5a/Branch_1/Conv2d_1a_3x3/BatchNorm/batchnorm/mul | Conv2D | [[1 256 7 7]] | 404.667 | 50176 | 3091456 | 45124096 | GPU_0_bfc | 3041280 | 0 | 0 | 0 | void tensorflow::functor::PadInputCustomKernelNCHW<float, 4>(int, float const*, tensorflow::functor::Dimension<4>, float*, tensorflow::functor::Dimension<4>, tensorflow::functor::Dimension<(4)-(2)>) | 6.00 | 0 | 0.00 | 0.00 | 0.00 | 0.00 | 0.00 | true | 0;0;0;0;0 |
| 165 | InceptionV2/InceptionV2/Mixed_5a/Branch_1/Conv2d_1a_3x3/BatchNorm/batchnorm/add_1 | Add | [[1 256 7 7]] | 26 | 50176 | 0 | 44923392 | GPU_0_bfc | 0 | 0 | 0 | 0 | void Eigen::internal::EigenMetaKernel<Eigen::TensorEvaluator<Eigen::TensorAssignOp<Eigen::TensorMap<Eigen::Tensor<float, 2, 1, int>, 16, Eigen::MakePointer>, Eigen::TensorCwiseBinaryOp<Eigen::internal::scalar_sum_op<float, float>, Eigen::TensorMap<Eigen::Tensor<float const, 2, 1, int>, 16, Eigen::MakePointer> const, Eigen::TensorBroadcastingOp<Eigen::array<long, 2ul> const, Eigen::TensorMap<Eigen::Tensor<float const, 2, 1, int>, 16, Eigen::MakePointer> const> const> const> const, Eigen::GpuDevice>, int>(Eigen::TensorEvaluator<Eigen::TensorAssignOp<Eigen::TensorMap<Eigen::Tensor<float, 2, 1, int>, 16, Eigen::MakePointer>, Eigen::TensorCwiseBinaryOp<Eigen::internal::scalar_sum_op<float, float>, Eigen::TensorMap<Eigen::Tensor<float const, 2, 1, int>, 16, Eigen::MakePointer> const, Eigen::TensorBroadcastingOp<Eigen::array<long, 2ul> const, Eigen::TensorMap<Eigen::Tensor<float const, 2, 1, int>, 16, Eigen::MakePointer> const> const> const> const, Eigen::GpuDevice>, int) | 5.00 | 12544 | 0.00 | 0.00 | 0.00 | 0.00 | 2.51 | true | 12544;12544;12544;12544;12544 |
| 166 | InceptionV2/InceptionV2/Mixed_5a/Branch_1/Conv2d_1a_3x3/Relu | Relu | [[1 256 7 7]] | 19 | 50176 | 0 | 44923392 | GPU_0_bfc | 0 | 0 | 0 | 0 | void Eigen::internal::EigenMetaKernel<Eigen::TensorEvaluator<Eigen::TensorAssignOp<Eigen::TensorMap<Eigen::Tensor<float, 1, 1, long>, 16, Eigen::MakePointer>, Eigen::TensorCwiseBinaryOp<Eigen::internal::scalar_max_op<float const, float const>, Eigen::TensorMap<Eigen::Tensor<float const, 1, 1, long>, 16, Eigen::MakePointer> const, Eigen::TensorCwiseNullaryOp<Eigen::internal::scalar_constant_op<float const>, Eigen::TensorMap<Eigen::Tensor<float const, 1, 1, long>, 16, Eigen::MakePointer> const> const> const> const, Eigen::GpuDevice>, long>(Eigen::TensorEvaluator<Eigen::TensorAssignOp<Eigen::TensorMap<Eigen::Tensor<float, 1, 1, long>, 16, Eigen::MakePointer>, Eigen::TensorCwiseBinaryOp<Eigen::internal::scalar_max_op<float const, float const>, Eigen::TensorMap<Eigen::Tensor<float const, 1, 1, long>, 16, Eigen::MakePointer> const, Eigen::TensorCwiseNullaryOp<Eigen::internal::scalar_constant_op<float const>, Eigen::TensorMap<Eigen::Tensor<float const, 1, 1, long>, 16, Eigen::MakePointer> const> const> const> const, Eigen::GpuDevice>, long) | 3.00 | 0 | 0.00 | 0.00 | 0.00 | 0.00 | 0.00 | true | 0;0;0;0;0 |
| 168 | InceptionV2/InceptionV2/Mixed_5b/Branch_3/AvgPool_0a_3x3/AvgPool | AvgPool | [[1 1024 7 7]] | 41.333 | 200704 | 200704 | 45124096 | GPU_0_bfc | 0 | 0 | 0 | 0 | void cudnn::detail::pooling_fw_4d_kernel<float, float, cudnn::detail::averpooling_func<float>, 2, false>(cudnnTensorStruct, float const*, cudnnTensorStruct, float*, cudnnPoolingStruct, float, float, int, cudnn::reduced_divisor, cudnn::reduced_divisor) | 10.00 | 1087642 | 0.00 | 0.00 | 0.00 | 0.00 | 108.76 | true | 1087642;1087642;1087642;1087642;1087642 |
| 169 | InceptionV2/InceptionV2/Mixed_5b/Branch_2/Conv2d_0a_1x1/BatchNorm/batchnorm/mul | Conv2D | [[1 160 7 7]] | 261 | 31488 | 887552 | 45155584 | GPU_0_bfc | 856064 | 0 | 0 | 0 | void im2col4d_kernel<float, int>(im2col4d_params, cudnnConvolutionStruct, cudnnTensor4dStruct, float const*, float*, int) | 105.00 | 0 | 0.00 | 0.00 | 0.00 | 0.00 | 0.00 | true | 0;0;0;0;0 |
| 169 | InceptionV2/InceptionV2/Mixed_5b/Branch_2/Conv2d_0a_1x1/BatchNorm/batchnorm/mul | Conv2D | [[1 160 7 7]] | 261 | 31488 | 887552 | 45155584 | GPU_0_bfc | 856064 | 0 | 0 | 0 | void cudnn::detail::explicit_convolve_sgemm<float, int, 128, 5, 5, 3, 3, 3, 0, true>(int, int, int, float const*, int, float const*, int, float*, kernel_conv_params, int, int, float, float, int, float*, float*) | 62.00 | 20979360 | 0.00 | 0.00 | 0.00 | 0.00 | 338.38 | true | 20979360;20979360;20979360;20979360;20979360 |
| 169 | InceptionV2/InceptionV2/Mixed_5b/Branch_2/Conv2d_0a_1x1/BatchNorm/batchnorm/mul | Conv2D | [[1 160 7 7]] | 261 | 31488 | 887552 | 45155584 | GPU_0_bfc | 856064 | 0 | 0 | 0 | void tensorflow::functor::ShuffleInTensor3Simple<float, 2, 1, 0, false>(int, float const*, tensorflow::functor::Dimension<3>, float*) | 11.00 | 0 | 0.00 | 0.00 | 0.00 | 0.00 | 0.00 | true | 0;0;0;0;0 |
| 170 | InceptionV2/InceptionV2/Mixed_5b/Branch_1/Conv2d_0a_1x1/BatchNorm/batchnorm/mul | Conv2D | [[1 192 7 7]] | 268.333 | 37632 | 1024768 | 45193216 | GPU_0_bfc | 987136 | 0 | 0 | 0 | void im2col4d_kernel<float, int>(im2col4d_params, cudnnConvolutionStruct, cudnnTensor4dStruct, float const*, float*, int) | 105.00 | 0 | 0.00 | 0.00 | 0.00 | 0.00 | 0.00 | true | 0;0;0;0;0 |
| 170 | InceptionV2/InceptionV2/Mixed_5b/Branch_1/Conv2d_0a_1x1/BatchNorm/batchnorm/mul | Conv2D | [[1 192 7 7]] | 268.333 | 37632 | 1024768 | 45193216 | GPU_0_bfc | 987136 | 0 | 0 | 0 | void cudnn::detail::explicit_convolve_sgemm<float, int, 128, 5, 5, 3, 3, 3, 0, true>(int, int, int, float const*, int, float const*, int, float*, kernel_conv_params, int, int, float, float, int, float*, float*) | 62.00 | 25175232 | 0.00 | 0.00 | 0.00 | 0.00 | 406.05 | true | 25175232;25175232;25175232;25175232;25175232 |
| 170 | InceptionV2/InceptionV2/Mixed_5b/Branch_1/Conv2d_0a_1x1/BatchNorm/batchnorm/mul | Conv2D | [[1 192 7 7]] | 268.333 | 37632 | 1024768 | 45193216 | GPU_0_bfc | 987136 | 0 | 0 | 0 | void tensorflow::functor::ShuffleInTensor3Simple<float, 2, 1, 0, false>(int, float const*, tensorflow::functor::Dimension<3>, float*) | 12.00 | 0 | 0.00 | 0.00 | 0.00 | 0.00 | 0.00 | true | 0;0;0;0;0 |
| 171 | InceptionV2/InceptionV2/Mixed_5b/Branch_0/Conv2d_0a_1x1/BatchNorm/batchnorm/mul | Conv2D | [[1 352 7 7]] | 267 | 69120 | 1711616 | 45262336 | GPU_0_bfc | 1642496 | 0 | 0 | 0 | void im2col4d_kernel<float, int>(im2col4d_params, cudnnConvolutionStruct, cudnnTensor4dStruct, float const*, float*, int) | 105.00 | 0 | 0.00 | 0.00 | 0.00 | 0.00 | 0.00 | true | 0;0;0;0;0 |
| 171 | InceptionV2/InceptionV2/Mixed_5b/Branch_0/Conv2d_0a_1x1/BatchNorm/batchnorm/mul | Conv2D | [[1 352 7 7]] | 267 | 69120 | 1711616 | 45262336 | GPU_0_bfc | 1642496 | 0 | 0 | 0 | void cudnn::detail::explicit_convolve_sgemm<float, int, 1024, 5, 5, 3, 3, 3, 0, true>(int, int, int, float const*, int, float const*, int, float*, kernel_conv_params, int, int, float, float, int, float*, float*) | 62.83 | 46154592 | 0.00 | 0.00 | 0.00 | 0.00 | 734.56 | true | 46154592;46154592;46154592;46154592;46154592 |
| 171 | InceptionV2/InceptionV2/Mixed_5b/Branch_0/Conv2d_0a_1x1/BatchNorm/batchnorm/mul | Conv2D | [[1 352 7 7]] | 267 | 69120 | 1711616 | 45262336 | GPU_0_bfc | 1642496 | 0 | 0 | 0 | void tensorflow::functor::ShuffleInTensor3Simple<float, 2, 1, 0, false>(int, float const*, tensorflow::functor::Dimension<3>, float*) | 20.17 | 0 | 0.00 | 0.00 | 0.00 | 0.00 | 0.00 | true | 0;0;0;0;0 |
| 172 | InceptionV2/InceptionV2/Mixed_5b/Branch_3/Conv2d_0b_1x1/BatchNorm/batchnorm/mul | Conv2D | [[1 128 7 7]] | 321.333 | 25088 | 750080 | 45086720 | GPU_0_bfc | 724992 | 0 | 0 | 0 | void im2col4d_kernel<float, int>(im2col4d_params, cudnnConvolutionStruct, cudnnTensor4dStruct, float const*, float*, int) | 164.17 | 0 | 0.00 | 0.00 | 0.00 | 0.00 | 0.00 | true | 0;0;0;0;0 |
| 172 | InceptionV2/InceptionV2/Mixed_5b/Branch_3/Conv2d_0b_1x1/BatchNorm/batchnorm/mul | Conv2D | [[1 128 7 7]] | 321.333 | 25088 | 750080 | 45086720 | GPU_0_bfc | 724992 | 0 | 0 | 0 | void cudnn::detail::explicit_convolve_sgemm<float, int, 128, 5, 5, 3, 3, 3, 0, true>(int, int, int, float const*, int, float const*, int, float*, kernel_conv_params, int, int, float, float, int, float*, float*) | 62.00 | 16783488 | 0.00 | 0.00 | 0.00 | 0.00 | 270.70 | true | 16783488;16783488;16783488;16783488;16783488 |
| 172 | InceptionV2/InceptionV2/Mixed_5b/Branch_3/Conv2d_0b_1x1/BatchNorm/batchnorm/mul | Conv2D | [[1 128 7 7]] | 321.333 | 25088 | 750080 | 45086720 | GPU_0_bfc | 724992 | 0 | 0 | 0 | void tensorflow::functor::ShuffleInTensor3Simple<float, 2, 1, 0, false>(int, float const*, tensorflow::functor::Dimension<3>, float*) | 10.00 | 0 | 0.00 | 0.00 | 0.00 | 0.00 | 0.00 | true | 0;0;0;0;0 |
| 173 | InceptionV2/InceptionV2/Mixed_5b/Branch_2/Conv2d_0a_1x1/BatchNorm/batchnorm/add_1 | Add | [[1 160 7 7]] | 26.333 | 31488 | 0 | 44886016 | GPU_0_bfc | 0 | 0 | 0 | 0 | void Eigen::internal::EigenMetaKernel<Eigen::TensorEvaluator<Eigen::TensorAssignOp<Eigen::TensorMap<Eigen::Tensor<float, 2, 1, int>, 16, Eigen::MakePointer>, Eigen::TensorCwiseBinaryOp<Eigen::internal::scalar_sum_op<float, float>, Eigen::TensorMap<Eigen::Tensor<float const, 2, 1, int>, 16, Eigen::MakePointer> const, Eigen::TensorBroadcastingOp<Eigen::array<long, 2ul> const, Eigen::TensorMap<Eigen::Tensor<float const, 2, 1, int>, 16, Eigen::MakePointer> const> const> const> const, Eigen::GpuDevice>, int>(Eigen::TensorEvaluator<Eigen::TensorAssignOp<Eigen::TensorMap<Eigen::Tensor<float, 2, 1, int>, 16, Eigen::MakePointer>, Eigen::TensorCwiseBinaryOp<Eigen::internal::scalar_sum_op<float, float>, Eigen::TensorMap<Eigen::Tensor<float const, 2, 1, int>, 16, Eigen::MakePointer> const, Eigen::TensorBroadcastingOp<Eigen::array<long, 2ul> const, Eigen::TensorMap<Eigen::Tensor<float const, 2, 1, int>, 16, Eigen::MakePointer> const> const> const> const, Eigen::GpuDevice>, int) | 5.00 | 7840 | 0.00 | 0.00 | 0.00 | 0.00 | 1.57 | true | 7840;7840;7840;7840;7840 |
| 174 | InceptionV2/InceptionV2/Mixed_5b/Branch_1/Conv2d_0a_1x1/BatchNorm/batchnorm/add_1 | Add | [[1 192 7 7]] | 20.667 | 37632 | 0 | 44886016 | GPU_0_bfc | 0 | 0 | 0 | 0 | void Eigen::internal::EigenMetaKernel<Eigen::TensorEvaluator<Eigen::TensorAssignOp<Eigen::TensorMap<Eigen::Tensor<float, 2, 1, int>, 16, Eigen::MakePointer>, Eigen::TensorCwiseBinaryOp<Eigen::internal::scalar_sum_op<float, float>, Eigen::TensorMap<Eigen::Tensor<float const, 2, 1, int>, 16, Eigen::MakePointer> const, Eigen::TensorBroadcastingOp<Eigen::array<long, 2ul> const, Eigen::TensorMap<Eigen::Tensor<float const, 2, 1, int>, 16, Eigen::MakePointer> const> const> const> const, Eigen::GpuDevice>, int>(Eigen::TensorEvaluator<Eigen::TensorAssignOp<Eigen::TensorMap<Eigen::Tensor<float, 2, 1, int>, 16, Eigen::MakePointer>, Eigen::TensorCwiseBinaryOp<Eigen::internal::scalar_sum_op<float, float>, Eigen::TensorMap<Eigen::Tensor<float const, 2, 1, int>, 16, Eigen::MakePointer> const, Eigen::TensorBroadcastingOp<Eigen::array<long, 2ul> const, Eigen::TensorMap<Eigen::Tensor<float const, 2, 1, int>, 16, Eigen::MakePointer> const> const> const> const, Eigen::GpuDevice>, int) | 5.00 | 9408 | 0.00 | 0.00 | 0.00 | 0.00 | 1.88 | true | 9408;9408;9408;9408;9408 |
| 175 | InceptionV2/InceptionV2/Mixed_5b/Branch_0/Conv2d_0a_1x1/BatchNorm/batchnorm/add_1 | Add | [[1 352 7 7]] | 20.667 | 69120 | 0 | 44886016 | GPU_0_bfc | 0 | 0 | 0 | 0 | void Eigen::internal::EigenMetaKernel<Eigen::TensorEvaluator<Eigen::TensorAssignOp<Eigen::TensorMap<Eigen::Tensor<float, 2, 1, int>, 16, Eigen::MakePointer>, Eigen::TensorCwiseBinaryOp<Eigen::internal::scalar_sum_op<float, float>, Eigen::TensorMap<Eigen::Tensor<float const, 2, 1, int>, 16, Eigen::MakePointer> const, Eigen::TensorBroadcastingOp<Eigen::array<long, 2ul> const, Eigen::TensorMap<Eigen::Tensor<float const, 2, 1, int>, 16, Eigen::MakePointer> const> const> const> const, Eigen::GpuDevice>, int>(Eigen::TensorEvaluator<Eigen::TensorAssignOp<Eigen::TensorMap<Eigen::Tensor<float, 2, 1, int>, 16, Eigen::MakePointer>, Eigen::TensorCwiseBinaryOp<Eigen::internal::scalar_sum_op<float, float>, Eigen::TensorMap<Eigen::Tensor<float const, 2, 1, int>, 16, Eigen::MakePointer> const, Eigen::TensorBroadcastingOp<Eigen::array<long, 2ul> const, Eigen::TensorMap<Eigen::Tensor<float const, 2, 1, int>, 16, Eigen::MakePointer> const> const> const> const, Eigen::GpuDevice>, int) | 5.00 | 17248 | 0.00 | 0.00 | 0.00 | 0.00 | 3.45 | true | 17248;17248;17248;17248;17248 |
| 176 | InceptionV2/InceptionV2/Mixed_5b/Branch_3/Conv2d_0b_1x1/BatchNorm/batchnorm/add_1 | Add | [[1 128 7 7]] | 19 | 25088 | 0 | 44886016 | GPU_0_bfc | 0 | 0 | 0 | 0 | void Eigen::internal::EigenMetaKernel<Eigen::TensorEvaluator<Eigen::TensorAssignOp<Eigen::TensorMap<Eigen::Tensor<float, 2, 1, int>, 16, Eigen::MakePointer>, Eigen::TensorCwiseBinaryOp<Eigen::internal::scalar_sum_op<float, float>, Eigen::TensorMap<Eigen::Tensor<float const, 2, 1, int>, 16, Eigen::MakePointer> const, Eigen::TensorBroadcastingOp<Eigen::array<long, 2ul> const, Eigen::TensorMap<Eigen::Tensor<float const, 2, 1, int>, 16, Eigen::MakePointer> const> const> const> const, Eigen::GpuDevice>, int>(Eigen::TensorEvaluator<Eigen::TensorAssignOp<Eigen::TensorMap<Eigen::Tensor<float, 2, 1, int>, 16, Eigen::MakePointer>, Eigen::TensorCwiseBinaryOp<Eigen::internal::scalar_sum_op<float, float>, Eigen::TensorMap<Eigen::Tensor<float const, 2, 1, int>, 16, Eigen::MakePointer> const, Eigen::TensorBroadcastingOp<Eigen::array<long, 2ul> const, Eigen::TensorMap<Eigen::Tensor<float const, 2, 1, int>, 16, Eigen::MakePointer> const> const> const> const, Eigen::GpuDevice>, int) | 5.00 | 6272 | 0.00 | 0.00 | 0.00 | 0.00 | 1.25 | true | 6272;6272;6272;6272;6272 |
| 177 | InceptionV2/InceptionV2/Mixed_5b/Branch_2/Conv2d_0a_1x1/Relu | Relu | [[1 160 7 7]] | 19.667 | 31488 | 0 | 44886016 | GPU_0_bfc | 0 | 0 | 0 | 0 | void Eigen::internal::EigenMetaKernel<Eigen::TensorEvaluator<Eigen::TensorAssignOp<Eigen::TensorMap<Eigen::Tensor<float, 1, 1, long>, 16, Eigen::MakePointer>, Eigen::TensorCwiseBinaryOp<Eigen::internal::scalar_max_op<float const, float const>, Eigen::TensorMap<Eigen::Tensor<float const, 1, 1, long>, 16, Eigen::MakePointer> const, Eigen::TensorCwiseNullaryOp<Eigen::internal::scalar_constant_op<float const>, Eigen::TensorMap<Eigen::Tensor<float const, 1, 1, long>, 16, Eigen::MakePointer> const> const> const> const, Eigen::GpuDevice>, long>(Eigen::TensorEvaluator<Eigen::TensorAssignOp<Eigen::TensorMap<Eigen::Tensor<float, 1, 1, long>, 16, Eigen::MakePointer>, Eigen::TensorCwiseBinaryOp<Eigen::internal::scalar_max_op<float const, float const>, Eigen::TensorMap<Eigen::Tensor<float const, 1, 1, long>, 16, Eigen::MakePointer> const, Eigen::TensorCwiseNullaryOp<Eigen::internal::scalar_constant_op<float const>, Eigen::TensorMap<Eigen::Tensor<float const, 1, 1, long>, 16, Eigen::MakePointer> const> const> const> const, Eigen::GpuDevice>, long) | 3.00 | 0 | 0.00 | 0.00 | 0.00 | 0.00 | 0.00 | true | 0;0;0;0;0 |
| 178 | InceptionV2/InceptionV2/Mixed_5b/Branch_1/Conv2d_0a_1x1/Relu | Relu | [[1 192 7 7]] | 17 | 37632 | 0 | 44886016 | GPU_0_bfc | 0 | 0 | 0 | 0 | void Eigen::internal::EigenMetaKernel<Eigen::TensorEvaluator<Eigen::TensorAssignOp<Eigen::TensorMap<Eigen::Tensor<float, 1, 1, long>, 16, Eigen::MakePointer>, Eigen::TensorCwiseBinaryOp<Eigen::internal::scalar_max_op<float const, float const>, Eigen::TensorMap<Eigen::Tensor<float const, 1, 1, long>, 16, Eigen::MakePointer> const, Eigen::TensorCwiseNullaryOp<Eigen::internal::scalar_constant_op<float const>, Eigen::TensorMap<Eigen::Tensor<float const, 1, 1, long>, 16, Eigen::MakePointer> const> const> const> const, Eigen::GpuDevice>, long>(Eigen::TensorEvaluator<Eigen::TensorAssignOp<Eigen::TensorMap<Eigen::Tensor<float, 1, 1, long>, 16, Eigen::MakePointer>, Eigen::TensorCwiseBinaryOp<Eigen::internal::scalar_max_op<float const, float const>, Eigen::TensorMap<Eigen::Tensor<float const, 1, 1, long>, 16, Eigen::MakePointer> const, Eigen::TensorCwiseNullaryOp<Eigen::internal::scalar_constant_op<float const>, Eigen::TensorMap<Eigen::Tensor<float const, 1, 1, long>, 16, Eigen::MakePointer> const> const> const> const, Eigen::GpuDevice>, long) | 3.00 | 0 | 0.00 | 0.00 | 0.00 | 0.00 | 0.00 | true | 0;0;0;0;0 |
| 179 | InceptionV2/InceptionV2/Mixed_5b/Branch_2/Conv2d_0b_3x3/BatchNorm/batchnorm/mul | Conv2D | [[1 224 7 7]] | 176.333 | 44032 | 4919296 | 44930048 | GPU_0_bfc | 4875264 | 0 | 0 | 0 | maxwell_scudnn_winograd_128x128_ldg1_ldg4_tile148n_nt | 51.33 | 38406144 | 0.00 | 0.00 | 0.00 | 0.00 | 748.18 | true | 38406144;38406144;38406144;38406144;38406144 |
| 179 | InceptionV2/InceptionV2/Mixed_5b/Branch_2/Conv2d_0b_3x3/BatchNorm/batchnorm/mul | Conv2D | [[1 224 7 7]] | 176.333 | 44032 | 4919296 | 44930048 | GPU_0_bfc | 4875264 | 0 | 0 | 0 | void tensorflow::functor::ShuffleInTensor3Simple<float, 2, 1, 0, false>(int, float const*, tensorflow::functor::Dimension<3>, float*) | 17.00 | 0 | 0.00 | 0.00 | 0.00 | 0.00 | 0.00 | true | 0;0;0;0;0 |
| 179 | InceptionV2/InceptionV2/Mixed_5b/Branch_2/Conv2d_0b_3x3/BatchNorm/batchnorm/mul | Conv2D | [[1 224 7 7]] | 176.333 | 44032 | 4919296 | 44930048 | GPU_0_bfc | 4875264 | 0 | 0 | 0 | void cudnn::winograd::generateWinogradTilesKernel<0, float, float>(cudnn::winograd::GenerateWinogradTilesParams<float, float>) | 8.17 | 2078720 | 0.00 | 0.00 | 0.00 | 0.00 | 254.53 | true | 2078720;2078720;2078720;2078720;2078720 |
| 180 | InceptionV2/InceptionV2/Mixed_5b/Branch_1/Conv2d_0b_3x3/BatchNorm/batchnorm/mul | Conv2D | [[1 320 7 7]] | 196.333 | 62720 | 8419840 | 44961280 | GPU_0_bfc | 8357120 | 0 | 0 | 0 | maxwell_scudnn_winograd_128x128_ldg1_ldg4_tile148n_nt | 60.67 | 65761280 | 0.00 | 0.00 | 0.00 | 0.00 | 1083.97 | true | 65761280;65761280;65761280;65761280;65761280 |
| 180 | InceptionV2/InceptionV2/Mixed_5b/Branch_1/Conv2d_0b_3x3/BatchNorm/batchnorm/mul | Conv2D | [[1 320 7 7]] | 196.333 | 62720 | 8419840 | 44961280 | GPU_0_bfc | 8357120 | 0 | 0 | 0 | void tensorflow::functor::ShuffleInTensor3Simple<float, 2, 1, 0, false>(int, float const*, tensorflow::functor::Dimension<3>, float*) | 27.17 | 0 | 0.00 | 0.00 | 0.00 | 0.00 | 0.00 | true | 0;0;0;0;0 |
| 180 | InceptionV2/InceptionV2/Mixed_5b/Branch_1/Conv2d_0b_3x3/BatchNorm/batchnorm/mul | Conv2D | [[1 320 7 7]] | 196.333 | 62720 | 8419840 | 44961280 | GPU_0_bfc | 8357120 | 0 | 0 | 0 | void cudnn::winograd::generateWinogradTilesKernel<0, float, float>(cudnn::winograd::GenerateWinogradTilesParams<float, float>) | 14.00 | 3563520 | 0.00 | 0.00 | 0.00 | 0.00 | 254.54 | true | 3563520;3563520;3563520;3563520;3563520 |
| 181 | InceptionV2/InceptionV2/Mixed_5b/Branch_2/Conv2d_0b_3x3/BatchNorm/batchnorm/add_1 | Add | [[1 224 7 7]] | 24.333 | 44032 | 0 | 44923648 | GPU_0_bfc | 0 | 0 | 0 | 0 | void Eigen::internal::EigenMetaKernel<Eigen::TensorEvaluator<Eigen::TensorAssignOp<Eigen::TensorMap<Eigen::Tensor<float, 2, 1, int>, 16, Eigen::MakePointer>, Eigen::TensorCwiseBinaryOp<Eigen::internal::scalar_sum_op<float, float>, Eigen::TensorMap<Eigen::Tensor<float const, 2, 1, int>, 16, Eigen::MakePointer> const, Eigen::TensorBroadcastingOp<Eigen::array<long, 2ul> const, Eigen::TensorMap<Eigen::Tensor<float const, 2, 1, int>, 16, Eigen::MakePointer> const> const> const> const, Eigen::GpuDevice>, int>(Eigen::TensorEvaluator<Eigen::TensorAssignOp<Eigen::TensorMap<Eigen::Tensor<float, 2, 1, int>, 16, Eigen::MakePointer>, Eigen::TensorCwiseBinaryOp<Eigen::internal::scalar_sum_op<float, float>, Eigen::TensorMap<Eigen::Tensor<float const, 2, 1, int>, 16, Eigen::MakePointer> const, Eigen::TensorBroadcastingOp<Eigen::array<long, 2ul> const, Eigen::TensorMap<Eigen::Tensor<float const, 2, 1, int>, 16, Eigen::MakePointer> const> const> const> const, Eigen::GpuDevice>, int) | 5.00 | 10976 | 0.00 | 0.00 | 0.00 | 0.00 | 2.20 | true | 10976;10976;10976;10976;10976 |
| 182 | InceptionV2/InceptionV2/Mixed_5b/Branch_1/Conv2d_0b_3x3/BatchNorm/batchnorm/add_1 | Add | [[1 320 7 7]] | 21.333 | 62720 | 0 | 44923648 | GPU_0_bfc | 0 | 0 | 0 | 0 | void Eigen::internal::EigenMetaKernel<Eigen::TensorEvaluator<Eigen::TensorAssignOp<Eigen::TensorMap<Eigen::Tensor<float, 2, 1, int>, 16, Eigen::MakePointer>, Eigen::TensorCwiseBinaryOp<Eigen::internal::scalar_sum_op<float, float>, Eigen::TensorMap<Eigen::Tensor<float const, 2, 1, int>, 16, Eigen::MakePointer> const, Eigen::TensorBroadcastingOp<Eigen::array<long, 2ul> const, Eigen::TensorMap<Eigen::Tensor<float const, 2, 1, int>, 16, Eigen::MakePointer> const> const> const> const, Eigen::GpuDevice>, int>(Eigen::TensorEvaluator<Eigen::TensorAssignOp<Eigen::TensorMap<Eigen::Tensor<float, 2, 1, int>, 16, Eigen::MakePointer>, Eigen::TensorCwiseBinaryOp<Eigen::internal::scalar_sum_op<float, float>, Eigen::TensorMap<Eigen::Tensor<float const, 2, 1, int>, 16, Eigen::MakePointer> const, Eigen::TensorBroadcastingOp<Eigen::array<long, 2ul> const, Eigen::TensorMap<Eigen::Tensor<float const, 2, 1, int>, 16, Eigen::MakePointer> const> const> const> const, Eigen::GpuDevice>, int) | 5.00 | 15680 | 0.00 | 0.00 | 0.00 | 0.00 | 3.14 | true | 15680;15680;15680;15680;15680 |
| 183 | InceptionV2/InceptionV2/Mixed_5b/Branch_2/Conv2d_0b_3x3/Relu | Relu | [[1 224 7 7]] | 17.667 | 44032 | 0 | 44923648 | GPU_0_bfc | 0 | 0 | 0 | 0 | void Eigen::internal::EigenMetaKernel<Eigen::TensorEvaluator<Eigen::TensorAssignOp<Eigen::TensorMap<Eigen::Tensor<float, 1, 1, long>, 16, Eigen::MakePointer>, Eigen::TensorCwiseBinaryOp<Eigen::internal::scalar_max_op<float const, float const>, Eigen::TensorMap<Eigen::Tensor<float const, 1, 1, long>, 16, Eigen::MakePointer> const, Eigen::TensorCwiseNullaryOp<Eigen::internal::scalar_constant_op<float const>, Eigen::TensorMap<Eigen::Tensor<float const, 1, 1, long>, 16, Eigen::MakePointer> const> const> const> const, Eigen::GpuDevice>, long>(Eigen::TensorEvaluator<Eigen::TensorAssignOp<Eigen::TensorMap<Eigen::Tensor<float, 1, 1, long>, 16, Eigen::MakePointer>, Eigen::TensorCwiseBinaryOp<Eigen::internal::scalar_max_op<float const, float const>, Eigen::TensorMap<Eigen::Tensor<float const, 1, 1, long>, 16, Eigen::MakePointer> const, Eigen::TensorCwiseNullaryOp<Eigen::internal::scalar_constant_op<float const>, Eigen::TensorMap<Eigen::Tensor<float const, 1, 1, long>, 16, Eigen::MakePointer> const> const> const> const, Eigen::GpuDevice>, long) | 3.00 | 0 | 0.00 | 0.00 | 0.00 | 0.00 | 0.00 | true | 0;0;0;0;0 |
| 184 | InceptionV2/InceptionV2/Mixed_5b/Branch_2/Conv2d_0c_3x3/BatchNorm/batchnorm/mul | Conv2D | [[1 224 7 7]] | 199.333 | 44032 | 6868992 | 44967680 | GPU_0_bfc | 6824960 | 0 | 0 | 0 | maxwell_scudnn_winograd_128x128_ldg1_ldg4_tile148n_nt | 68.50 | 53659648 | 0.00 | 0.00 | 0.00 | 0.00 | 783.35 | true | 53659648;53659648;53659648;53659648;53659648 |
| 184 | InceptionV2/InceptionV2/Mixed_5b/Branch_2/Conv2d_0c_3x3/BatchNorm/batchnorm/mul | Conv2D | [[1 224 7 7]] | 199.333 | 44032 | 6868992 | 44967680 | GPU_0_bfc | 6824960 | 0 | 0 | 0 | void tensorflow::functor::ShuffleInTensor3Simple<float, 2, 1, 0, false>(int, float const*, tensorflow::functor::Dimension<3>, float*) | 23.00 | 0 | 0.00 | 0.00 | 0.00 | 0.00 | 0.00 | true | 0;0;0;0;0 |
| 184 | InceptionV2/InceptionV2/Mixed_5b/Branch_2/Conv2d_0c_3x3/BatchNorm/batchnorm/mul | Conv2D | [[1 224 7 7]] | 199.333 | 44032 | 6868992 | 44967680 | GPU_0_bfc | 6824960 | 0 | 0 | 0 | void cudnn::winograd::generateWinogradTilesKernel<0, float, float>(cudnn::winograd::GenerateWinogradTilesParams<float, float>) | 11.00 | 2910208 | 0.00 | 0.00 | 0.00 | 0.00 | 264.56 | true | 2910208;2910208;2910208;2910208;2910208 |
| 185 | InceptionV2/InceptionV2/Mixed_5b/Branch_2/Conv2d_0c_3x3/BatchNorm/batchnorm/add_1 | Add | [[1 224 7 7]] | 24 | 44032 | 0 | 44923648 | GPU_0_bfc | 0 | 0 | 0 | 0 | void Eigen::internal::EigenMetaKernel<Eigen::TensorEvaluator<Eigen::TensorAssignOp<Eigen::TensorMap<Eigen::Tensor<float, 2, 1, int>, 16, Eigen::MakePointer>, Eigen::TensorCwiseBinaryOp<Eigen::internal::scalar_sum_op<float, float>, Eigen::TensorMap<Eigen::Tensor<float const, 2, 1, int>, 16, Eigen::MakePointer> const, Eigen::TensorBroadcastingOp<Eigen::array<long, 2ul> const, Eigen::TensorMap<Eigen::Tensor<float const, 2, 1, int>, 16, Eigen::MakePointer> const> const> const> const, Eigen::GpuDevice>, int>(Eigen::TensorEvaluator<Eigen::TensorAssignOp<Eigen::TensorMap<Eigen::Tensor<float, 2, 1, int>, 16, Eigen::MakePointer>, Eigen::TensorCwiseBinaryOp<Eigen::internal::scalar_sum_op<float, float>, Eigen::TensorMap<Eigen::Tensor<float const, 2, 1, int>, 16, Eigen::MakePointer> const, Eigen::TensorBroadcastingOp<Eigen::array<long, 2ul> const, Eigen::TensorMap<Eigen::Tensor<float const, 2, 1, int>, 16, Eigen::MakePointer> const> const> const> const, Eigen::GpuDevice>, int) | 5.00 | 10976 | 0.00 | 0.00 | 0.00 | 0.00 | 2.20 | true | 10976;10976;10976;10976;10976 |
| 187 | InceptionV2/InceptionV2/Mixed_5b/Branch_0/Conv2d_0a_1x1/Relu | Relu | [[1 1024 7 7]] | 19.667 | 294656 | 0 | 45017344 | GPU_0_bfc | 0 | 0 | 0 | 0 | void Eigen::internal::EigenMetaKernel<Eigen::TensorEvaluator<Eigen::TensorAssignOp<Eigen::TensorMap<Eigen::Tensor<float, 1, 1, long>, 16, Eigen::MakePointer>, Eigen::TensorCwiseBinaryOp<Eigen::internal::scalar_max_op<float const, float const>, Eigen::TensorMap<Eigen::Tensor<float const, 1, 1, long>, 16, Eigen::MakePointer> const, Eigen::TensorCwiseNullaryOp<Eigen::internal::scalar_constant_op<float const>, Eigen::TensorMap<Eigen::Tensor<float const, 1, 1, long>, 16, Eigen::MakePointer> const> const> const> const, Eigen::GpuDevice>, long>(Eigen::TensorEvaluator<Eigen::TensorAssignOp<Eigen::TensorMap<Eigen::Tensor<float, 1, 1, long>, 16, Eigen::MakePointer>, Eigen::TensorCwiseBinaryOp<Eigen::internal::scalar_max_op<float const, float const>, Eigen::TensorMap<Eigen::Tensor<float const, 1, 1, long>, 16, Eigen::MakePointer> const, Eigen::TensorCwiseNullaryOp<Eigen::internal::scalar_constant_op<float const>, Eigen::TensorMap<Eigen::Tensor<float const, 1, 1, long>, 16, Eigen::MakePointer> const> const> const> const, Eigen::GpuDevice>, long) | 3.00 | 0 | 0.00 | 0.00 | 0.00 | 0.00 | 0.00 | true | 0;0;0;0;0 |
| 188 | InceptionV2/InceptionV2/Mixed_5c/Branch_3/MaxPool_0a_3x3/MaxPool | MaxPool | [[1 1024 7 7]] | 36.667 | 200704 | 200704 | 45218048 | GPU_0_bfc | 0 | 0 | 0 | 0 | void cudnn::detail::pooling_fw_4d_kernel<float, float, cudnn::detail::maxpooling_func<float, (cudnnNanPropagation_t)0>, 0, false>(cudnnTensorStruct, float const*, cudnnTensorStruct, float*, cudnnPoolingStruct, float, float, int, cudnn::reduced_divisor, cudnn::reduced_divisor) | 7.17 | 50176 | 0.00 | 0.00 | 0.00 | 0.00 | 7.00 | true | 50176;50176;50176;50176;50176 |
| 189 | InceptionV2/InceptionV2/Mixed_5c/Branch_2/Conv2d_0a_1x1/BatchNorm/batchnorm/mul | Conv2D | [[1 192 7 7]] | 265 | 37632 | 1024768 | 45255680 | GPU_0_bfc | 987136 | 0 | 0 | 0 | void im2col4d_kernel<float, int>(im2col4d_params, cudnnConvolutionStruct, cudnnTensor4dStruct, float const*, float*, int) | 105.00 | 0 | 0.00 | 0.00 | 0.00 | 0.00 | 0.00 | true | 0;0;0;0;0 |
| 189 | InceptionV2/InceptionV2/Mixed_5c/Branch_2/Conv2d_0a_1x1/BatchNorm/batchnorm/mul | Conv2D | [[1 192 7 7]] | 265 | 37632 | 1024768 | 45255680 | GPU_0_bfc | 987136 | 0 | 0 | 0 | void cudnn::detail::explicit_convolve_sgemm<float, int, 128, 5, 5, 3, 3, 3, 0, true>(int, int, int, float const*, int, float const*, int, float*, kernel_conv_params, int, int, float, float, int, float*, float*) | 62.00 | 25175232 | 0.00 | 0.00 | 0.00 | 0.00 | 406.05 | true | 25175232;25175232;25175232;25175232;25175232 |
| 189 | InceptionV2/InceptionV2/Mixed_5c/Branch_2/Conv2d_0a_1x1/BatchNorm/batchnorm/mul | Conv2D | [[1 192 7 7]] | 265 | 37632 | 1024768 | 45255680 | GPU_0_bfc | 987136 | 0 | 0 | 0 | void tensorflow::functor::ShuffleInTensor3Simple<float, 2, 1, 0, false>(int, float const*, tensorflow::functor::Dimension<3>, float*) | 13.00 | 0 | 0.00 | 0.00 | 0.00 | 0.00 | 0.00 | true | 0;0;0;0;0 |
| 190 | InceptionV2/InceptionV2/Mixed_5c/Branch_1/Conv2d_0a_1x1/BatchNorm/batchnorm/mul | Conv2D | [[1 192 7 7]] | 251 | 37632 | 1024768 | 45293312 | GPU_0_bfc | 987136 | 0 | 0 | 0 | void im2col4d_kernel<float, int>(im2col4d_params, cudnnConvolutionStruct, cudnnTensor4dStruct, float const*, float*, int) | 105.00 | 0 | 0.00 | 0.00 | 0.00 | 0.00 | 0.00 | true | 0;0;0;0;0 |
| 190 | InceptionV2/InceptionV2/Mixed_5c/Branch_1/Conv2d_0a_1x1/BatchNorm/batchnorm/mul | Conv2D | [[1 192 7 7]] | 251 | 37632 | 1024768 | 45293312 | GPU_0_bfc | 987136 | 0 | 0 | 0 | void cudnn::detail::explicit_convolve_sgemm<float, int, 128, 5, 5, 3, 3, 3, 0, true>(int, int, int, float const*, int, float const*, int, float*, kernel_conv_params, int, int, float, float, int, float*, float*) | 62.00 | 25175232 | 0.00 | 0.00 | 0.00 | 0.00 | 406.05 | true | 25175232;25175232;25175232;25175232;25175232 |
| 190 | InceptionV2/InceptionV2/Mixed_5c/Branch_1/Conv2d_0a_1x1/BatchNorm/batchnorm/mul | Conv2D | [[1 192 7 7]] | 251 | 37632 | 1024768 | 45293312 | GPU_0_bfc | 987136 | 0 | 0 | 0 | void tensorflow::functor::ShuffleInTensor3Simple<float, 2, 1, 0, false>(int, float const*, tensorflow::functor::Dimension<3>, float*) | 12.00 | 0 | 0.00 | 0.00 | 0.00 | 0.00 | 0.00 | true | 0;0;0;0;0 |
| 191 | InceptionV2/InceptionV2/Mixed_5c/Branch_0/Conv2d_0a_1x1/BatchNorm/batchnorm/mul | Conv2D | [[1 352 7 7]] | 272.333 | 100608 | 1743104 | 45393920 | GPU_0_bfc | 1642496 | 0 | 0 | 0 | void im2col4d_kernel<float, int>(im2col4d_params, cudnnConvolutionStruct, cudnnTensor4dStruct, float const*, float*, int) | 105.00 | 0 | 0.00 | 0.00 | 0.00 | 0.00 | 0.00 | true | 0;0;0;0;0 |
| 191 | InceptionV2/InceptionV2/Mixed_5c/Branch_0/Conv2d_0a_1x1/BatchNorm/batchnorm/mul | Conv2D | [[1 352 7 7]] | 272.333 | 100608 | 1743104 | 45393920 | GPU_0_bfc | 1642496 | 0 | 0 | 0 | void cudnn::detail::explicit_convolve_sgemm<float, int, 1024, 5, 5, 3, 3, 3, 0, true>(int, int, int, float const*, int, float const*, int, float*, kernel_conv_params, int, int, float, float, int, float*, float*) | 62.17 | 46154592 | 0.00 | 0.00 | 0.00 | 0.00 | 742.43 | true | 46154592;46154592;46154592;46154592;46154592 |
| 191 | InceptionV2/InceptionV2/Mixed_5c/Branch_0/Conv2d_0a_1x1/BatchNorm/batchnorm/mul | Conv2D | [[1 352 7 7]] | 272.333 | 100608 | 1743104 | 45393920 | GPU_0_bfc | 1642496 | 0 | 0 | 0 | void tensorflow::functor::ShuffleInTensor3Simple<float, 2, 1, 0, false>(int, float const*, tensorflow::functor::Dimension<3>, float*) | 20.00 | 0 | 0.00 | 0.00 | 0.00 | 0.00 | 0.00 | true | 0;0;0;0;0 |
| 192 | InceptionV2/InceptionV2/Mixed_5c/Branch_3/Conv2d_0b_1x1/BatchNorm/batchnorm/mul | Conv2D | [[1 128 7 7]] | 314.667 | 25088 | 818944 | 45124352 | GPU_0_bfc | 793856 | 0 | 0 | 0 | void im2col4d_kernel<float, int>(im2col4d_params, cudnnConvolutionStruct, cudnnTensor4dStruct, float const*, float*, int) | 165.00 | 0 | 0.00 | 0.00 | 0.00 | 0.00 | 0.00 | true | 0;0;0;0;0 |
| 192 | InceptionV2/InceptionV2/Mixed_5c/Branch_3/Conv2d_0b_1x1/BatchNorm/batchnorm/mul | Conv2D | [[1 128 7 7]] | 314.667 | 25088 | 818944 | 45124352 | GPU_0_bfc | 793856 | 0 | 0 | 0 | void cudnn::detail::explicit_convolve_sgemm<float, int, 128, 5, 5, 3, 3, 3, 0, true>(int, int, int, float const*, int, float const*, int, float*, kernel_conv_params, int, int, float, float, int, float*, float*) | 62.00 | 16783488 | 0.00 | 0.00 | 0.00 | 0.00 | 270.70 | true | 16783488;16783488;16783488;16783488;16783488 |
| 192 | InceptionV2/InceptionV2/Mixed_5c/Branch_3/Conv2d_0b_1x1/BatchNorm/batchnorm/mul | Conv2D | [[1 128 7 7]] | 314.667 | 25088 | 818944 | 45124352 | GPU_0_bfc | 793856 | 0 | 0 | 0 | void tensorflow::functor::ShuffleInTensor3Simple<float, 2, 1, 0, false>(int, float const*, tensorflow::functor::Dimension<3>, float*) | 10.00 | 0 | 0.00 | 0.00 | 0.00 | 0.00 | 0.00 | true | 0;0;0;0;0 |
| 193 | InceptionV2/InceptionV2/Mixed_5c/Branch_2/Conv2d_0a_1x1/BatchNorm/batchnorm/add_1 | Add | [[1 192 7 7]] | 29.333 | 37632 | 0 | 44923648 | GPU_0_bfc | 0 | 0 | 0 | 0 | void Eigen::internal::EigenMetaKernel<Eigen::TensorEvaluator<Eigen::TensorAssignOp<Eigen::TensorMap<Eigen::Tensor<float, 2, 1, int>, 16, Eigen::MakePointer>, Eigen::TensorCwiseBinaryOp<Eigen::internal::scalar_sum_op<float, float>, Eigen::TensorMap<Eigen::Tensor<float const, 2, 1, int>, 16, Eigen::MakePointer> const, Eigen::TensorBroadcastingOp<Eigen::array<long, 2ul> const, Eigen::TensorMap<Eigen::Tensor<float const, 2, 1, int>, 16, Eigen::MakePointer> const> const> const> const, Eigen::GpuDevice>, int>(Eigen::TensorEvaluator<Eigen::TensorAssignOp<Eigen::TensorMap<Eigen::Tensor<float, 2, 1, int>, 16, Eigen::MakePointer>, Eigen::TensorCwiseBinaryOp<Eigen::internal::scalar_sum_op<float, float>, Eigen::TensorMap<Eigen::Tensor<float const, 2, 1, int>, 16, Eigen::MakePointer> const, Eigen::TensorBroadcastingOp<Eigen::array<long, 2ul> const, Eigen::TensorMap<Eigen::Tensor<float const, 2, 1, int>, 16, Eigen::MakePointer> const> const> const> const, Eigen::GpuDevice>, int) | 5.00 | 9408 | 0.00 | 0.00 | 0.00 | 0.00 | 1.88 | true | 9408;9408;9408;9408;9408 |
| 194 | InceptionV2/InceptionV2/Mixed_5c/Branch_1/Conv2d_0a_1x1/BatchNorm/batchnorm/add_1 | Add | [[1 192 7 7]] | 20.333 | 37632 | 0 | 44923648 | GPU_0_bfc | 0 | 0 | 0 | 0 | void Eigen::internal::EigenMetaKernel<Eigen::TensorEvaluator<Eigen::TensorAssignOp<Eigen::TensorMap<Eigen::Tensor<float, 2, 1, int>, 16, Eigen::MakePointer>, Eigen::TensorCwiseBinaryOp<Eigen::internal::scalar_sum_op<float, float>, Eigen::TensorMap<Eigen::Tensor<float const, 2, 1, int>, 16, Eigen::MakePointer> const, Eigen::TensorBroadcastingOp<Eigen::array<long, 2ul> const, Eigen::TensorMap<Eigen::Tensor<float const, 2, 1, int>, 16, Eigen::MakePointer> const> const> const> const, Eigen::GpuDevice>, int>(Eigen::TensorEvaluator<Eigen::TensorAssignOp<Eigen::TensorMap<Eigen::Tensor<float, 2, 1, int>, 16, Eigen::MakePointer>, Eigen::TensorCwiseBinaryOp<Eigen::internal::scalar_sum_op<float, float>, Eigen::TensorMap<Eigen::Tensor<float const, 2, 1, int>, 16, Eigen::MakePointer> const, Eigen::TensorBroadcastingOp<Eigen::array<long, 2ul> const, Eigen::TensorMap<Eigen::Tensor<float const, 2, 1, int>, 16, Eigen::MakePointer> const> const> const> const, Eigen::GpuDevice>, int) | 5.00 | 9408 | 0.00 | 0.00 | 0.00 | 0.00 | 1.88 | true | 9408;9408;9408;9408;9408 |
| 195 | InceptionV2/InceptionV2/Mixed_5c/Branch_0/Conv2d_0a_1x1/BatchNorm/batchnorm/add_1 | Add | [[1 352 7 7]] | 20.333 | 100608 | 0 | 44923648 | GPU_0_bfc | 0 | 0 | 0 | 0 | void Eigen::internal::EigenMetaKernel<Eigen::TensorEvaluator<Eigen::TensorAssignOp<Eigen::TensorMap<Eigen::Tensor<float, 2, 1, int>, 16, Eigen::MakePointer>, Eigen::TensorCwiseBinaryOp<Eigen::internal::scalar_sum_op<float, float>, Eigen::TensorMap<Eigen::Tensor<float const, 2, 1, int>, 16, Eigen::MakePointer> const, Eigen::TensorBroadcastingOp<Eigen::array<long, 2ul> const, Eigen::TensorMap<Eigen::Tensor<float const, 2, 1, int>, 16, Eigen::MakePointer> const> const> const> const, Eigen::GpuDevice>, int>(Eigen::TensorEvaluator<Eigen::TensorAssignOp<Eigen::TensorMap<Eigen::Tensor<float, 2, 1, int>, 16, Eigen::MakePointer>, Eigen::TensorCwiseBinaryOp<Eigen::internal::scalar_sum_op<float, float>, Eigen::TensorMap<Eigen::Tensor<float const, 2, 1, int>, 16, Eigen::MakePointer> const, Eigen::TensorBroadcastingOp<Eigen::array<long, 2ul> const, Eigen::TensorMap<Eigen::Tensor<float const, 2, 1, int>, 16, Eigen::MakePointer> const> const> const> const, Eigen::GpuDevice>, int) | 5.00 | 17248 | 0.00 | 0.00 | 0.00 | 0.00 | 3.45 | true | 17248;17248;17248;17248;17248 |
| 196 | InceptionV2/InceptionV2/Mixed_5c/Branch_3/Conv2d_0b_1x1/BatchNorm/batchnorm/add_1 | Add | [[1 128 7 7]] | 20.667 | 25088 | 0 | 44923648 | GPU_0_bfc | 0 | 0 | 0 | 0 | void Eigen::internal::EigenMetaKernel<Eigen::TensorEvaluator<Eigen::TensorAssignOp<Eigen::TensorMap<Eigen::Tensor<float, 2, 1, int>, 16, Eigen::MakePointer>, Eigen::TensorCwiseBinaryOp<Eigen::internal::scalar_sum_op<float, float>, Eigen::TensorMap<Eigen::Tensor<float const, 2, 1, int>, 16, Eigen::MakePointer> const, Eigen::TensorBroadcastingOp<Eigen::array<long, 2ul> const, Eigen::TensorMap<Eigen::Tensor<float const, 2, 1, int>, 16, Eigen::MakePointer> const> const> const> const, Eigen::GpuDevice>, int>(Eigen::TensorEvaluator<Eigen::TensorAssignOp<Eigen::TensorMap<Eigen::Tensor<float, 2, 1, int>, 16, Eigen::MakePointer>, Eigen::TensorCwiseBinaryOp<Eigen::internal::scalar_sum_op<float, float>, Eigen::TensorMap<Eigen::Tensor<float const, 2, 1, int>, 16, Eigen::MakePointer> const, Eigen::TensorBroadcastingOp<Eigen::array<long, 2ul> const, Eigen::TensorMap<Eigen::Tensor<float const, 2, 1, int>, 16, Eigen::MakePointer> const> const> const> const, Eigen::GpuDevice>, int) | 5.00 | 6272 | 0.00 | 0.00 | 0.00 | 0.00 | 1.25 | true | 6272;6272;6272;6272;6272 |
| 197 | InceptionV2/InceptionV2/Mixed_5c/Branch_2/Conv2d_0a_1x1/Relu | Relu | [[1 192 7 7]] | 17 | 37632 | 0 | 44923648 | GPU_0_bfc | 0 | 0 | 0 | 0 | void Eigen::internal::EigenMetaKernel<Eigen::TensorEvaluator<Eigen::TensorAssignOp<Eigen::TensorMap<Eigen::Tensor<float, 1, 1, long>, 16, Eigen::MakePointer>, Eigen::TensorCwiseBinaryOp<Eigen::internal::scalar_max_op<float const, float const>, Eigen::TensorMap<Eigen::Tensor<float const, 1, 1, long>, 16, Eigen::MakePointer> const, Eigen::TensorCwiseNullaryOp<Eigen::internal::scalar_constant_op<float const>, Eigen::TensorMap<Eigen::Tensor<float const, 1, 1, long>, 16, Eigen::MakePointer> const> const> const> const, Eigen::GpuDevice>, long>(Eigen::TensorEvaluator<Eigen::TensorAssignOp<Eigen::TensorMap<Eigen::Tensor<float, 1, 1, long>, 16, Eigen::MakePointer>, Eigen::TensorCwiseBinaryOp<Eigen::internal::scalar_max_op<float const, float const>, Eigen::TensorMap<Eigen::Tensor<float const, 1, 1, long>, 16, Eigen::MakePointer> const, Eigen::TensorCwiseNullaryOp<Eigen::internal::scalar_constant_op<float const>, Eigen::TensorMap<Eigen::Tensor<float const, 1, 1, long>, 16, Eigen::MakePointer> const> const> const> const, Eigen::GpuDevice>, long) | 3.00 | 0 | 0.00 | 0.00 | 0.00 | 0.00 | 0.00 | true | 0;0;0;0;0 |
| 198 | InceptionV2/InceptionV2/Mixed_5c/Branch_1/Conv2d_0a_1x1/Relu | Relu | [[1 192 7 7]] | 19.333 | 37632 | 0 | 44923648 | GPU_0_bfc | 0 | 0 | 0 | 0 | void Eigen::internal::EigenMetaKernel<Eigen::TensorEvaluator<Eigen::TensorAssignOp<Eigen::TensorMap<Eigen::Tensor<float, 1, 1, long>, 16, Eigen::MakePointer>, Eigen::TensorCwiseBinaryOp<Eigen::internal::scalar_max_op<float const, float const>, Eigen::TensorMap<Eigen::Tensor<float const, 1, 1, long>, 16, Eigen::MakePointer> const, Eigen::TensorCwiseNullaryOp<Eigen::internal::scalar_constant_op<float const>, Eigen::TensorMap<Eigen::Tensor<float const, 1, 1, long>, 16, Eigen::MakePointer> const> const> const> const, Eigen::GpuDevice>, long>(Eigen::TensorEvaluator<Eigen::TensorAssignOp<Eigen::TensorMap<Eigen::Tensor<float, 1, 1, long>, 16, Eigen::MakePointer>, Eigen::TensorCwiseBinaryOp<Eigen::internal::scalar_max_op<float const, float const>, Eigen::TensorMap<Eigen::Tensor<float const, 1, 1, long>, 16, Eigen::MakePointer> const, Eigen::TensorCwiseNullaryOp<Eigen::internal::scalar_constant_op<float const>, Eigen::TensorMap<Eigen::Tensor<float const, 1, 1, long>, 16, Eigen::MakePointer> const> const> const> const, Eigen::GpuDevice>, long) | 3.00 | 0 | 0.00 | 0.00 | 0.00 | 0.00 | 0.00 | true | 0;0;0;0;0 |
| 199 | InceptionV2/InceptionV2/Mixed_5c/Branch_2/Conv2d_0b_3x3/BatchNorm/batchnorm/mul | Conv2D | [[1 224 7 7]] | 183.333 | 44032 | 5894144 | 44967680 | GPU_0_bfc | 5850112 | 0 | 0 | 0 | maxwell_scudnn_winograd_128x128_ldg1_ldg4_tile148n_nt | 60.00 | 46032896 | 0.00 | 0.00 | 0.00 | 0.00 | 767.21 | true | 46032896;46032896;46032896;46032896;46032896 |
| 199 | InceptionV2/InceptionV2/Mixed_5c/Branch_2/Conv2d_0b_3x3/BatchNorm/batchnorm/mul | Conv2D | [[1 224 7 7]] | 183.333 | 44032 | 5894144 | 44967680 | GPU_0_bfc | 5850112 | 0 | 0 | 0 | void tensorflow::functor::ShuffleInTensor3Simple<float, 2, 1, 0, false>(int, float const*, tensorflow::functor::Dimension<3>, float*) | 19.00 | 0 | 0.00 | 0.00 | 0.00 | 0.00 | 0.00 | true | 0;0;0;0;0 |
| 199 | InceptionV2/InceptionV2/Mixed_5c/Branch_2/Conv2d_0b_3x3/BatchNorm/batchnorm/mul | Conv2D | [[1 224 7 7]] | 183.333 | 44032 | 5894144 | 44967680 | GPU_0_bfc | 5850112 | 0 | 0 | 0 | void cudnn::winograd::generateWinogradTilesKernel<0, float, float>(cudnn::winograd::GenerateWinogradTilesParams<float, float>) | 10.00 | 2494464 | 0.00 | 0.00 | 0.00 | 0.00 | 249.45 | true | 2494464;2494464;2494464;2494464;2494464 |
| 200 | InceptionV2/InceptionV2/Mixed_5c/Branch_1/Conv2d_0b_3x3/BatchNorm/batchnorm/mul | Conv2D | [[1 320 7 7]] | 202 | 62720 | 8419840 | 44992768 | GPU_0_bfc | 8357120 | 0 | 0 | 0 | maxwell_scudnn_winograd_128x128_ldg1_ldg4_tile148n_nt | 60.50 | 65761280 | 0.00 | 0.00 | 0.00 | 0.00 | 1086.96 | true | 65761280;65761280;65761280;65761280;65761280 |
| 200 | InceptionV2/InceptionV2/Mixed_5c/Branch_1/Conv2d_0b_3x3/BatchNorm/batchnorm/mul | Conv2D | [[1 320 7 7]] | 202 | 62720 | 8419840 | 44992768 | GPU_0_bfc | 8357120 | 0 | 0 | 0 | void tensorflow::functor::ShuffleInTensor3Simple<float, 2, 1, 0, false>(int, float const*, tensorflow::functor::Dimension<3>, float*) | 27.00 | 0 | 0.00 | 0.00 | 0.00 | 0.00 | 0.00 | true | 0;0;0;0;0 |
| 200 | InceptionV2/InceptionV2/Mixed_5c/Branch_1/Conv2d_0b_3x3/BatchNorm/batchnorm/mul | Conv2D | [[1 320 7 7]] | 202 | 62720 | 8419840 | 44992768 | GPU_0_bfc | 8357120 | 0 | 0 | 0 | void cudnn::winograd::generateWinogradTilesKernel<0, float, float>(cudnn::winograd::GenerateWinogradTilesParams<float, float>) | 14.00 | 3563520 | 0.00 | 0.00 | 0.00 | 0.00 | 254.54 | true | 3563520;3563520;3563520;3563520;3563520 |
| 201 | InceptionV2/InceptionV2/Mixed_5c/Branch_2/Conv2d_0b_3x3/BatchNorm/batchnorm/add_1 | Add | [[1 224 7 7]] | 24.667 | 44032 | 0 | 44955136 | GPU_0_bfc | 0 | 0 | 0 | 0 | void Eigen::internal::EigenMetaKernel<Eigen::TensorEvaluator<Eigen::TensorAssignOp<Eigen::TensorMap<Eigen::Tensor<float, 2, 1, int>, 16, Eigen::MakePointer>, Eigen::TensorCwiseBinaryOp<Eigen::internal::scalar_sum_op<float, float>, Eigen::TensorMap<Eigen::Tensor<float const, 2, 1, int>, 16, Eigen::MakePointer> const, Eigen::TensorBroadcastingOp<Eigen::array<long, 2ul> const, Eigen::TensorMap<Eigen::Tensor<float const, 2, 1, int>, 16, Eigen::MakePointer> const> const> const> const, Eigen::GpuDevice>, int>(Eigen::TensorEvaluator<Eigen::TensorAssignOp<Eigen::TensorMap<Eigen::Tensor<float, 2, 1, int>, 16, Eigen::MakePointer>, Eigen::TensorCwiseBinaryOp<Eigen::internal::scalar_sum_op<float, float>, Eigen::TensorMap<Eigen::Tensor<float const, 2, 1, int>, 16, Eigen::MakePointer> const, Eigen::TensorBroadcastingOp<Eigen::array<long, 2ul> const, Eigen::TensorMap<Eigen::Tensor<float const, 2, 1, int>, 16, Eigen::MakePointer> const> const> const> const, Eigen::GpuDevice>, int) | 5.00 | 10976 | 0.00 | 0.00 | 0.00 | 0.00 | 2.20 | true | 10976;10976;10976;10976;10976 |
| 202 | InceptionV2/InceptionV2/Mixed_5c/Branch_1/Conv2d_0b_3x3/BatchNorm/batchnorm/add_1 | Add | [[1 320 7 7]] | 20.333 | 62720 | 0 | 44955136 | GPU_0_bfc | 0 | 0 | 0 | 0 | void Eigen::internal::EigenMetaKernel<Eigen::TensorEvaluator<Eigen::TensorAssignOp<Eigen::TensorMap<Eigen::Tensor<float, 2, 1, int>, 16, Eigen::MakePointer>, Eigen::TensorCwiseBinaryOp<Eigen::internal::scalar_sum_op<float, float>, Eigen::TensorMap<Eigen::Tensor<float const, 2, 1, int>, 16, Eigen::MakePointer> const, Eigen::TensorBroadcastingOp<Eigen::array<long, 2ul> const, Eigen::TensorMap<Eigen::Tensor<float const, 2, 1, int>, 16, Eigen::MakePointer> const> const> const> const, Eigen::GpuDevice>, int>(Eigen::TensorEvaluator<Eigen::TensorAssignOp<Eigen::TensorMap<Eigen::Tensor<float, 2, 1, int>, 16, Eigen::MakePointer>, Eigen::TensorCwiseBinaryOp<Eigen::internal::scalar_sum_op<float, float>, Eigen::TensorMap<Eigen::Tensor<float const, 2, 1, int>, 16, Eigen::MakePointer> const, Eigen::TensorBroadcastingOp<Eigen::array<long, 2ul> const, Eigen::TensorMap<Eigen::Tensor<float const, 2, 1, int>, 16, Eigen::MakePointer> const> const> const> const, Eigen::GpuDevice>, int) | 5.00 | 15680 | 0.00 | 0.00 | 0.00 | 0.00 | 3.14 | true | 15680;15680;15680;15680;15680 |
| 203 | InceptionV2/InceptionV2/Mixed_5c/Branch_2/Conv2d_0b_3x3/Relu | Relu | [[1 224 7 7]] | 17.667 | 44032 | 0 | 44955136 | GPU_0_bfc | 0 | 0 | 0 | 0 | void Eigen::internal::EigenMetaKernel<Eigen::TensorEvaluator<Eigen::TensorAssignOp<Eigen::TensorMap<Eigen::Tensor<float, 1, 1, long>, 16, Eigen::MakePointer>, Eigen::TensorCwiseBinaryOp<Eigen::internal::scalar_max_op<float const, float const>, Eigen::TensorMap<Eigen::Tensor<float const, 1, 1, long>, 16, Eigen::MakePointer> const, Eigen::TensorCwiseNullaryOp<Eigen::internal::scalar_constant_op<float const>, Eigen::TensorMap<Eigen::Tensor<float const, 1, 1, long>, 16, Eigen::MakePointer> const> const> const> const, Eigen::GpuDevice>, long>(Eigen::TensorEvaluator<Eigen::TensorAssignOp<Eigen::TensorMap<Eigen::Tensor<float, 1, 1, long>, 16, Eigen::MakePointer>, Eigen::TensorCwiseBinaryOp<Eigen::internal::scalar_max_op<float const, float const>, Eigen::TensorMap<Eigen::Tensor<float const, 1, 1, long>, 16, Eigen::MakePointer> const, Eigen::TensorCwiseNullaryOp<Eigen::internal::scalar_constant_op<float const>, Eigen::TensorMap<Eigen::Tensor<float const, 1, 1, long>, 16, Eigen::MakePointer> const> const> const> const, Eigen::GpuDevice>, long) | 3.00 | 0 | 0.00 | 0.00 | 0.00 | 0.00 | 0.00 | true | 0;0;0;0;0 |
| 204 | InceptionV2/InceptionV2/Mixed_5c/Branch_2/Conv2d_0c_3x3/BatchNorm/batchnorm/mul | Conv2D | [[1 224 7 7]] | 197 | 75264 | 6900224 | 45030400 | GPU_0_bfc | 6824960 | 0 | 0 | 0 | maxwell_scudnn_winograd_128x128_ldg1_ldg4_tile148n_nt | 68.83 | 53659648 | 0.00 | 0.00 | 0.00 | 0.00 | 779.56 | true | 53659648;53659648;53659648;53659648;53659648 |
| 204 | InceptionV2/InceptionV2/Mixed_5c/Branch_2/Conv2d_0c_3x3/BatchNorm/batchnorm/mul | Conv2D | [[1 224 7 7]] | 197 | 75264 | 6900224 | 45030400 | GPU_0_bfc | 6824960 | 0 | 0 | 0 | void tensorflow::functor::ShuffleInTensor3Simple<float, 2, 1, 0, false>(int, float const*, tensorflow::functor::Dimension<3>, float*) | 23.00 | 0 | 0.00 | 0.00 | 0.00 | 0.00 | 0.00 | true | 0;0;0;0;0 |
| 204 | InceptionV2/InceptionV2/Mixed_5c/Branch_2/Conv2d_0c_3x3/BatchNorm/batchnorm/mul | Conv2D | [[1 224 7 7]] | 197 | 75264 | 6900224 | 45030400 | GPU_0_bfc | 6824960 | 0 | 0 | 0 | void cudnn::winograd::generateWinogradTilesKernel<0, float, float>(cudnn::winograd::GenerateWinogradTilesParams<float, float>) | 11.00 | 2910208 | 0.00 | 0.00 | 0.00 | 0.00 | 264.56 | true | 2910208;2910208;2910208;2910208;2910208 |
| 205 | InceptionV2/InceptionV2/Mixed_5c/Branch_2/Conv2d_0c_3x3/BatchNorm/batchnorm/add_1 | Add | [[1 224 7 7]] | 27.333 | 75264 | 0 | 44986368 | GPU_0_bfc | 0 | 0 | 0 | 0 | void Eigen::internal::EigenMetaKernel<Eigen::TensorEvaluator<Eigen::TensorAssignOp<Eigen::TensorMap<Eigen::Tensor<float, 2, 1, int>, 16, Eigen::MakePointer>, Eigen::TensorCwiseBinaryOp<Eigen::internal::scalar_sum_op<float, float>, Eigen::TensorMap<Eigen::Tensor<float const, 2, 1, int>, 16, Eigen::MakePointer> const, Eigen::TensorBroadcastingOp<Eigen::array<long, 2ul> const, Eigen::TensorMap<Eigen::Tensor<float const, 2, 1, int>, 16, Eigen::MakePointer> const> const> const> const, Eigen::GpuDevice>, int>(Eigen::TensorEvaluator<Eigen::TensorAssignOp<Eigen::TensorMap<Eigen::Tensor<float, 2, 1, int>, 16, Eigen::MakePointer>, Eigen::TensorCwiseBinaryOp<Eigen::internal::scalar_sum_op<float, float>, Eigen::TensorMap<Eigen::Tensor<float const, 2, 1, int>, 16, Eigen::MakePointer> const, Eigen::TensorBroadcastingOp<Eigen::array<long, 2ul> const, Eigen::TensorMap<Eigen::Tensor<float const, 2, 1, int>, 16, Eigen::MakePointer> const> const> const> const, Eigen::GpuDevice>, int) | 5.00 | 10976 | 0.00 | 0.00 | 0.00 | 0.00 | 2.20 | true | 10976;10976;10976;10976;10976 |
| 207 | InceptionV2/InceptionV2/Mixed_5c/Branch_0/Conv2d_0a_1x1/Relu | Relu | [[1 1024 7 7]] | 20.667 | 200704 | 0 | 44923392 | GPU_0_bfc | 0 | 0 | 0 | 0 | void Eigen::internal::EigenMetaKernel<Eigen::TensorEvaluator<Eigen::TensorAssignOp<Eigen::TensorMap<Eigen::Tensor<float, 1, 1, long>, 16, Eigen::MakePointer>, Eigen::TensorCwiseBinaryOp<Eigen::internal::scalar_max_op<float const, float const>, Eigen::TensorMap<Eigen::Tensor<float const, 1, 1, long>, 16, Eigen::MakePointer> const, Eigen::TensorCwiseNullaryOp<Eigen::internal::scalar_constant_op<float const>, Eigen::TensorMap<Eigen::Tensor<float const, 1, 1, long>, 16, Eigen::MakePointer> const> const> const> const, Eigen::GpuDevice>, long>(Eigen::TensorEvaluator<Eigen::TensorAssignOp<Eigen::TensorMap<Eigen::Tensor<float, 1, 1, long>, 16, Eigen::MakePointer>, Eigen::TensorCwiseBinaryOp<Eigen::internal::scalar_max_op<float const, float const>, Eigen::TensorMap<Eigen::Tensor<float const, 1, 1, long>, 16, Eigen::MakePointer> const, Eigen::TensorCwiseNullaryOp<Eigen::internal::scalar_constant_op<float const>, Eigen::TensorMap<Eigen::Tensor<float const, 1, 1, long>, 16, Eigen::MakePointer> const> const> const> const, Eigen::GpuDevice>, long) | 3.00 | 0 | 0.00 | 0.00 | 0.00 | 0.00 | 0.00 | true | 0;0;0;0;0 |
| 208 | InceptionV2/Logits/AvgPool_1a_7x7/AvgPool | AvgPool | [[1 1024 1 1]] | 40.667 | 4096 | 4096 | 44927488 | GPU_0_bfc | 0 | 0 | 0 | 0 | void cudnn::detail::pooling_fw_4d_kernel<float, float, cudnn::detail::averpooling_func<float>, 2, false>(cudnnTensorStruct, float const*, cudnnTensorStruct, float*, cudnnPoolingStruct, float, float, int, cudnn::reduced_divisor, cudnn::reduced_divisor) | 11.00 | 72074 | 0.00 | 0.00 | 0.00 | 0.00 | 6.55 | true | 72074;72074;72074;72074;72074 |
| 209 | InceptionV2/Logits/Conv2d_1c_1x1/convolution | Conv2D | [[1 1001 1 1]] | 297.333 | 4096 | 4104192 | 44730880 | GPU_0_bfc | 4100096 | 0 | 0 | 0 | void cudnn::detail::implicit_convolve_sgemm<float, float, 128, 5, 5, 3, 3, 3, 1, true, false, true>(int, int, int, float const*, int, float*, float*, kernel_conv_params, int, float, float, int, float*, float*, int, int) | 173.17 | 67109865 | 0.00 | 0.00 | 0.00 | 0.00 | 387.54 | true | 67109865;67109865;67109865;67109865;67109865 |
| 209 | InceptionV2/Logits/Conv2d_1c_1x1/convolution | Conv2D | [[1 1001 1 1]] | 297.333 | 4096 | 4104192 | 44730880 | GPU_0_bfc | 4100096 | 0 | 0 | 0 | void tensorflow::functor::ShuffleInTensor3Simple<float, 2, 1, 0, false>(int, float const*, tensorflow::functor::Dimension<3>, float*) | 53.83 | 0 | 0.00 | 0.00 | 0.00 | 0.00 | 0.00 | true | 0;0;0;0;0 |
| 210 | InceptionV2/Logits/Conv2d_1c_1x1/BiasAdd | BiasAdd | [[1 1001 1 1]] | 24 | 4096 | 0 | 44726784 | GPU_0_bfc | 0 | 0 | 0 | 0 | void tensorflow::BiasNCHWKernel<float>(int, float const*, float const*, float*, int, int) | 4.00 | 1001 | 0.00 | 0.00 | 0.00 | 0.00 | 0.25 | true | 1001;1001;1001;1001;1001 |
| 214 | InceptionV2/Predictions/Softmax | Softmax | [[1 1001]] | 65 | 4096 | 8192 | 44726784 | GPU_0_bfc | 8192 | 0 | 0 | 0 | void tensorflow::functor::RowReduceKernel<cub::TransformInputIterator<float, tensorflow::(anonymous namespace)::SubtractAndExpFunctor<float, float>, cub::CountingInputIterator<int, long>, long>, float*, cub::Sum>(cub::TransformInputIterator<float, tensorflow::(anonymous namespace)::SubtractAndExpFunctor<float, float>, cub::CountingInputIterator<int, long>, long>, float*, int, int, cub::Sum, std::iterator_traits<cub::TransformInputIterator<float, tensorflow::(anonymous namespace)::SubtractAndExpFunctor<float, float>, cub::CountingInputIterator<int, long>, long> >::value_type) | 12.00 | 10431 | 0.00 | 0.00 | 0.00 | 0.00 | 0.87 | true | 10431;10431;10431;10431;10431 |
| 214 | InceptionV2/Predictions/Softmax | Softmax | [[1 1001]] | 65 | 4096 | 8192 | 44726784 | GPU_0_bfc | 8192 | 0 | 0 | 0 | void tensorflow::functor::RowReduceKernel<float const*, float*, cub::Max>(float const*, float*, int, int, cub::Max, std::iterator_traits<float const*>::value_type) | 6.00 | 0 | 0.00 | 0.00 | 0.00 | 0.00 | 0.00 | true | 0;0;0;0;0 |
| 214 | InceptionV2/Predictions/Softmax | Softmax | [[1 1001]] | 65 | 4096 | 8192 | 44726784 | GPU_0_bfc | 8192 | 0 | 0 | 0 | void tensorflow::(anonymous namespace)::GenerateNormalizedProb<float, float>(float const*, float const*, float const*, float*, int, int, bool) | 4.00 | 24024 | 0.00 | 0.00 | 0.00 | 0.00 | 6.01 | true | 24024;24024;24024;24024;24024 |
Showing 1 to 346 of 346 entries