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 |
|---|
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 |
|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|
| 0 | resnext0_conv0_fwd | Convolution | [1,3,224,224] | 20053 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | cudnn::gemm::computeOffsetsKernel(cudnn::gemm::ComputeOffsetsParams) | 0.00 | 0 | 0.00 | 0.00 | 0.00 | 0.00 | NaN | true | |
| 0 | resnext0_conv0_fwd | Convolution | [1,3,224,224] | 20053 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | volta_scudnn_128x64_relu_medium_nn_v1 | 0.00 | 0 | 0.00 | 0.00 | 0.00 | 0.00 | NaN | true | |
| 1 | resnext0_batchnorm0_fwd | BatchNorm | [1,64,112,112] | 196.667 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | void cudnn::detail::bn_fw_inf_1C11_kernel_NCHW<float, float, true, 1>(float, float, cudnnTensorStruct, float const*, cudnnTensorStruct, float*, cudnnTensorStruct, float const*, float const*, float const*, float const*, float) | 0.00 | 0 | 0.00 | 0.00 | 0.00 | 0.00 | NaN | true | |
| 2 | resnext0_relu0_fwd | Activation | [1,64,112,112] | 161.333 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | void op_generic_tensor_kernel<2, float, float, float, 256, (cudnnGenericOp_t)8, (cudnnNanPropagation_t)0, (cudnnDimOrder_t)0, 1>(cudnnTensorStruct, float*, cudnnTensorStruct, float const*, cudnnTensorStruct, float const*, float, float, float, float, dimArray, reducedDivisorArray, bool) | 0.00 | 0 | 0.00 | 0.00 | 0.00 | 0.00 | NaN | true | |
| 3 | resnext0_pool0_fwd | Pooling | [1,64,112,112] | 2855.333 | 0 | 0 | 0 | 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) | 0.00 | 0 | 0.00 | 0.00 | 0.00 | 0.00 | NaN | true | |
| 4 | resnext0_stage1_conv0_fwd | Convolution | [1,64,56,56] | 3170 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | cudnn::gemm::computeOffsetsKernel(cudnn::gemm::ComputeOffsetsParams) | 0.00 | 0 | 0.00 | 0.00 | 0.00 | 0.00 | NaN | true | |
| 4 | resnext0_stage1_conv0_fwd | Convolution | [1,64,56,56] | 3170 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | volta_scudnn_128x64_relu_interior_nn_v1 | 0.00 | 0 | 0.00 | 0.00 | 0.00 | 0.00 | NaN | true | |
| 5 | resnext0_stage1_batchnorm0_fwd | BatchNorm | [1,128,56,56] | 124.667 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | void cudnn::detail::bn_fw_inf_1C11_kernel_NCHW<float, float, true, 1>(float, float, cudnnTensorStruct, float const*, cudnnTensorStruct, float*, cudnnTensorStruct, float const*, float const*, float const*, float const*, float) | 0.00 | 0 | 0.00 | 0.00 | 0.00 | 0.00 | NaN | true | |
| 6 | resnext0_stage1_relu0_fwd | Activation | [1,128,56,56] | 76 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | void op_generic_tensor_kernel<2, float, float, float, 256, (cudnnGenericOp_t)8, (cudnnNanPropagation_t)0, (cudnnDimOrder_t)0, 1>(cudnnTensorStruct, float*, cudnnTensorStruct, float const*, cudnnTensorStruct, float const*, float, float, float, float, dimArray, reducedDivisorArray, bool) | 0.00 | 0 | 0.00 | 0.00 | 0.00 | 0.00 | NaN | true | |
| 7 | resnext0_stage1_conv1_fwd | Convolution | [1,128,56,56] | 11195.333 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | void conv2d_grouped_direct_kernel<float, float, float, float, float, true, false, 0, 4, 3>(cudnnTensorStruct, float const*, cudnnFilterStruct, float const*, cudnnConvolutionStruct, cudnnTensorStruct, float*, float, float, cudnn::reduced_divisor, cudnn::reduced_divisor, cudnn::reduced_divisor, cudnn::reduced_divisor, cudnn::reduced_divisor, int, float const*, float const*, cudnnActivationStruct) | 0.00 | 0 | 0.00 | 0.00 | 0.00 | 0.00 | NaN | true | |
| 8 | resnext0_stage1_batchnorm1_fwd | BatchNorm | [1,128,56,56] | 131 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | void cudnn::detail::bn_fw_inf_1C11_kernel_NCHW<float, float, true, 1>(float, float, cudnnTensorStruct, float const*, cudnnTensorStruct, float*, cudnnTensorStruct, float const*, float const*, float const*, float const*, float) | 0.00 | 0 | 0.00 | 0.00 | 0.00 | 0.00 | NaN | true | |
| 9 | resnext0_stage1_relu1_fwd | Activation | [1,128,56,56] | 78 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | void op_generic_tensor_kernel<2, float, float, float, 256, (cudnnGenericOp_t)8, (cudnnNanPropagation_t)0, (cudnnDimOrder_t)0, 1>(cudnnTensorStruct, float*, cudnnTensorStruct, float const*, cudnnTensorStruct, float const*, float, float, float, float, dimArray, reducedDivisorArray, bool) | 0.00 | 0 | 0.00 | 0.00 | 0.00 | 0.00 | NaN | true | |
| 10 | resnext0_stage1_conv2_fwd | Convolution | [1,128,56,56] | 10943.333 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | cudnn::gemm::computeOffsetsKernel(cudnn::gemm::ComputeOffsetsParams) | 0.00 | 0 | 0.00 | 0.00 | 0.00 | 0.00 | NaN | true | |
| 10 | resnext0_stage1_conv2_fwd | Convolution | [1,128,56,56] | 10943.333 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | volta_scudnn_128x128_relu_interior_nn_v1 | 0.00 | 0 | 0.00 | 0.00 | 0.00 | 0.00 | NaN | true | |
| 11 | resnext0_stage1_batchnorm2_fwd | BatchNorm | [1,256,56,56] | 160.333 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | void cudnn::detail::bn_fw_inf_1C11_kernel_NCHW<float, float, true, 1>(float, float, cudnnTensorStruct, float const*, cudnnTensorStruct, float*, cudnnTensorStruct, float const*, float const*, float const*, float const*, float) | 0.00 | 0 | 0.00 | 0.00 | 0.00 | 0.00 | NaN | true | |
| 12 | resnext0_stage1_conv3_fwd | Convolution | [1,64,56,56] | 5874.667 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | cudnn::gemm::computeOffsetsKernel(cudnn::gemm::ComputeOffsetsParams) | 0.00 | 0 | 0.00 | 0.00 | 0.00 | 0.00 | NaN | true | |
| 12 | resnext0_stage1_conv3_fwd | Convolution | [1,64,56,56] | 5874.667 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | volta_scudnn_128x64_relu_interior_nn_v1 | 0.00 | 0 | 0.00 | 0.00 | 0.00 | 0.00 | NaN | true | |
| 13 | resnext0_stage1_batchnorm3_fwd | BatchNorm | [1,256,56,56] | 160.333 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | void cudnn::detail::bn_fw_inf_1C11_kernel_NCHW<float, float, true, 1>(float, float, cudnnTensorStruct, float const*, cudnnTensorStruct, float*, cudnnTensorStruct, float const*, float const*, float const*, float const*, float) | 0.00 | 0 | 0.00 | 0.00 | 0.00 | 0.00 | NaN | true | |
| 14 | add_resnext0_stage1_activation0 | add_relu | [1,256,56,56] | 195 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | void mxnet::op::mxnet_op::mxnet_generic_kernel<mxnet::op::AddReluKernel, float*, float*, float*, mxnet::OpReqType>(int, float*, float*, float*, mxnet::OpReqType) | 0.00 | 0 | 0.00 | 0.00 | 0.00 | 0.00 | NaN | true | |
| 15 | resnext0_stage1_conv4_fwd | Convolution | [1,256,56,56] | 11123.333 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | cudnn::gemm::computeOffsetsKernel(cudnn::gemm::ComputeOffsetsParams) | 0.00 | 0 | 0.00 | 0.00 | 0.00 | 0.00 | NaN | true | |
| 15 | resnext0_stage1_conv4_fwd | Convolution | [1,256,56,56] | 11123.333 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | volta_scudnn_128x64_relu_interior_nn_v1 | 0.00 | 0 | 0.00 | 0.00 | 0.00 | 0.00 | NaN | true | |
| 16 | resnext0_stage1_batchnorm4_fwd | BatchNorm | [1,128,56,56] | 114.333 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | void cudnn::detail::bn_fw_inf_1C11_kernel_NCHW<float, float, true, 1>(float, float, cudnnTensorStruct, float const*, cudnnTensorStruct, float*, cudnnTensorStruct, float const*, float const*, float const*, float const*, float) | 0.00 | 0 | 0.00 | 0.00 | 0.00 | 0.00 | NaN | true | |
| 17 | resnext0_stage1_relu2_fwd | Activation | [1,128,56,56] | 72.333 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | void op_generic_tensor_kernel<2, float, float, float, 256, (cudnnGenericOp_t)8, (cudnnNanPropagation_t)0, (cudnnDimOrder_t)0, 1>(cudnnTensorStruct, float*, cudnnTensorStruct, float const*, cudnnTensorStruct, float const*, float, float, float, float, dimArray, reducedDivisorArray, bool) | 0.00 | 0 | 0.00 | 0.00 | 0.00 | 0.00 | NaN | true | |
| 18 | resnext0_stage1_conv5_fwd | Convolution | [1,128,56,56] | 10960.333 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | void conv2d_grouped_direct_kernel<float, float, float, float, float, true, false, 0, 4, 3>(cudnnTensorStruct, float const*, cudnnFilterStruct, float const*, cudnnConvolutionStruct, cudnnTensorStruct, float*, float, float, cudnn::reduced_divisor, cudnn::reduced_divisor, cudnn::reduced_divisor, cudnn::reduced_divisor, cudnn::reduced_divisor, int, float const*, float const*, cudnnActivationStruct) | 0.00 | 0 | 0.00 | 0.00 | 0.00 | 0.00 | NaN | true | |
| 19 | resnext0_stage1_batchnorm5_fwd | BatchNorm | [1,128,56,56] | 114.667 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | void cudnn::detail::bn_fw_inf_1C11_kernel_NCHW<float, float, true, 1>(float, float, cudnnTensorStruct, float const*, cudnnTensorStruct, float*, cudnnTensorStruct, float const*, float const*, float const*, float const*, float) | 0.00 | 0 | 0.00 | 0.00 | 0.00 | 0.00 | NaN | true | |
| 20 | resnext0_stage1_relu3_fwd | Activation | [1,128,56,56] | 76.333 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | void op_generic_tensor_kernel<2, float, float, float, 256, (cudnnGenericOp_t)8, (cudnnNanPropagation_t)0, (cudnnDimOrder_t)0, 1>(cudnnTensorStruct, float*, cudnnTensorStruct, float const*, cudnnTensorStruct, float const*, float, float, float, float, dimArray, reducedDivisorArray, bool) | 0.00 | 0 | 0.00 | 0.00 | 0.00 | 0.00 | NaN | true | |
| 21 | resnext0_stage1_conv6_fwd | Convolution | [1,128,56,56] | 10835 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | cudnn::gemm::computeOffsetsKernel(cudnn::gemm::ComputeOffsetsParams) | 0.00 | 0 | 0.00 | 0.00 | 0.00 | 0.00 | NaN | true | |
| 21 | resnext0_stage1_conv6_fwd | Convolution | [1,128,56,56] | 10835 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | volta_scudnn_128x128_relu_interior_nn_v1 | 0.00 | 0 | 0.00 | 0.00 | 0.00 | 0.00 | NaN | true | |
| 22 | resnext0_stage1_batchnorm6_fwd | BatchNorm | [1,256,56,56] | 153 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | void cudnn::detail::bn_fw_inf_1C11_kernel_NCHW<float, float, true, 1>(float, float, cudnnTensorStruct, float const*, cudnnTensorStruct, float*, cudnnTensorStruct, float const*, float const*, float const*, float const*, float) | 0.00 | 0 | 0.00 | 0.00 | 0.00 | 0.00 | NaN | true | |
| 23 | add_resnext0_stage1_activation1 | add_relu | [1,256,56,56] | 200.333 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | void mxnet::op::mxnet_op::mxnet_generic_kernel<mxnet::op::AddReluKernel, float*, float*, float*, mxnet::OpReqType>(int, float*, float*, float*, mxnet::OpReqType) | 0.00 | 0 | 0.00 | 0.00 | 0.00 | 0.00 | NaN | true | |
| 24 | resnext0_stage1_conv7_fwd | Convolution | [1,256,56,56] | 11024 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | cudnn::gemm::computeOffsetsKernel(cudnn::gemm::ComputeOffsetsParams) | 0.00 | 0 | 0.00 | 0.00 | 0.00 | 0.00 | NaN | true | |
| 24 | resnext0_stage1_conv7_fwd | Convolution | [1,256,56,56] | 11024 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | volta_scudnn_128x64_relu_interior_nn_v1 | 0.00 | 0 | 0.00 | 0.00 | 0.00 | 0.00 | NaN | true | |
| 25 | resnext0_stage1_batchnorm7_fwd | BatchNorm | [1,128,56,56] | 119 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | void cudnn::detail::bn_fw_inf_1C11_kernel_NCHW<float, float, true, 1>(float, float, cudnnTensorStruct, float const*, cudnnTensorStruct, float*, cudnnTensorStruct, float const*, float const*, float const*, float const*, float) | 0.00 | 0 | 0.00 | 0.00 | 0.00 | 0.00 | NaN | true | |
| 26 | resnext0_stage1_relu4_fwd | Activation | [1,128,56,56] | 75.333 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | void op_generic_tensor_kernel<2, float, float, float, 256, (cudnnGenericOp_t)8, (cudnnNanPropagation_t)0, (cudnnDimOrder_t)0, 1>(cudnnTensorStruct, float*, cudnnTensorStruct, float const*, cudnnTensorStruct, float const*, float, float, float, float, dimArray, reducedDivisorArray, bool) | 0.00 | 0 | 0.00 | 0.00 | 0.00 | 0.00 | NaN | true | |
| 27 | resnext0_stage1_conv8_fwd | Convolution | [1,128,56,56] | 10958 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | void conv2d_grouped_direct_kernel<float, float, float, float, float, true, false, 0, 4, 3>(cudnnTensorStruct, float const*, cudnnFilterStruct, float const*, cudnnConvolutionStruct, cudnnTensorStruct, float*, float, float, cudnn::reduced_divisor, cudnn::reduced_divisor, cudnn::reduced_divisor, cudnn::reduced_divisor, cudnn::reduced_divisor, int, float const*, float const*, cudnnActivationStruct) | 0.00 | 0 | 0.00 | 0.00 | 0.00 | 0.00 | NaN | true | |
| 28 | resnext0_stage1_batchnorm8_fwd | BatchNorm | [1,128,56,56] | 119 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | void cudnn::detail::bn_fw_inf_1C11_kernel_NCHW<float, float, true, 1>(float, float, cudnnTensorStruct, float const*, cudnnTensorStruct, float*, cudnnTensorStruct, float const*, float const*, float const*, float const*, float) | 0.00 | 0 | 0.00 | 0.00 | 0.00 | 0.00 | NaN | true | |
| 29 | resnext0_stage1_relu5_fwd | Activation | [1,128,56,56] | 76 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | void op_generic_tensor_kernel<2, float, float, float, 256, (cudnnGenericOp_t)8, (cudnnNanPropagation_t)0, (cudnnDimOrder_t)0, 1>(cudnnTensorStruct, float*, cudnnTensorStruct, float const*, cudnnTensorStruct, float const*, float, float, float, float, dimArray, reducedDivisorArray, bool) | 0.00 | 0 | 0.00 | 0.00 | 0.00 | 0.00 | NaN | true | |
| 30 | resnext0_stage1_conv9_fwd | Convolution | [1,128,56,56] | 11030 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | cudnn::gemm::computeOffsetsKernel(cudnn::gemm::ComputeOffsetsParams) | 0.00 | 0 | 0.00 | 0.00 | 0.00 | 0.00 | NaN | true | |
| 30 | resnext0_stage1_conv9_fwd | Convolution | [1,128,56,56] | 11030 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | volta_scudnn_128x128_relu_interior_nn_v1 | 0.00 | 0 | 0.00 | 0.00 | 0.00 | 0.00 | NaN | true | |
| 31 | resnext0_stage1_batchnorm9_fwd | BatchNorm | [1,256,56,56] | 168.667 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | void cudnn::detail::bn_fw_inf_1C11_kernel_NCHW<float, float, true, 1>(float, float, cudnnTensorStruct, float const*, cudnnTensorStruct, float*, cudnnTensorStruct, float const*, float const*, float const*, float const*, float) | 0.00 | 0 | 0.00 | 0.00 | 0.00 | 0.00 | NaN | true | |
| 32 | add_resnext0_stage1_activation2 | add_relu | [1,256,56,56] | 207 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | void mxnet::op::mxnet_op::mxnet_generic_kernel<mxnet::op::AddReluKernel, float*, float*, float*, mxnet::OpReqType>(int, float*, float*, float*, mxnet::OpReqType) | 0.00 | 0 | 0.00 | 0.00 | 0.00 | 0.00 | NaN | true | |
| 33 | resnext0_stage2_conv0_fwd | Convolution | [1,256,56,56] | 20302.667 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | cudnn::gemm::computeOffsetsKernel(cudnn::gemm::ComputeOffsetsParams) | 0.00 | 0 | 0.00 | 0.00 | 0.00 | 0.00 | NaN | true | |
| 33 | resnext0_stage2_conv0_fwd | Convolution | [1,256,56,56] | 20302.667 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | volta_scudnn_128x128_relu_interior_nn_v1 | 0.00 | 0 | 0.00 | 0.00 | 0.00 | 0.00 | NaN | true | |
| 34 | resnext0_stage2_batchnorm0_fwd | BatchNorm | [1,256,56,56] | 155.667 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | void cudnn::detail::bn_fw_inf_1C11_kernel_NCHW<float, float, true, 1>(float, float, cudnnTensorStruct, float const*, cudnnTensorStruct, float*, cudnnTensorStruct, float const*, float const*, float const*, float const*, float) | 0.00 | 0 | 0.00 | 0.00 | 0.00 | 0.00 | NaN | true | |
| 35 | resnext0_stage2_relu0_fwd | Activation | [1,256,56,56] | 146.333 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | void op_generic_tensor_kernel<2, float, float, float, 256, (cudnnGenericOp_t)8, (cudnnNanPropagation_t)0, (cudnnDimOrder_t)0, 1>(cudnnTensorStruct, float*, cudnnTensorStruct, float const*, cudnnTensorStruct, float const*, float, float, float, float, dimArray, reducedDivisorArray, bool) | 0.00 | 0 | 0.00 | 0.00 | 0.00 | 0.00 | NaN | true | |
| 36 | resnext0_stage2_conv1_fwd | Convolution | [1,256,56,56] | 6171 | 0 | 0 | 0 | 0 | 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) | 0.00 | 0 | 0.00 | 0.00 | 0.00 | 0.00 | NaN | true | |
| 36 | resnext0_stage2_conv1_fwd | Convolution | [1,256,56,56] | 6171 | 0 | 0 | 0 | 0 | 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) | 0.00 | 0 | 0.00 | 0.00 | 0.00 | 0.00 | NaN | true | |
| 36 | resnext0_stage2_conv1_fwd | Convolution | [1,256,56,56] | 6171 | 0 | 0 | 0 | 0 | 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) | 0.00 | 0 | 0.00 | 0.00 | 0.00 | 0.00 | NaN | true | |
| 36 | resnext0_stage2_conv1_fwd | Convolution | [1,256,56,56] | 6171 | 0 | 0 | 0 | 0 | 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) | 0.00 | 0 | 0.00 | 0.00 | 0.00 | 0.00 | NaN | true | |
| 36 | resnext0_stage2_conv1_fwd | Convolution | [1,256,56,56] | 6171 | 0 | 0 | 0 | 0 | 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) | 0.00 | 0 | 0.00 | 0.00 | 0.00 | 0.00 | NaN | true | |
| 36 | resnext0_stage2_conv1_fwd | Convolution | [1,256,56,56] | 6171 | 0 | 0 | 0 | 0 | 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) | 0.00 | 0 | 0.00 | 0.00 | 0.00 | 0.00 | NaN | true | |
| 36 | resnext0_stage2_conv1_fwd | Convolution | [1,256,56,56] | 6171 | 0 | 0 | 0 | 0 | 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) | 0.00 | 0 | 0.00 | 0.00 | 0.00 | 0.00 | NaN | true | |
| 36 | resnext0_stage2_conv1_fwd | Convolution | [1,256,56,56] | 6171 | 0 | 0 | 0 | 0 | 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) | 0.00 | 0 | 0.00 | 0.00 | 0.00 | 0.00 | NaN | true | |
| 36 | resnext0_stage2_conv1_fwd | Convolution | [1,256,56,56] | 6171 | 0 | 0 | 0 | 0 | 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) | 0.00 | 0 | 0.00 | 0.00 | 0.00 | 0.00 | NaN | true | |
| 36 | resnext0_stage2_conv1_fwd | Convolution | [1,256,56,56] | 6171 | 0 | 0 | 0 | 0 | 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) | 0.00 | 0 | 0.00 | 0.00 | 0.00 | 0.00 | NaN | true | |
| 36 | resnext0_stage2_conv1_fwd | Convolution | [1,256,56,56] | 6171 | 0 | 0 | 0 | 0 | 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) | 0.00 | 0 | 0.00 | 0.00 | 0.00 | 0.00 | NaN | true | |
| 36 | resnext0_stage2_conv1_fwd | Convolution | [1,256,56,56] | 6171 | 0 | 0 | 0 | 0 | 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) | 0.00 | 0 | 0.00 | 0.00 | 0.00 | 0.00 | NaN | true | |
| 36 | resnext0_stage2_conv1_fwd | Convolution | [1,256,56,56] | 6171 | 0 | 0 | 0 | 0 | 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) | 0.00 | 0 | 0.00 | 0.00 | 0.00 | 0.00 | NaN | true | |
| 36 | resnext0_stage2_conv1_fwd | Convolution | [1,256,56,56] | 6171 | 0 | 0 | 0 | 0 | 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) | 0.00 | 0 | 0.00 | 0.00 | 0.00 | 0.00 | NaN | true | |
| 36 | resnext0_stage2_conv1_fwd | Convolution | [1,256,56,56] | 6171 | 0 | 0 | 0 | 0 | 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) | 0.00 | 0 | 0.00 | 0.00 | 0.00 | 0.00 | NaN | true | |
| 36 | resnext0_stage2_conv1_fwd | Convolution | [1,256,56,56] | 6171 | 0 | 0 | 0 | 0 | 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) | 0.00 | 0 | 0.00 | 0.00 | 0.00 | 0.00 | NaN | true | |
| 36 | resnext0_stage2_conv1_fwd | Convolution | [1,256,56,56] | 6171 | 0 | 0 | 0 | 0 | 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) | 0.00 | 0 | 0.00 | 0.00 | 0.00 | 0.00 | NaN | true | |
| 36 | resnext0_stage2_conv1_fwd | Convolution | [1,256,56,56] | 6171 | 0 | 0 | 0 | 0 | 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) | 0.00 | 0 | 0.00 | 0.00 | 0.00 | 0.00 | NaN | true | |
| 36 | resnext0_stage2_conv1_fwd | Convolution | [1,256,56,56] | 6171 | 0 | 0 | 0 | 0 | 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) | 0.00 | 0 | 0.00 | 0.00 | 0.00 | 0.00 | NaN | true | |
| 36 | resnext0_stage2_conv1_fwd | Convolution | [1,256,56,56] | 6171 | 0 | 0 | 0 | 0 | 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) | 0.00 | 0 | 0.00 | 0.00 | 0.00 | 0.00 | NaN | true | |
| 36 | resnext0_stage2_conv1_fwd | Convolution | [1,256,56,56] | 6171 | 0 | 0 | 0 | 0 | 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) | 0.00 | 0 | 0.00 | 0.00 | 0.00 | 0.00 | NaN | true | |
| 36 | resnext0_stage2_conv1_fwd | Convolution | [1,256,56,56] | 6171 | 0 | 0 | 0 | 0 | 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) | 0.00 | 0 | 0.00 | 0.00 | 0.00 | 0.00 | NaN | true | |
| 36 | resnext0_stage2_conv1_fwd | Convolution | [1,256,56,56] | 6171 | 0 | 0 | 0 | 0 | 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) | 0.00 | 0 | 0.00 | 0.00 | 0.00 | 0.00 | NaN | true | |
| 36 | resnext0_stage2_conv1_fwd | Convolution | [1,256,56,56] | 6171 | 0 | 0 | 0 | 0 | 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) | 0.00 | 0 | 0.00 | 0.00 | 0.00 | 0.00 | NaN | true | |
| 36 | resnext0_stage2_conv1_fwd | Convolution | [1,256,56,56] | 6171 | 0 | 0 | 0 | 0 | 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) | 0.00 | 0 | 0.00 | 0.00 | 0.00 | 0.00 | NaN | true | |
| 36 | resnext0_stage2_conv1_fwd | Convolution | [1,256,56,56] | 6171 | 0 | 0 | 0 | 0 | 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) | 0.00 | 0 | 0.00 | 0.00 | 0.00 | 0.00 | NaN | true | |
| 36 | resnext0_stage2_conv1_fwd | Convolution | [1,256,56,56] | 6171 | 0 | 0 | 0 | 0 | 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) | 0.00 | 0 | 0.00 | 0.00 | 0.00 | 0.00 | NaN | true | |
| 36 | resnext0_stage2_conv1_fwd | Convolution | [1,256,56,56] | 6171 | 0 | 0 | 0 | 0 | 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) | 0.00 | 0 | 0.00 | 0.00 | 0.00 | 0.00 | NaN | true | |
| 36 | resnext0_stage2_conv1_fwd | Convolution | [1,256,56,56] | 6171 | 0 | 0 | 0 | 0 | 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) | 0.00 | 0 | 0.00 | 0.00 | 0.00 | 0.00 | NaN | true | |
| 36 | resnext0_stage2_conv1_fwd | Convolution | [1,256,56,56] | 6171 | 0 | 0 | 0 | 0 | 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) | 0.00 | 0 | 0.00 | 0.00 | 0.00 | 0.00 | NaN | true | |
| 36 | resnext0_stage2_conv1_fwd | Convolution | [1,256,56,56] | 6171 | 0 | 0 | 0 | 0 | 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) | 0.00 | 0 | 0.00 | 0.00 | 0.00 | 0.00 | NaN | true | |
| 36 | resnext0_stage2_conv1_fwd | Convolution | [1,256,56,56] | 6171 | 0 | 0 | 0 | 0 | 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) | 0.00 | 0 | 0.00 | 0.00 | 0.00 | 0.00 | NaN | true | |
| 37 | resnext0_stage2_batchnorm1_fwd | BatchNorm | [1,256,28,28] | 106.667 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | void cudnn::detail::bn_fw_inf_1C11_kernel_NCHW<float, float, true, 1>(float, float, cudnnTensorStruct, float const*, cudnnTensorStruct, float*, cudnnTensorStruct, float const*, float const*, float const*, float const*, float) | 0.00 | 0 | 0.00 | 0.00 | 0.00 | 0.00 | NaN | true | |
| 38 | resnext0_stage2_relu1_fwd | Activation | [1,256,28,28] | 45.333 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | void op_generic_tensor_kernel<2, float, float, float, 256, (cudnnGenericOp_t)8, (cudnnNanPropagation_t)0, (cudnnDimOrder_t)0, 1>(cudnnTensorStruct, float*, cudnnTensorStruct, float const*, cudnnTensorStruct, float const*, float, float, float, float, dimArray, reducedDivisorArray, bool) | 0.00 | 0 | 0.00 | 0.00 | 0.00 | 0.00 | NaN | true | |
| 39 | resnext0_stage2_conv2_fwd | Convolution | [1,256,28,28] | 10578 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | cudnn::gemm::computeOffsetsKernel(cudnn::gemm::ComputeOffsetsParams) | 0.00 | 0 | 0.00 | 0.00 | 0.00 | 0.00 | NaN | true | |
| 39 | resnext0_stage2_conv2_fwd | Convolution | [1,256,28,28] | 10578 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | volta_scudnn_128x64_relu_interior_nn_v1 | 0.00 | 0 | 0.00 | 0.00 | 0.00 | 0.00 | NaN | true | |
| 40 | resnext0_stage2_batchnorm2_fwd | BatchNorm | [1,512,28,28] | 134 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | void cudnn::detail::bn_fw_inf_1C11_kernel_NCHW<float, float, true, 1>(float, float, cudnnTensorStruct, float const*, cudnnTensorStruct, float*, cudnnTensorStruct, float const*, float const*, float const*, float const*, float) | 0.00 | 0 | 0.00 | 0.00 | 0.00 | 0.00 | NaN | true | |
| 41 | resnext0_stage2_conv3_fwd | Convolution | [1,256,56,56] | 10683.333 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | cudnn::gemm::computeOffsetsKernel(cudnn::gemm::ComputeOffsetsParams) | 0.00 | 0 | 0.00 | 0.00 | 0.00 | 0.00 | NaN | true | |
| 41 | resnext0_stage2_conv3_fwd | Convolution | [1,256,56,56] | 10683.333 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | volta_scudnn_128x64_relu_interior_nn_v1 | 0.00 | 0 | 0.00 | 0.00 | 0.00 | 0.00 | NaN | true | |
| 42 | resnext0_stage2_batchnorm3_fwd | BatchNorm | [1,512,28,28] | 122.333 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | void cudnn::detail::bn_fw_inf_1C11_kernel_NCHW<float, float, true, 1>(float, float, cudnnTensorStruct, float const*, cudnnTensorStruct, float*, cudnnTensorStruct, float const*, float const*, float const*, float const*, float) | 0.00 | 0 | 0.00 | 0.00 | 0.00 | 0.00 | NaN | true | |
| 43 | add_resnext0_stage2_activation0 | add_relu | [1,512,28,28] | 87.667 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | void mxnet::op::mxnet_op::mxnet_generic_kernel<mxnet::op::AddReluKernel, float*, float*, float*, mxnet::OpReqType>(int, float*, float*, float*, mxnet::OpReqType) | 0.00 | 0 | 0.00 | 0.00 | 0.00 | 0.00 | NaN | true | |
| 44 | resnext0_stage2_conv4_fwd | Convolution | [1,512,28,28] | 10399.667 | 0 | 0 | 0 | 0 | 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) | 0.00 | 0 | 0.00 | 0.00 | 0.00 | 0.00 | NaN | true | |
| 45 | resnext0_stage2_batchnorm4_fwd | BatchNorm | [1,256,28,28] | 103 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | void cudnn::detail::bn_fw_inf_1C11_kernel_NCHW<float, float, true, 1>(float, float, cudnnTensorStruct, float const*, cudnnTensorStruct, float*, cudnnTensorStruct, float const*, float const*, float const*, float const*, float) | 0.00 | 0 | 0.00 | 0.00 | 0.00 | 0.00 | NaN | true | |
| 46 | resnext0_stage2_relu2_fwd | Activation | [1,256,28,28] | 35.333 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | void op_generic_tensor_kernel<2, float, float, float, 256, (cudnnGenericOp_t)8, (cudnnNanPropagation_t)0, (cudnnDimOrder_t)0, 1>(cudnnTensorStruct, float*, cudnnTensorStruct, float const*, cudnnTensorStruct, float const*, float, float, float, float, dimArray, reducedDivisorArray, bool) | 0.00 | 0 | 0.00 | 0.00 | 0.00 | 0.00 | NaN | true | |
| 47 | resnext0_stage2_conv5_fwd | Convolution | [1,256,28,28] | 5784 | 0 | 0 | 0 | 0 | 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) | 0.00 | 0 | 0.00 | 0.00 | 0.00 | 0.00 | NaN | true | |
| 47 | resnext0_stage2_conv5_fwd | Convolution | [1,256,28,28] | 5784 | 0 | 0 | 0 | 0 | 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) | 0.00 | 0 | 0.00 | 0.00 | 0.00 | 0.00 | NaN | true | |
| 47 | resnext0_stage2_conv5_fwd | Convolution | [1,256,28,28] | 5784 | 0 | 0 | 0 | 0 | 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) | 0.00 | 0 | 0.00 | 0.00 | 0.00 | 0.00 | NaN | true | |
| 47 | resnext0_stage2_conv5_fwd | Convolution | [1,256,28,28] | 5784 | 0 | 0 | 0 | 0 | 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) | 0.00 | 0 | 0.00 | 0.00 | 0.00 | 0.00 | NaN | true | |
| 47 | resnext0_stage2_conv5_fwd | Convolution | [1,256,28,28] | 5784 | 0 | 0 | 0 | 0 | 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) | 0.00 | 0 | 0.00 | 0.00 | 0.00 | 0.00 | NaN | true | |
| 47 | resnext0_stage2_conv5_fwd | Convolution | [1,256,28,28] | 5784 | 0 | 0 | 0 | 0 | 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) | 0.00 | 0 | 0.00 | 0.00 | 0.00 | 0.00 | NaN | true | |
| 47 | resnext0_stage2_conv5_fwd | Convolution | [1,256,28,28] | 5784 | 0 | 0 | 0 | 0 | 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) | 0.00 | 0 | 0.00 | 0.00 | 0.00 | 0.00 | NaN | true | |
| 47 | resnext0_stage2_conv5_fwd | Convolution | [1,256,28,28] | 5784 | 0 | 0 | 0 | 0 | 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) | 0.00 | 0 | 0.00 | 0.00 | 0.00 | 0.00 | NaN | true | |
| 47 | resnext0_stage2_conv5_fwd | Convolution | [1,256,28,28] | 5784 | 0 | 0 | 0 | 0 | 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) | 0.00 | 0 | 0.00 | 0.00 | 0.00 | 0.00 | NaN | true | |
| 47 | resnext0_stage2_conv5_fwd | Convolution | [1,256,28,28] | 5784 | 0 | 0 | 0 | 0 | 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) | 0.00 | 0 | 0.00 | 0.00 | 0.00 | 0.00 | NaN | true | |
| 47 | resnext0_stage2_conv5_fwd | Convolution | [1,256,28,28] | 5784 | 0 | 0 | 0 | 0 | 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) | 0.00 | 0 | 0.00 | 0.00 | 0.00 | 0.00 | NaN | true | |
| 47 | resnext0_stage2_conv5_fwd | Convolution | [1,256,28,28] | 5784 | 0 | 0 | 0 | 0 | 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) | 0.00 | 0 | 0.00 | 0.00 | 0.00 | 0.00 | NaN | true | |
| 47 | resnext0_stage2_conv5_fwd | Convolution | [1,256,28,28] | 5784 | 0 | 0 | 0 | 0 | 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) | 0.00 | 0 | 0.00 | 0.00 | 0.00 | 0.00 | NaN | true | |
| 47 | resnext0_stage2_conv5_fwd | Convolution | [1,256,28,28] | 5784 | 0 | 0 | 0 | 0 | 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) | 0.00 | 0 | 0.00 | 0.00 | 0.00 | 0.00 | NaN | true | |
| 47 | resnext0_stage2_conv5_fwd | Convolution | [1,256,28,28] | 5784 | 0 | 0 | 0 | 0 | 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) | 0.00 | 0 | 0.00 | 0.00 | 0.00 | 0.00 | NaN | true | |
| 47 | resnext0_stage2_conv5_fwd | Convolution | [1,256,28,28] | 5784 | 0 | 0 | 0 | 0 | 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) | 0.00 | 0 | 0.00 | 0.00 | 0.00 | 0.00 | NaN | true | |
| 47 | resnext0_stage2_conv5_fwd | Convolution | [1,256,28,28] | 5784 | 0 | 0 | 0 | 0 | 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) | 0.00 | 0 | 0.00 | 0.00 | 0.00 | 0.00 | NaN | true | |
| 47 | resnext0_stage2_conv5_fwd | Convolution | [1,256,28,28] | 5784 | 0 | 0 | 0 | 0 | 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) | 0.00 | 0 | 0.00 | 0.00 | 0.00 | 0.00 | NaN | true | |
| 47 | resnext0_stage2_conv5_fwd | Convolution | [1,256,28,28] | 5784 | 0 | 0 | 0 | 0 | 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) | 0.00 | 0 | 0.00 | 0.00 | 0.00 | 0.00 | NaN | true | |
| 47 | resnext0_stage2_conv5_fwd | Convolution | [1,256,28,28] | 5784 | 0 | 0 | 0 | 0 | 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) | 0.00 | 0 | 0.00 | 0.00 | 0.00 | 0.00 | NaN | true | |
| 47 | resnext0_stage2_conv5_fwd | Convolution | [1,256,28,28] | 5784 | 0 | 0 | 0 | 0 | 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) | 0.00 | 0 | 0.00 | 0.00 | 0.00 | 0.00 | NaN | true | |
| 47 | resnext0_stage2_conv5_fwd | Convolution | [1,256,28,28] | 5784 | 0 | 0 | 0 | 0 | 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) | 0.00 | 0 | 0.00 | 0.00 | 0.00 | 0.00 | NaN | true | |
| 47 | resnext0_stage2_conv5_fwd | Convolution | [1,256,28,28] | 5784 | 0 | 0 | 0 | 0 | 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) | 0.00 | 0 | 0.00 | 0.00 | 0.00 | 0.00 | NaN | true | |
| 47 | resnext0_stage2_conv5_fwd | Convolution | [1,256,28,28] | 5784 | 0 | 0 | 0 | 0 | 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) | 0.00 | 0 | 0.00 | 0.00 | 0.00 | 0.00 | NaN | true | |
| 47 | resnext0_stage2_conv5_fwd | Convolution | [1,256,28,28] | 5784 | 0 | 0 | 0 | 0 | 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) | 0.00 | 0 | 0.00 | 0.00 | 0.00 | 0.00 | NaN | true | |
| 47 | resnext0_stage2_conv5_fwd | Convolution | [1,256,28,28] | 5784 | 0 | 0 | 0 | 0 | 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) | 0.00 | 0 | 0.00 | 0.00 | 0.00 | 0.00 | NaN | true | |
| 47 | resnext0_stage2_conv5_fwd | Convolution | [1,256,28,28] | 5784 | 0 | 0 | 0 | 0 | 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) | 0.00 | 0 | 0.00 | 0.00 | 0.00 | 0.00 | NaN | true | |
| 47 | resnext0_stage2_conv5_fwd | Convolution | [1,256,28,28] | 5784 | 0 | 0 | 0 | 0 | 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) | 0.00 | 0 | 0.00 | 0.00 | 0.00 | 0.00 | NaN | true | |
| 47 | resnext0_stage2_conv5_fwd | Convolution | [1,256,28,28] | 5784 | 0 | 0 | 0 | 0 | 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) | 0.00 | 0 | 0.00 | 0.00 | 0.00 | 0.00 | NaN | true | |
| 47 | resnext0_stage2_conv5_fwd | Convolution | [1,256,28,28] | 5784 | 0 | 0 | 0 | 0 | 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) | 0.00 | 0 | 0.00 | 0.00 | 0.00 | 0.00 | NaN | true | |
| 47 | resnext0_stage2_conv5_fwd | Convolution | [1,256,28,28] | 5784 | 0 | 0 | 0 | 0 | 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) | 0.00 | 0 | 0.00 | 0.00 | 0.00 | 0.00 | NaN | true | |
| 47 | resnext0_stage2_conv5_fwd | Convolution | [1,256,28,28] | 5784 | 0 | 0 | 0 | 0 | 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) | 0.00 | 0 | 0.00 | 0.00 | 0.00 | 0.00 | NaN | true | |
| 48 | resnext0_stage2_batchnorm5_fwd | BatchNorm | [1,256,28,28] | 109.667 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | void cudnn::detail::bn_fw_inf_1C11_kernel_NCHW<float, float, true, 1>(float, float, cudnnTensorStruct, float const*, cudnnTensorStruct, float*, cudnnTensorStruct, float const*, float const*, float const*, float const*, float) | 0.00 | 0 | 0.00 | 0.00 | 0.00 | 0.00 | NaN | true | |
| 49 | resnext0_stage2_relu3_fwd | Activation | [1,256,28,28] | 39 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | void op_generic_tensor_kernel<2, float, float, float, 256, (cudnnGenericOp_t)8, (cudnnNanPropagation_t)0, (cudnnDimOrder_t)0, 1>(cudnnTensorStruct, float*, cudnnTensorStruct, float const*, cudnnTensorStruct, float const*, float, float, float, float, dimArray, reducedDivisorArray, bool) | 0.00 | 0 | 0.00 | 0.00 | 0.00 | 0.00 | NaN | true | |
| 50 | resnext0_stage2_conv6_fwd | Convolution | [1,256,28,28] | 10492.333 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | cudnn::gemm::computeOffsetsKernel(cudnn::gemm::ComputeOffsetsParams) | 0.00 | 0 | 0.00 | 0.00 | 0.00 | 0.00 | NaN | true | |
| 50 | resnext0_stage2_conv6_fwd | Convolution | [1,256,28,28] | 10492.333 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | volta_scudnn_128x64_relu_interior_nn_v1 | 0.00 | 0 | 0.00 | 0.00 | 0.00 | 0.00 | NaN | true | |
| 51 | resnext0_stage2_batchnorm6_fwd | BatchNorm | [1,512,28,28] | 121 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | void cudnn::detail::bn_fw_inf_1C11_kernel_NCHW<float, float, true, 1>(float, float, cudnnTensorStruct, float const*, cudnnTensorStruct, float*, cudnnTensorStruct, float const*, float const*, float const*, float const*, float) | 0.00 | 0 | 0.00 | 0.00 | 0.00 | 0.00 | NaN | true | |
| 52 | add_resnext0_stage2_activation1 | add_relu | [1,512,28,28] | 100.333 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | void mxnet::op::mxnet_op::mxnet_generic_kernel<mxnet::op::AddReluKernel, float*, float*, float*, mxnet::OpReqType>(int, float*, float*, float*, mxnet::OpReqType) | 0.00 | 0 | 0.00 | 0.00 | 0.00 | 0.00 | NaN | true | |
| 53 | resnext0_stage2_conv7_fwd | Convolution | [1,512,28,28] | 10681.667 | 0 | 0 | 0 | 0 | 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) | 0.00 | 0 | 0.00 | 0.00 | 0.00 | 0.00 | NaN | true | |
| 54 | resnext0_stage2_batchnorm7_fwd | BatchNorm | [1,256,28,28] | 114 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | void cudnn::detail::bn_fw_inf_1C11_kernel_NCHW<float, float, true, 1>(float, float, cudnnTensorStruct, float const*, cudnnTensorStruct, float*, cudnnTensorStruct, float const*, float const*, float const*, float const*, float) | 0.00 | 0 | 0.00 | 0.00 | 0.00 | 0.00 | NaN | true | |
| 55 | resnext0_stage2_relu4_fwd | Activation | [1,256,28,28] | 36.667 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | void op_generic_tensor_kernel<2, float, float, float, 256, (cudnnGenericOp_t)8, (cudnnNanPropagation_t)0, (cudnnDimOrder_t)0, 1>(cudnnTensorStruct, float*, cudnnTensorStruct, float const*, cudnnTensorStruct, float const*, float, float, float, float, dimArray, reducedDivisorArray, bool) | 0.00 | 0 | 0.00 | 0.00 | 0.00 | 0.00 | NaN | true | |
| 56 | resnext0_stage2_conv8_fwd | Convolution | [1,256,28,28] | 5849 | 0 | 0 | 0 | 0 | 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) | 0.00 | 0 | 0.00 | 0.00 | 0.00 | 0.00 | NaN | true | |
| 56 | resnext0_stage2_conv8_fwd | Convolution | [1,256,28,28] | 5849 | 0 | 0 | 0 | 0 | 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) | 0.00 | 0 | 0.00 | 0.00 | 0.00 | 0.00 | NaN | true | |
| 56 | resnext0_stage2_conv8_fwd | Convolution | [1,256,28,28] | 5849 | 0 | 0 | 0 | 0 | 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) | 0.00 | 0 | 0.00 | 0.00 | 0.00 | 0.00 | NaN | true | |
| 56 | resnext0_stage2_conv8_fwd | Convolution | [1,256,28,28] | 5849 | 0 | 0 | 0 | 0 | 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) | 0.00 | 0 | 0.00 | 0.00 | 0.00 | 0.00 | NaN | true | |
| 56 | resnext0_stage2_conv8_fwd | Convolution | [1,256,28,28] | 5849 | 0 | 0 | 0 | 0 | 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) | 0.00 | 0 | 0.00 | 0.00 | 0.00 | 0.00 | NaN | true | |
| 56 | resnext0_stage2_conv8_fwd | Convolution | [1,256,28,28] | 5849 | 0 | 0 | 0 | 0 | 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) | 0.00 | 0 | 0.00 | 0.00 | 0.00 | 0.00 | NaN | true | |
| 56 | resnext0_stage2_conv8_fwd | Convolution | [1,256,28,28] | 5849 | 0 | 0 | 0 | 0 | 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) | 0.00 | 0 | 0.00 | 0.00 | 0.00 | 0.00 | NaN | true | |
| 56 | resnext0_stage2_conv8_fwd | Convolution | [1,256,28,28] | 5849 | 0 | 0 | 0 | 0 | 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) | 0.00 | 0 | 0.00 | 0.00 | 0.00 | 0.00 | NaN | true | |
| 56 | resnext0_stage2_conv8_fwd | Convolution | [1,256,28,28] | 5849 | 0 | 0 | 0 | 0 | 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) | 0.00 | 0 | 0.00 | 0.00 | 0.00 | 0.00 | NaN | true | |
| 56 | resnext0_stage2_conv8_fwd | Convolution | [1,256,28,28] | 5849 | 0 | 0 | 0 | 0 | 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) | 0.00 | 0 | 0.00 | 0.00 | 0.00 | 0.00 | NaN | true | |
| 56 | resnext0_stage2_conv8_fwd | Convolution | [1,256,28,28] | 5849 | 0 | 0 | 0 | 0 | 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) | 0.00 | 0 | 0.00 | 0.00 | 0.00 | 0.00 | NaN | true | |
| 56 | resnext0_stage2_conv8_fwd | Convolution | [1,256,28,28] | 5849 | 0 | 0 | 0 | 0 | 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) | 0.00 | 0 | 0.00 | 0.00 | 0.00 | 0.00 | NaN | true | |
| 56 | resnext0_stage2_conv8_fwd | Convolution | [1,256,28,28] | 5849 | 0 | 0 | 0 | 0 | 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) | 0.00 | 0 | 0.00 | 0.00 | 0.00 | 0.00 | NaN | true | |
| 56 | resnext0_stage2_conv8_fwd | Convolution | [1,256,28,28] | 5849 | 0 | 0 | 0 | 0 | 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) | 0.00 | 0 | 0.00 | 0.00 | 0.00 | 0.00 | NaN | true | |
| 56 | resnext0_stage2_conv8_fwd | Convolution | [1,256,28,28] | 5849 | 0 | 0 | 0 | 0 | 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) | 0.00 | 0 | 0.00 | 0.00 | 0.00 | 0.00 | NaN | true | |
| 56 | resnext0_stage2_conv8_fwd | Convolution | [1,256,28,28] | 5849 | 0 | 0 | 0 | 0 | 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) | 0.00 | 0 | 0.00 | 0.00 | 0.00 | 0.00 | NaN | true | |
| 56 | resnext0_stage2_conv8_fwd | Convolution | [1,256,28,28] | 5849 | 0 | 0 | 0 | 0 | 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) | 0.00 | 0 | 0.00 | 0.00 | 0.00 | 0.00 | NaN | true | |
| 56 | resnext0_stage2_conv8_fwd | Convolution | [1,256,28,28] | 5849 | 0 | 0 | 0 | 0 | 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) | 0.00 | 0 | 0.00 | 0.00 | 0.00 | 0.00 | NaN | true | |
| 56 | resnext0_stage2_conv8_fwd | Convolution | [1,256,28,28] | 5849 | 0 | 0 | 0 | 0 | 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) | 0.00 | 0 | 0.00 | 0.00 | 0.00 | 0.00 | NaN | true | |
| 56 | resnext0_stage2_conv8_fwd | Convolution | [1,256,28,28] | 5849 | 0 | 0 | 0 | 0 | 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) | 0.00 | 0 | 0.00 | 0.00 | 0.00 | 0.00 | NaN | true | |
| 56 | resnext0_stage2_conv8_fwd | Convolution | [1,256,28,28] | 5849 | 0 | 0 | 0 | 0 | 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) | 0.00 | 0 | 0.00 | 0.00 | 0.00 | 0.00 | NaN | true | |
| 56 | resnext0_stage2_conv8_fwd | Convolution | [1,256,28,28] | 5849 | 0 | 0 | 0 | 0 | 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) | 0.00 | 0 | 0.00 | 0.00 | 0.00 | 0.00 | NaN | true | |
| 56 | resnext0_stage2_conv8_fwd | Convolution | [1,256,28,28] | 5849 | 0 | 0 | 0 | 0 | 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) | 0.00 | 0 | 0.00 | 0.00 | 0.00 | 0.00 | NaN | true | |
| 56 | resnext0_stage2_conv8_fwd | Convolution | [1,256,28,28] | 5849 | 0 | 0 | 0 | 0 | 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) | 0.00 | 0 | 0.00 | 0.00 | 0.00 | 0.00 | NaN | true | |
| 56 | resnext0_stage2_conv8_fwd | Convolution | [1,256,28,28] | 5849 | 0 | 0 | 0 | 0 | 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) | 0.00 | 0 | 0.00 | 0.00 | 0.00 | 0.00 | NaN | true | |
| 56 | resnext0_stage2_conv8_fwd | Convolution | [1,256,28,28] | 5849 | 0 | 0 | 0 | 0 | 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) | 0.00 | 0 | 0.00 | 0.00 | 0.00 | 0.00 | NaN | true | |
| 56 | resnext0_stage2_conv8_fwd | Convolution | [1,256,28,28] | 5849 | 0 | 0 | 0 | 0 | 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) | 0.00 | 0 | 0.00 | 0.00 | 0.00 | 0.00 | NaN | true | |
| 56 | resnext0_stage2_conv8_fwd | Convolution | [1,256,28,28] | 5849 | 0 | 0 | 0 | 0 | 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) | 0.00 | 0 | 0.00 | 0.00 | 0.00 | 0.00 | NaN | true | |
| 56 | resnext0_stage2_conv8_fwd | Convolution | [1,256,28,28] | 5849 | 0 | 0 | 0 | 0 | 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) | 0.00 | 0 | 0.00 | 0.00 | 0.00 | 0.00 | NaN | true | |
| 56 | resnext0_stage2_conv8_fwd | Convolution | [1,256,28,28] | 5849 | 0 | 0 | 0 | 0 | 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) | 0.00 | 0 | 0.00 | 0.00 | 0.00 | 0.00 | NaN | true | |
| 56 | resnext0_stage2_conv8_fwd | Convolution | [1,256,28,28] | 5849 | 0 | 0 | 0 | 0 | 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) | 0.00 | 0 | 0.00 | 0.00 | 0.00 | 0.00 | NaN | true | |
| 56 | resnext0_stage2_conv8_fwd | Convolution | [1,256,28,28] | 5849 | 0 | 0 | 0 | 0 | 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) | 0.00 | 0 | 0.00 | 0.00 | 0.00 | 0.00 | NaN | true | |
| 57 | resnext0_stage2_batchnorm8_fwd | BatchNorm | [1,256,28,28] | 108.333 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | void cudnn::detail::bn_fw_inf_1C11_kernel_NCHW<float, float, true, 1>(float, float, cudnnTensorStruct, float const*, cudnnTensorStruct, float*, cudnnTensorStruct, float const*, float const*, float const*, float const*, float) | 0.00 | 0 | 0.00 | 0.00 | 0.00 | 0.00 | NaN | true | |
| 58 | resnext0_stage2_relu5_fwd | Activation | [1,256,28,28] | 39 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | void op_generic_tensor_kernel<2, float, float, float, 256, (cudnnGenericOp_t)8, (cudnnNanPropagation_t)0, (cudnnDimOrder_t)0, 1>(cudnnTensorStruct, float*, cudnnTensorStruct, float const*, cudnnTensorStruct, float const*, float, float, float, float, dimArray, reducedDivisorArray, bool) | 0.00 | 0 | 0.00 | 0.00 | 0.00 | 0.00 | NaN | true | |
| 59 | resnext0_stage2_conv9_fwd | Convolution | [1,256,28,28] | 10626.667 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | cudnn::gemm::computeOffsetsKernel(cudnn::gemm::ComputeOffsetsParams) | 0.00 | 0 | 0.00 | 0.00 | 0.00 | 0.00 | NaN | true | |
| 59 | resnext0_stage2_conv9_fwd | Convolution | [1,256,28,28] | 10626.667 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | volta_scudnn_128x64_relu_interior_nn_v1 | 0.00 | 0 | 0.00 | 0.00 | 0.00 | 0.00 | NaN | true | |
| 60 | resnext0_stage2_batchnorm9_fwd | BatchNorm | [1,512,28,28] | 124.667 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | void cudnn::detail::bn_fw_inf_1C11_kernel_NCHW<float, float, true, 1>(float, float, cudnnTensorStruct, float const*, cudnnTensorStruct, float*, cudnnTensorStruct, float const*, float const*, float const*, float const*, float) | 0.00 | 0 | 0.00 | 0.00 | 0.00 | 0.00 | NaN | true | |
| 61 | add_resnext0_stage2_activation2 | add_relu | [1,512,28,28] | 99.333 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | void mxnet::op::mxnet_op::mxnet_generic_kernel<mxnet::op::AddReluKernel, float*, float*, float*, mxnet::OpReqType>(int, float*, float*, float*, mxnet::OpReqType) | 0.00 | 0 | 0.00 | 0.00 | 0.00 | 0.00 | NaN | true | |
| 62 | resnext0_stage2_conv10_fwd | Convolution | [1,512,28,28] | 10634.333 | 0 | 0 | 0 | 0 | 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) | 0.00 | 0 | 0.00 | 0.00 | 0.00 | 0.00 | NaN | true | |
| 63 | resnext0_stage2_batchnorm10_fwd | BatchNorm | [1,256,28,28] | 112.333 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | void cudnn::detail::bn_fw_inf_1C11_kernel_NCHW<float, float, true, 1>(float, float, cudnnTensorStruct, float const*, cudnnTensorStruct, float*, cudnnTensorStruct, float const*, float const*, float const*, float const*, float) | 0.00 | 0 | 0.00 | 0.00 | 0.00 | 0.00 | NaN | true | |
| 64 | resnext0_stage2_relu6_fwd | Activation | [1,256,28,28] | 39.333 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | void op_generic_tensor_kernel<2, float, float, float, 256, (cudnnGenericOp_t)8, (cudnnNanPropagation_t)0, (cudnnDimOrder_t)0, 1>(cudnnTensorStruct, float*, cudnnTensorStruct, float const*, cudnnTensorStruct, float const*, float, float, float, float, dimArray, reducedDivisorArray, bool) | 0.00 | 0 | 0.00 | 0.00 | 0.00 | 0.00 | NaN | true | |
| 65 | resnext0_stage2_conv11_fwd | Convolution | [1,256,28,28] | 5948 | 0 | 0 | 0 | 0 | 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) | 0.00 | 0 | 0.00 | 0.00 | 0.00 | 0.00 | NaN | true | |
| 65 | resnext0_stage2_conv11_fwd | Convolution | [1,256,28,28] | 5948 | 0 | 0 | 0 | 0 | 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) | 0.00 | 0 | 0.00 | 0.00 | 0.00 | 0.00 | NaN | true | |
| 65 | resnext0_stage2_conv11_fwd | Convolution | [1,256,28,28] | 5948 | 0 | 0 | 0 | 0 | 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) | 0.00 | 0 | 0.00 | 0.00 | 0.00 | 0.00 | NaN | true | |
| 65 | resnext0_stage2_conv11_fwd | Convolution | [1,256,28,28] | 5948 | 0 | 0 | 0 | 0 | 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) | 0.00 | 0 | 0.00 | 0.00 | 0.00 | 0.00 | NaN | true | |
| 65 | resnext0_stage2_conv11_fwd | Convolution | [1,256,28,28] | 5948 | 0 | 0 | 0 | 0 | 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) | 0.00 | 0 | 0.00 | 0.00 | 0.00 | 0.00 | NaN | true | |
| 65 | resnext0_stage2_conv11_fwd | Convolution | [1,256,28,28] | 5948 | 0 | 0 | 0 | 0 | 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) | 0.00 | 0 | 0.00 | 0.00 | 0.00 | 0.00 | NaN | true | |
| 65 | resnext0_stage2_conv11_fwd | Convolution | [1,256,28,28] | 5948 | 0 | 0 | 0 | 0 | 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) | 0.00 | 0 | 0.00 | 0.00 | 0.00 | 0.00 | NaN | true | |
| 65 | resnext0_stage2_conv11_fwd | Convolution | [1,256,28,28] | 5948 | 0 | 0 | 0 | 0 | 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) | 0.00 | 0 | 0.00 | 0.00 | 0.00 | 0.00 | NaN | true | |
| 65 | resnext0_stage2_conv11_fwd | Convolution | [1,256,28,28] | 5948 | 0 | 0 | 0 | 0 | 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) | 0.00 | 0 | 0.00 | 0.00 | 0.00 | 0.00 | NaN | true | |
| 65 | resnext0_stage2_conv11_fwd | Convolution | [1,256,28,28] | 5948 | 0 | 0 | 0 | 0 | 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) | 0.00 | 0 | 0.00 | 0.00 | 0.00 | 0.00 | NaN | true | |
| 65 | resnext0_stage2_conv11_fwd | Convolution | [1,256,28,28] | 5948 | 0 | 0 | 0 | 0 | 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) | 0.00 | 0 | 0.00 | 0.00 | 0.00 | 0.00 | NaN | true | |
| 65 | resnext0_stage2_conv11_fwd | Convolution | [1,256,28,28] | 5948 | 0 | 0 | 0 | 0 | 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) | 0.00 | 0 | 0.00 | 0.00 | 0.00 | 0.00 | NaN | true | |
| 65 | resnext0_stage2_conv11_fwd | Convolution | [1,256,28,28] | 5948 | 0 | 0 | 0 | 0 | 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) | 0.00 | 0 | 0.00 | 0.00 | 0.00 | 0.00 | NaN | true | |
| 65 | resnext0_stage2_conv11_fwd | Convolution | [1,256,28,28] | 5948 | 0 | 0 | 0 | 0 | 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) | 0.00 | 0 | 0.00 | 0.00 | 0.00 | 0.00 | NaN | true | |
| 65 | resnext0_stage2_conv11_fwd | Convolution | [1,256,28,28] | 5948 | 0 | 0 | 0 | 0 | 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) | 0.00 | 0 | 0.00 | 0.00 | 0.00 | 0.00 | NaN | true | |
| 65 | resnext0_stage2_conv11_fwd | Convolution | [1,256,28,28] | 5948 | 0 | 0 | 0 | 0 | 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) | 0.00 | 0 | 0.00 | 0.00 | 0.00 | 0.00 | NaN | true | |
| 65 | resnext0_stage2_conv11_fwd | Convolution | [1,256,28,28] | 5948 | 0 | 0 | 0 | 0 | 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) | 0.00 | 0 | 0.00 | 0.00 | 0.00 | 0.00 | NaN | true | |
| 65 | resnext0_stage2_conv11_fwd | Convolution | [1,256,28,28] | 5948 | 0 | 0 | 0 | 0 | 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) | 0.00 | 0 | 0.00 | 0.00 | 0.00 | 0.00 | NaN | true | |
| 65 | resnext0_stage2_conv11_fwd | Convolution | [1,256,28,28] | 5948 | 0 | 0 | 0 | 0 | 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) | 0.00 | 0 | 0.00 | 0.00 | 0.00 | 0.00 | NaN | true | |
| 65 | resnext0_stage2_conv11_fwd | Convolution | [1,256,28,28] | 5948 | 0 | 0 | 0 | 0 | 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) | 0.00 | 0 | 0.00 | 0.00 | 0.00 | 0.00 | NaN | true | |
| 65 | resnext0_stage2_conv11_fwd | Convolution | [1,256,28,28] | 5948 | 0 | 0 | 0 | 0 | 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) | 0.00 | 0 | 0.00 | 0.00 | 0.00 | 0.00 | NaN | true | |
| 65 | resnext0_stage2_conv11_fwd | Convolution | [1,256,28,28] | 5948 | 0 | 0 | 0 | 0 | 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) | 0.00 | 0 | 0.00 | 0.00 | 0.00 | 0.00 | NaN | true | |
| 65 | resnext0_stage2_conv11_fwd | Convolution | [1,256,28,28] | 5948 | 0 | 0 | 0 | 0 | 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) | 0.00 | 0 | 0.00 | 0.00 | 0.00 | 0.00 | NaN | true | |
| 65 | resnext0_stage2_conv11_fwd | Convolution | [1,256,28,28] | 5948 | 0 | 0 | 0 | 0 | 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) | 0.00 | 0 | 0.00 | 0.00 | 0.00 | 0.00 | NaN | true | |
| 65 | resnext0_stage2_conv11_fwd | Convolution | [1,256,28,28] | 5948 | 0 | 0 | 0 | 0 | 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) | 0.00 | 0 | 0.00 | 0.00 | 0.00 | 0.00 | NaN | true | |
| 65 | resnext0_stage2_conv11_fwd | Convolution | [1,256,28,28] | 5948 | 0 | 0 | 0 | 0 | 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) | 0.00 | 0 | 0.00 | 0.00 | 0.00 | 0.00 | NaN | true | |
| 65 | resnext0_stage2_conv11_fwd | Convolution | [1,256,28,28] | 5948 | 0 | 0 | 0 | 0 | 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) | 0.00 | 0 | 0.00 | 0.00 | 0.00 | 0.00 | NaN | true | |
| 65 | resnext0_stage2_conv11_fwd | Convolution | [1,256,28,28] | 5948 | 0 | 0 | 0 | 0 | 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) | 0.00 | 0 | 0.00 | 0.00 | 0.00 | 0.00 | NaN | true | |
| 65 | resnext0_stage2_conv11_fwd | Convolution | [1,256,28,28] | 5948 | 0 | 0 | 0 | 0 | 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) | 0.00 | 0 | 0.00 | 0.00 | 0.00 | 0.00 | NaN | true | |
| 65 | resnext0_stage2_conv11_fwd | Convolution | [1,256,28,28] | 5948 | 0 | 0 | 0 | 0 | 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) | 0.00 | 0 | 0.00 | 0.00 | 0.00 | 0.00 | NaN | true | |
| 65 | resnext0_stage2_conv11_fwd | Convolution | [1,256,28,28] | 5948 | 0 | 0 | 0 | 0 | 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) | 0.00 | 0 | 0.00 | 0.00 | 0.00 | 0.00 | NaN | true | |
| 65 | resnext0_stage2_conv11_fwd | Convolution | [1,256,28,28] | 5948 | 0 | 0 | 0 | 0 | 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) | 0.00 | 0 | 0.00 | 0.00 | 0.00 | 0.00 | NaN | true | |
| 66 | resnext0_stage2_batchnorm11_fwd | BatchNorm | [1,256,28,28] | 109.333 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | void cudnn::detail::bn_fw_inf_1C11_kernel_NCHW<float, float, true, 1>(float, float, cudnnTensorStruct, float const*, cudnnTensorStruct, float*, cudnnTensorStruct, float const*, float const*, float const*, float const*, float) | 0.00 | 0 | 0.00 | 0.00 | 0.00 | 0.00 | NaN | true | |
| 67 | resnext0_stage2_relu7_fwd | Activation | [1,256,28,28] | 38 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | void op_generic_tensor_kernel<2, float, float, float, 256, (cudnnGenericOp_t)8, (cudnnNanPropagation_t)0, (cudnnDimOrder_t)0, 1>(cudnnTensorStruct, float*, cudnnTensorStruct, float const*, cudnnTensorStruct, float const*, float, float, float, float, dimArray, reducedDivisorArray, bool) | 0.00 | 0 | 0.00 | 0.00 | 0.00 | 0.00 | NaN | true | |
| 68 | resnext0_stage2_conv12_fwd | Convolution | [1,256,28,28] | 10774.667 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | cudnn::gemm::computeOffsetsKernel(cudnn::gemm::ComputeOffsetsParams) | 0.00 | 0 | 0.00 | 0.00 | 0.00 | 0.00 | NaN | true | |
| 68 | resnext0_stage2_conv12_fwd | Convolution | [1,256,28,28] | 10774.667 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | volta_scudnn_128x64_relu_interior_nn_v1 | 0.00 | 0 | 0.00 | 0.00 | 0.00 | 0.00 | NaN | true | |
| 69 | resnext0_stage2_batchnorm12_fwd | BatchNorm | [1,512,28,28] | 124.667 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | void cudnn::detail::bn_fw_inf_1C11_kernel_NCHW<float, float, true, 1>(float, float, cudnnTensorStruct, float const*, cudnnTensorStruct, float*, cudnnTensorStruct, float const*, float const*, float const*, float const*, float) | 0.00 | 0 | 0.00 | 0.00 | 0.00 | 0.00 | NaN | true | |
| 70 | add_resnext0_stage2_activation3 | add_relu | [1,512,28,28] | 103 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | void mxnet::op::mxnet_op::mxnet_generic_kernel<mxnet::op::AddReluKernel, float*, float*, float*, mxnet::OpReqType>(int, float*, float*, float*, mxnet::OpReqType) | 0.00 | 0 | 0.00 | 0.00 | 0.00 | 0.00 | NaN | true | |
| 71 | resnext0_stage3_conv0_fwd | Convolution | [1,512,28,28] | 20506.667 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | cudnn::gemm::computeOffsetsKernel(cudnn::gemm::ComputeOffsetsParams) | 0.00 | 0 | 0.00 | 0.00 | 0.00 | 0.00 | NaN | true | |
| 71 | resnext0_stage3_conv0_fwd | Convolution | [1,512,28,28] | 20506.667 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | volta_scudnn_128x64_relu_interior_nn_v1 | 0.00 | 0 | 0.00 | 0.00 | 0.00 | 0.00 | NaN | true | |
| 72 | resnext0_stage3_batchnorm0_fwd | BatchNorm | [1,512,28,28] | 128.667 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | void cudnn::detail::bn_fw_inf_1C11_kernel_NCHW<float, float, true, 1>(float, float, cudnnTensorStruct, float const*, cudnnTensorStruct, float*, cudnnTensorStruct, float const*, float const*, float const*, float const*, float) | 0.00 | 0 | 0.00 | 0.00 | 0.00 | 0.00 | NaN | true | |
| 73 | resnext0_stage3_relu0_fwd | Activation | [1,512,28,28] | 73.667 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | void op_generic_tensor_kernel<2, float, float, float, 256, (cudnnGenericOp_t)8, (cudnnNanPropagation_t)0, (cudnnDimOrder_t)0, 1>(cudnnTensorStruct, float*, cudnnTensorStruct, float const*, cudnnTensorStruct, float const*, float, float, float, float, dimArray, reducedDivisorArray, bool) | 0.00 | 0 | 0.00 | 0.00 | 0.00 | 0.00 | NaN | true | |
| 74 | resnext0_stage3_conv1_fwd | Convolution | [1,512,28,28] | 3578 | 0 | 0 | 0 | 0 | 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) | 0.00 | 0 | 0.00 | 0.00 | 0.00 | 0.00 | NaN | true | |
| 74 | resnext0_stage3_conv1_fwd | Convolution | [1,512,28,28] | 3578 | 0 | 0 | 0 | 0 | 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) | 0.00 | 0 | 0.00 | 0.00 | 0.00 | 0.00 | NaN | true | |
| 74 | resnext0_stage3_conv1_fwd | Convolution | [1,512,28,28] | 3578 | 0 | 0 | 0 | 0 | 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) | 0.00 | 0 | 0.00 | 0.00 | 0.00 | 0.00 | NaN | true | |
| 74 | resnext0_stage3_conv1_fwd | Convolution | [1,512,28,28] | 3578 | 0 | 0 | 0 | 0 | 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) | 0.00 | 0 | 0.00 | 0.00 | 0.00 | 0.00 | NaN | true | |
| 74 | resnext0_stage3_conv1_fwd | Convolution | [1,512,28,28] | 3578 | 0 | 0 | 0 | 0 | 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) | 0.00 | 0 | 0.00 | 0.00 | 0.00 | 0.00 | NaN | true | |
| 74 | resnext0_stage3_conv1_fwd | Convolution | [1,512,28,28] | 3578 | 0 | 0 | 0 | 0 | 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) | 0.00 | 0 | 0.00 | 0.00 | 0.00 | 0.00 | NaN | true | |
| 74 | resnext0_stage3_conv1_fwd | Convolution | [1,512,28,28] | 3578 | 0 | 0 | 0 | 0 | 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) | 0.00 | 0 | 0.00 | 0.00 | 0.00 | 0.00 | NaN | true | |
| 74 | resnext0_stage3_conv1_fwd | Convolution | [1,512,28,28] | 3578 | 0 | 0 | 0 | 0 | 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) | 0.00 | 0 | 0.00 | 0.00 | 0.00 | 0.00 | NaN | true | |
| 74 | resnext0_stage3_conv1_fwd | Convolution | [1,512,28,28] | 3578 | 0 | 0 | 0 | 0 | 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) | 0.00 | 0 | 0.00 | 0.00 | 0.00 | 0.00 | NaN | true | |
| 74 | resnext0_stage3_conv1_fwd | Convolution | [1,512,28,28] | 3578 | 0 | 0 | 0 | 0 | 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) | 0.00 | 0 | 0.00 | 0.00 | 0.00 | 0.00 | NaN | true | |
| 74 | resnext0_stage3_conv1_fwd | Convolution | [1,512,28,28] | 3578 | 0 | 0 | 0 | 0 | 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) | 0.00 | 0 | 0.00 | 0.00 | 0.00 | 0.00 | NaN | true | |
| 74 | resnext0_stage3_conv1_fwd | Convolution | [1,512,28,28] | 3578 | 0 | 0 | 0 | 0 | 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) | 0.00 | 0 | 0.00 | 0.00 | 0.00 | 0.00 | NaN | true | |
| 74 | resnext0_stage3_conv1_fwd | Convolution | [1,512,28,28] | 3578 | 0 | 0 | 0 | 0 | 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) | 0.00 | 0 | 0.00 | 0.00 | 0.00 | 0.00 | NaN | true | |
| 74 | resnext0_stage3_conv1_fwd | Convolution | [1,512,28,28] | 3578 | 0 | 0 | 0 | 0 | 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) | 0.00 | 0 | 0.00 | 0.00 | 0.00 | 0.00 | NaN | true | |
| 74 | resnext0_stage3_conv1_fwd | Convolution | [1,512,28,28] | 3578 | 0 | 0 | 0 | 0 | 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) | 0.00 | 0 | 0.00 | 0.00 | 0.00 | 0.00 | NaN | true | |
| 74 | resnext0_stage3_conv1_fwd | Convolution | [1,512,28,28] | 3578 | 0 | 0 | 0 | 0 | 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) | 0.00 | 0 | 0.00 | 0.00 | 0.00 | 0.00 | NaN | true | |
| 74 | resnext0_stage3_conv1_fwd | Convolution | [1,512,28,28] | 3578 | 0 | 0 | 0 | 0 | 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) | 0.00 | 0 | 0.00 | 0.00 | 0.00 | 0.00 | NaN | true | |
| 74 | resnext0_stage3_conv1_fwd | Convolution | [1,512,28,28] | 3578 | 0 | 0 | 0 | 0 | 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) | 0.00 | 0 | 0.00 | 0.00 | 0.00 | 0.00 | NaN | true | |
| 74 | resnext0_stage3_conv1_fwd | Convolution | [1,512,28,28] | 3578 | 0 | 0 | 0 | 0 | 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) | 0.00 | 0 | 0.00 | 0.00 | 0.00 | 0.00 | NaN | true | |
| 74 | resnext0_stage3_conv1_fwd | Convolution | [1,512,28,28] | 3578 | 0 | 0 | 0 | 0 | 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) | 0.00 | 0 | 0.00 | 0.00 | 0.00 | 0.00 | NaN | true | |
| 74 | resnext0_stage3_conv1_fwd | Convolution | [1,512,28,28] | 3578 | 0 | 0 | 0 | 0 | 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) | 0.00 | 0 | 0.00 | 0.00 | 0.00 | 0.00 | NaN | true | |
| 74 | resnext0_stage3_conv1_fwd | Convolution | [1,512,28,28] | 3578 | 0 | 0 | 0 | 0 | 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) | 0.00 | 0 | 0.00 | 0.00 | 0.00 | 0.00 | NaN | true | |
| 74 | resnext0_stage3_conv1_fwd | Convolution | [1,512,28,28] | 3578 | 0 | 0 | 0 | 0 | 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) | 0.00 | 0 | 0.00 | 0.00 | 0.00 | 0.00 | NaN | true | |
| 74 | resnext0_stage3_conv1_fwd | Convolution | [1,512,28,28] | 3578 | 0 | 0 | 0 | 0 | 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) | 0.00 | 0 | 0.00 | 0.00 | 0.00 | 0.00 | NaN | true | |
| 74 | resnext0_stage3_conv1_fwd | Convolution | [1,512,28,28] | 3578 | 0 | 0 | 0 | 0 | 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) | 0.00 | 0 | 0.00 | 0.00 | 0.00 | 0.00 | NaN | true | |
| 74 | resnext0_stage3_conv1_fwd | Convolution | [1,512,28,28] | 3578 | 0 | 0 | 0 | 0 | 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) | 0.00 | 0 | 0.00 | 0.00 | 0.00 | 0.00 | NaN | true | |
| 74 | resnext0_stage3_conv1_fwd | Convolution | [1,512,28,28] | 3578 | 0 | 0 | 0 | 0 | 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) | 0.00 | 0 | 0.00 | 0.00 | 0.00 | 0.00 | NaN | true | |
| 74 | resnext0_stage3_conv1_fwd | Convolution | [1,512,28,28] | 3578 | 0 | 0 | 0 | 0 | 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) | 0.00 | 0 | 0.00 | 0.00 | 0.00 | 0.00 | NaN | true | |
| 74 | resnext0_stage3_conv1_fwd | Convolution | [1,512,28,28] | 3578 | 0 | 0 | 0 | 0 | 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) | 0.00 | 0 | 0.00 | 0.00 | 0.00 | 0.00 | NaN | true | |
| 74 | resnext0_stage3_conv1_fwd | Convolution | [1,512,28,28] | 3578 | 0 | 0 | 0 | 0 | 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) | 0.00 | 0 | 0.00 | 0.00 | 0.00 | 0.00 | NaN | true | |
| 74 | resnext0_stage3_conv1_fwd | Convolution | [1,512,28,28] | 3578 | 0 | 0 | 0 | 0 | 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) | 0.00 | 0 | 0.00 | 0.00 | 0.00 | 0.00 | NaN | true | |
| 74 | resnext0_stage3_conv1_fwd | Convolution | [1,512,28,28] | 3578 | 0 | 0 | 0 | 0 | 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) | 0.00 | 0 | 0.00 | 0.00 | 0.00 | 0.00 | NaN | true | |
| 75 | resnext0_stage3_batchnorm1_fwd | BatchNorm | [1,512,14,14] | 69 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | void cudnn::detail::bn_fw_inf_1C11_kernel_NCHW<float, float, true, 1>(float, float, cudnnTensorStruct, float const*, cudnnTensorStruct, float*, cudnnTensorStruct, float const*, float const*, float const*, float const*, float) | 0.00 | 0 | 0.00 | 0.00 | 0.00 | 0.00 | NaN | true | |
| 76 | resnext0_stage3_relu1_fwd | Activation | [1,512,14,14] | 21.667 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | void op_generic_tensor_kernel<2, float, float, float, 256, (cudnnGenericOp_t)8, (cudnnNanPropagation_t)0, (cudnnDimOrder_t)0, 1>(cudnnTensorStruct, float*, cudnnTensorStruct, float const*, cudnnTensorStruct, float const*, float, float, float, float, dimArray, reducedDivisorArray, bool) | 0.00 | 0 | 0.00 | 0.00 | 0.00 | 0.00 | NaN | true | |
| 77 | resnext0_stage3_conv2_fwd | Convolution | [1,512,14,14] | 10733.667 | 0 | 0 | 0 | 0 | 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) | 0.00 | 0 | 0.00 | 0.00 | 0.00 | 0.00 | NaN | true | |
| 78 | resnext0_stage3_batchnorm2_fwd | BatchNorm | [1,1024,14,14] | 115 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | void cudnn::detail::bn_fw_inf_1C11_kernel_NCHW<float, float, true, 1>(float, float, cudnnTensorStruct, float const*, cudnnTensorStruct, float*, cudnnTensorStruct, float const*, float const*, float const*, float const*, float) | 0.00 | 0 | 0.00 | 0.00 | 0.00 | 0.00 | NaN | true | |
| 79 | resnext0_stage3_conv3_fwd | Convolution | [1,512,28,28] | 10754.667 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | cudnn::gemm::computeOffsetsKernel(cudnn::gemm::ComputeOffsetsParams) | 0.00 | 0 | 0.00 | 0.00 | 0.00 | 0.00 | NaN | true | |
| 79 | resnext0_stage3_conv3_fwd | Convolution | [1,512,28,28] | 10754.667 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | volta_scudnn_128x64_relu_interior_nn_v1 | 0.00 | 0 | 0.00 | 0.00 | 0.00 | 0.00 | NaN | true | |
| 80 | resnext0_stage3_batchnorm3_fwd | BatchNorm | [1,1024,14,14] | 107 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | void cudnn::detail::bn_fw_inf_1C11_kernel_NCHW<float, float, true, 1>(float, float, cudnnTensorStruct, float const*, cudnnTensorStruct, float*, cudnnTensorStruct, float const*, float const*, float const*, float const*, float) | 0.00 | 0 | 0.00 | 0.00 | 0.00 | 0.00 | NaN | true | |
| 81 | add_resnext0_stage3_activation0 | add_relu | [1,1024,14,14] | 51.333 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | void mxnet::op::mxnet_op::mxnet_generic_kernel<mxnet::op::AddReluKernel, float*, float*, float*, mxnet::OpReqType>(int, float*, float*, float*, mxnet::OpReqType) | 0.00 | 0 | 0.00 | 0.00 | 0.00 | 0.00 | NaN | true | |
| 82 | resnext0_stage3_conv4_fwd | Convolution | [1,1024,14,14] | 10849.667 | 0 | 0 | 0 | 0 | 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) | 0.00 | 0 | 0.00 | 0.00 | 0.00 | 0.00 | NaN | true | |
| 83 | resnext0_stage3_batchnorm4_fwd | BatchNorm | [1,512,14,14] | 89 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | void cudnn::detail::bn_fw_inf_1C11_kernel_NCHW<float, float, true, 1>(float, float, cudnnTensorStruct, float const*, cudnnTensorStruct, float*, cudnnTensorStruct, float const*, float const*, float const*, float const*, float) | 0.00 | 0 | 0.00 | 0.00 | 0.00 | 0.00 | NaN | true | |
| 84 | resnext0_stage3_relu2_fwd | Activation | [1,512,14,14] | 20.667 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | void op_generic_tensor_kernel<2, float, float, float, 256, (cudnnGenericOp_t)8, (cudnnNanPropagation_t)0, (cudnnDimOrder_t)0, 1>(cudnnTensorStruct, float*, cudnnTensorStruct, float const*, cudnnTensorStruct, float const*, float, float, float, float, dimArray, reducedDivisorArray, bool) | 0.00 | 0 | 0.00 | 0.00 | 0.00 | 0.00 | NaN | true | |
| 85 | resnext0_stage3_conv5_fwd | Convolution | [1,512,14,14] | 3488 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | void cudnn::winograd::generateWinogradTilesKernel<0, float, float>(cudnn::winograd::GenerateWinogradTilesParams<float, float>) | 0.00 | 0 | 0.00 | 0.00 | 0.00 | 0.00 | NaN | true | |
| 85 | resnext0_stage3_conv5_fwd | Convolution | [1,512,14,14] | 3488 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | volta_scudnn_winograd_128x128_ldg1_ldg4_relu_tile148t_nt_v1 | 0.00 | 0 | 0.00 | 0.00 | 0.00 | 0.00 | NaN | true | |
| 85 | resnext0_stage3_conv5_fwd | Convolution | [1,512,14,14] | 3488 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | void cudnn::winograd::generateWinogradTilesKernel<0, float, float>(cudnn::winograd::GenerateWinogradTilesParams<float, float>) | 0.00 | 0 | 0.00 | 0.00 | 0.00 | 0.00 | NaN | true | |
| 85 | resnext0_stage3_conv5_fwd | Convolution | [1,512,14,14] | 3488 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | volta_scudnn_winograd_128x128_ldg1_ldg4_relu_tile148t_nt_v1 | 0.00 | 0 | 0.00 | 0.00 | 0.00 | 0.00 | NaN | true | |
| 85 | resnext0_stage3_conv5_fwd | Convolution | [1,512,14,14] | 3488 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | void cudnn::winograd::generateWinogradTilesKernel<0, float, float>(cudnn::winograd::GenerateWinogradTilesParams<float, float>) | 0.00 | 0 | 0.00 | 0.00 | 0.00 | 0.00 | NaN | true | |
| 85 | resnext0_stage3_conv5_fwd | Convolution | [1,512,14,14] | 3488 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | volta_scudnn_winograd_128x128_ldg1_ldg4_relu_tile148t_nt_v1 | 0.00 | 0 | 0.00 | 0.00 | 0.00 | 0.00 | NaN | true | |
| 85 | resnext0_stage3_conv5_fwd | Convolution | [1,512,14,14] | 3488 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | void cudnn::winograd::generateWinogradTilesKernel<0, float, float>(cudnn::winograd::GenerateWinogradTilesParams<float, float>) | 0.00 | 0 | 0.00 | 0.00 | 0.00 | 0.00 | NaN | true | |
| 85 | resnext0_stage3_conv5_fwd | Convolution | [1,512,14,14] | 3488 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | volta_scudnn_winograd_128x128_ldg1_ldg4_relu_tile148t_nt_v1 | 0.00 | 0 | 0.00 | 0.00 | 0.00 | 0.00 | NaN | true | |
| 85 | resnext0_stage3_conv5_fwd | Convolution | [1,512,14,14] | 3488 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | void cudnn::winograd::generateWinogradTilesKernel<0, float, float>(cudnn::winograd::GenerateWinogradTilesParams<float, float>) | 0.00 | 0 | 0.00 | 0.00 | 0.00 | 0.00 | NaN | true | |
| 85 | resnext0_stage3_conv5_fwd | Convolution | [1,512,14,14] | 3488 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | volta_scudnn_winograd_128x128_ldg1_ldg4_relu_tile148t_nt_v1 | 0.00 | 0 | 0.00 | 0.00 | 0.00 | 0.00 | NaN | true | |
| 85 | resnext0_stage3_conv5_fwd | Convolution | [1,512,14,14] | 3488 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | void cudnn::winograd::generateWinogradTilesKernel<0, float, float>(cudnn::winograd::GenerateWinogradTilesParams<float, float>) | 0.00 | 0 | 0.00 | 0.00 | 0.00 | 0.00 | NaN | true | |
| 85 | resnext0_stage3_conv5_fwd | Convolution | [1,512,14,14] | 3488 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | volta_scudnn_winograd_128x128_ldg1_ldg4_relu_tile148t_nt_v1 | 0.00 | 0 | 0.00 | 0.00 | 0.00 | 0.00 | NaN | true | |
| 85 | resnext0_stage3_conv5_fwd | Convolution | [1,512,14,14] | 3488 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | void cudnn::winograd::generateWinogradTilesKernel<0, float, float>(cudnn::winograd::GenerateWinogradTilesParams<float, float>) | 0.00 | 0 | 0.00 | 0.00 | 0.00 | 0.00 | NaN | true | |
| 85 | resnext0_stage3_conv5_fwd | Convolution | [1,512,14,14] | 3488 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | volta_scudnn_winograd_128x128_ldg1_ldg4_relu_tile148t_nt_v1 | 0.00 | 0 | 0.00 | 0.00 | 0.00 | 0.00 | NaN | true | |
| 85 | resnext0_stage3_conv5_fwd | Convolution | [1,512,14,14] | 3488 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | void cudnn::winograd::generateWinogradTilesKernel<0, float, float>(cudnn::winograd::GenerateWinogradTilesParams<float, float>) | 0.00 | 0 | 0.00 | 0.00 | 0.00 | 0.00 | NaN | true | |
| 85 | resnext0_stage3_conv5_fwd | Convolution | [1,512,14,14] | 3488 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | volta_scudnn_winograd_128x128_ldg1_ldg4_relu_tile148t_nt_v1 | 0.00 | 0 | 0.00 | 0.00 | 0.00 | 0.00 | NaN | true | |
| 85 | resnext0_stage3_conv5_fwd | Convolution | [1,512,14,14] | 3488 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | void cudnn::winograd::generateWinogradTilesKernel<0, float, float>(cudnn::winograd::GenerateWinogradTilesParams<float, float>) | 0.00 | 0 | 0.00 | 0.00 | 0.00 | 0.00 | NaN | true | |
| 85 | resnext0_stage3_conv5_fwd | Convolution | [1,512,14,14] | 3488 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | volta_scudnn_winograd_128x128_ldg1_ldg4_relu_tile148t_nt_v1 | 0.00 | 0 | 0.00 | 0.00 | 0.00 | 0.00 | NaN | true | |
| 85 | resnext0_stage3_conv5_fwd | Convolution | [1,512,14,14] | 3488 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | void cudnn::winograd::generateWinogradTilesKernel<0, float, float>(cudnn::winograd::GenerateWinogradTilesParams<float, float>) | 0.00 | 0 | 0.00 | 0.00 | 0.00 | 0.00 | NaN | true | |
| 85 | resnext0_stage3_conv5_fwd | Convolution | [1,512,14,14] | 3488 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | volta_scudnn_winograd_128x128_ldg1_ldg4_relu_tile148t_nt_v1 | 0.00 | 0 | 0.00 | 0.00 | 0.00 | 0.00 | NaN | true | |
| 85 | resnext0_stage3_conv5_fwd | Convolution | [1,512,14,14] | 3488 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | void cudnn::winograd::generateWinogradTilesKernel<0, float, float>(cudnn::winograd::GenerateWinogradTilesParams<float, float>) | 0.00 | 0 | 0.00 | 0.00 | 0.00 | 0.00 | NaN | true | |
| 85 | resnext0_stage3_conv5_fwd | Convolution | [1,512,14,14] | 3488 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | volta_scudnn_winograd_128x128_ldg1_ldg4_relu_tile148t_nt_v1 | 0.00 | 0 | 0.00 | 0.00 | 0.00 | 0.00 | NaN | true | |
| 85 | resnext0_stage3_conv5_fwd | Convolution | [1,512,14,14] | 3488 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | void cudnn::winograd::generateWinogradTilesKernel<0, float, float>(cudnn::winograd::GenerateWinogradTilesParams<float, float>) | 0.00 | 0 | 0.00 | 0.00 | 0.00 | 0.00 | NaN | true | |
| 85 | resnext0_stage3_conv5_fwd | Convolution | [1,512,14,14] | 3488 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | volta_scudnn_winograd_128x128_ldg1_ldg4_relu_tile148t_nt_v1 | 0.00 | 0 | 0.00 | 0.00 | 0.00 | 0.00 | NaN | true | |
| 85 | resnext0_stage3_conv5_fwd | Convolution | [1,512,14,14] | 3488 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | void cudnn::winograd::generateWinogradTilesKernel<0, float, float>(cudnn::winograd::GenerateWinogradTilesParams<float, float>) | 0.00 | 0 | 0.00 | 0.00 | 0.00 | 0.00 | NaN | true | |
| 85 | resnext0_stage3_conv5_fwd | Convolution | [1,512,14,14] | 3488 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | volta_scudnn_winograd_128x128_ldg1_ldg4_relu_tile148t_nt_v1 | 0.00 | 0 | 0.00 | 0.00 | 0.00 | 0.00 | NaN | true | |
| 85 | resnext0_stage3_conv5_fwd | Convolution | [1,512,14,14] | 3488 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | void cudnn::winograd::generateWinogradTilesKernel<0, float, float>(cudnn::winograd::GenerateWinogradTilesParams<float, float>) | 0.00 | 0 | 0.00 | 0.00 | 0.00 | 0.00 | NaN | true | |
| 85 | resnext0_stage3_conv5_fwd | Convolution | [1,512,14,14] | 3488 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | volta_scudnn_winograd_128x128_ldg1_ldg4_relu_tile148t_nt_v1 | 0.00 | 0 | 0.00 | 0.00 | 0.00 | 0.00 | NaN | true | |
| 85 | resnext0_stage3_conv5_fwd | Convolution | [1,512,14,14] | 3488 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | void cudnn::winograd::generateWinogradTilesKernel<0, float, float>(cudnn::winograd::GenerateWinogradTilesParams<float, float>) | 0.00 | 0 | 0.00 | 0.00 | 0.00 | 0.00 | NaN | true | |
| 85 | resnext0_stage3_conv5_fwd | Convolution | [1,512,14,14] | 3488 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | volta_scudnn_winograd_128x128_ldg1_ldg4_relu_tile148t_nt_v1 | 0.00 | 0 | 0.00 | 0.00 | 0.00 | 0.00 | NaN | true | |
| 85 | resnext0_stage3_conv5_fwd | Convolution | [1,512,14,14] | 3488 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | void cudnn::winograd::generateWinogradTilesKernel<0, float, float>(cudnn::winograd::GenerateWinogradTilesParams<float, float>) | 0.00 | 0 | 0.00 | 0.00 | 0.00 | 0.00 | NaN | true | |
| 85 | resnext0_stage3_conv5_fwd | Convolution | [1,512,14,14] | 3488 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | volta_scudnn_winograd_128x128_ldg1_ldg4_relu_tile148t_nt_v1 | 0.00 | 0 | 0.00 | 0.00 | 0.00 | 0.00 | NaN | true | |
| 85 | resnext0_stage3_conv5_fwd | Convolution | [1,512,14,14] | 3488 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | void cudnn::winograd::generateWinogradTilesKernel<0, float, float>(cudnn::winograd::GenerateWinogradTilesParams<float, float>) | 0.00 | 0 | 0.00 | 0.00 | 0.00 | 0.00 | NaN | true | |
| 85 | resnext0_stage3_conv5_fwd | Convolution | [1,512,14,14] | 3488 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | volta_scudnn_winograd_128x128_ldg1_ldg4_relu_tile148t_nt_v1 | 0.00 | 0 | 0.00 | 0.00 | 0.00 | 0.00 | NaN | true | |
| 85 | resnext0_stage3_conv5_fwd | Convolution | [1,512,14,14] | 3488 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | void cudnn::winograd::generateWinogradTilesKernel<0, float, float>(cudnn::winograd::GenerateWinogradTilesParams<float, float>) | 0.00 | 0 | 0.00 | 0.00 | 0.00 | 0.00 | NaN | true | |
| 85 | resnext0_stage3_conv5_fwd | Convolution | [1,512,14,14] | 3488 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | volta_scudnn_winograd_128x128_ldg1_ldg4_relu_tile148t_nt_v1 | 0.00 | 0 | 0.00 | 0.00 | 0.00 | 0.00 | NaN | true | |
| 85 | resnext0_stage3_conv5_fwd | Convolution | [1,512,14,14] | 3488 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | void cudnn::winograd::generateWinogradTilesKernel<0, float, float>(cudnn::winograd::GenerateWinogradTilesParams<float, float>) | 0.00 | 0 | 0.00 | 0.00 | 0.00 | 0.00 | NaN | true | |
| 85 | resnext0_stage3_conv5_fwd | Convolution | [1,512,14,14] | 3488 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | volta_scudnn_winograd_128x128_ldg1_ldg4_relu_tile148t_nt_v1 | 0.00 | 0 | 0.00 | 0.00 | 0.00 | 0.00 | NaN | true | |
| 85 | resnext0_stage3_conv5_fwd | Convolution | [1,512,14,14] | 3488 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | void cudnn::winograd::generateWinogradTilesKernel<0, float, float>(cudnn::winograd::GenerateWinogradTilesParams<float, float>) | 0.00 | 0 | 0.00 | 0.00 | 0.00 | 0.00 | NaN | true | |
| 85 | resnext0_stage3_conv5_fwd | Convolution | [1,512,14,14] | 3488 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | volta_scudnn_winograd_128x128_ldg1_ldg4_relu_tile148t_nt_v1 | 0.00 | 0 | 0.00 | 0.00 | 0.00 | 0.00 | NaN | true | |
| 85 | resnext0_stage3_conv5_fwd | Convolution | [1,512,14,14] | 3488 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | void cudnn::winograd::generateWinogradTilesKernel<0, float, float>(cudnn::winograd::GenerateWinogradTilesParams<float, float>) | 0.00 | 0 | 0.00 | 0.00 | 0.00 | 0.00 | NaN | true | |
| 85 | resnext0_stage3_conv5_fwd | Convolution | [1,512,14,14] | 3488 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | volta_scudnn_winograd_128x128_ldg1_ldg4_relu_tile148t_nt_v1 | 0.00 | 0 | 0.00 | 0.00 | 0.00 | 0.00 | NaN | true | |
| 85 | resnext0_stage3_conv5_fwd | Convolution | [1,512,14,14] | 3488 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | void cudnn::winograd::generateWinogradTilesKernel<0, float, float>(cudnn::winograd::GenerateWinogradTilesParams<float, float>) | 0.00 | 0 | 0.00 | 0.00 | 0.00 | 0.00 | NaN | true | |
| 85 | resnext0_stage3_conv5_fwd | Convolution | [1,512,14,14] | 3488 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | volta_scudnn_winograd_128x128_ldg1_ldg4_relu_tile148t_nt_v1 | 0.00 | 0 | 0.00 | 0.00 | 0.00 | 0.00 | NaN | true | |
| 85 | resnext0_stage3_conv5_fwd | Convolution | [1,512,14,14] | 3488 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | void cudnn::winograd::generateWinogradTilesKernel<0, float, float>(cudnn::winograd::GenerateWinogradTilesParams<float, float>) | 0.00 | 0 | 0.00 | 0.00 | 0.00 | 0.00 | NaN | true | |
| 85 | resnext0_stage3_conv5_fwd | Convolution | [1,512,14,14] | 3488 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | volta_scudnn_winograd_128x128_ldg1_ldg4_relu_tile148t_nt_v1 | 0.00 | 0 | 0.00 | 0.00 | 0.00 | 0.00 | NaN | true | |
| 85 | resnext0_stage3_conv5_fwd | Convolution | [1,512,14,14] | 3488 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | void cudnn::winograd::generateWinogradTilesKernel<0, float, float>(cudnn::winograd::GenerateWinogradTilesParams<float, float>) | 0.00 | 0 | 0.00 | 0.00 | 0.00 | 0.00 | NaN | true | |
| 85 | resnext0_stage3_conv5_fwd | Convolution | [1,512,14,14] | 3488 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | volta_scudnn_winograd_128x128_ldg1_ldg4_relu_tile148t_nt_v1 | 0.00 | 0 | 0.00 | 0.00 | 0.00 | 0.00 | NaN | true | |
| 85 | resnext0_stage3_conv5_fwd | Convolution | [1,512,14,14] | 3488 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | void cudnn::winograd::generateWinogradTilesKernel<0, float, float>(cudnn::winograd::GenerateWinogradTilesParams<float, float>) | 0.00 | 0 | 0.00 | 0.00 | 0.00 | 0.00 | NaN | true | |
| 85 | resnext0_stage3_conv5_fwd | Convolution | [1,512,14,14] | 3488 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | volta_scudnn_winograd_128x128_ldg1_ldg4_relu_tile148t_nt_v1 | 0.00 | 0 | 0.00 | 0.00 | 0.00 | 0.00 | NaN | true | |
| 85 | resnext0_stage3_conv5_fwd | Convolution | [1,512,14,14] | 3488 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | void cudnn::winograd::generateWinogradTilesKernel<0, float, float>(cudnn::winograd::GenerateWinogradTilesParams<float, float>) | 0.00 | 0 | 0.00 | 0.00 | 0.00 | 0.00 | NaN | true | |
| 85 | resnext0_stage3_conv5_fwd | Convolution | [1,512,14,14] | 3488 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | volta_scudnn_winograd_128x128_ldg1_ldg4_relu_tile148t_nt_v1 | 0.00 | 0 | 0.00 | 0.00 | 0.00 | 0.00 | NaN | true | |
| 85 | resnext0_stage3_conv5_fwd | Convolution | [1,512,14,14] | 3488 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | void cudnn::winograd::generateWinogradTilesKernel<0, float, float>(cudnn::winograd::GenerateWinogradTilesParams<float, float>) | 0.00 | 0 | 0.00 | 0.00 | 0.00 | 0.00 | NaN | true | |
| 85 | resnext0_stage3_conv5_fwd | Convolution | [1,512,14,14] | 3488 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | volta_scudnn_winograd_128x128_ldg1_ldg4_relu_tile148t_nt_v1 | 0.00 | 0 | 0.00 | 0.00 | 0.00 | 0.00 | NaN | true | |
| 85 | resnext0_stage3_conv5_fwd | Convolution | [1,512,14,14] | 3488 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | void cudnn::winograd::generateWinogradTilesKernel<0, float, float>(cudnn::winograd::GenerateWinogradTilesParams<float, float>) | 0.00 | 0 | 0.00 | 0.00 | 0.00 | 0.00 | NaN | true | |
| 85 | resnext0_stage3_conv5_fwd | Convolution | [1,512,14,14] | 3488 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | volta_scudnn_winograd_128x128_ldg1_ldg4_relu_tile148t_nt_v1 | 0.00 | 0 | 0.00 | 0.00 | 0.00 | 0.00 | NaN | true | |
| 85 | resnext0_stage3_conv5_fwd | Convolution | [1,512,14,14] | 3488 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | void cudnn::winograd::generateWinogradTilesKernel<0, float, float>(cudnn::winograd::GenerateWinogradTilesParams<float, float>) | 0.00 | 0 | 0.00 | 0.00 | 0.00 | 0.00 | NaN | true | |
| 85 | resnext0_stage3_conv5_fwd | Convolution | [1,512,14,14] | 3488 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | volta_scudnn_winograd_128x128_ldg1_ldg4_relu_tile148t_nt_v1 | 0.00 | 0 | 0.00 | 0.00 | 0.00 | 0.00 | NaN | true | |
| 85 | resnext0_stage3_conv5_fwd | Convolution | [1,512,14,14] | 3488 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | void cudnn::winograd::generateWinogradTilesKernel<0, float, float>(cudnn::winograd::GenerateWinogradTilesParams<float, float>) | 0.00 | 0 | 0.00 | 0.00 | 0.00 | 0.00 | NaN | true | |
| 85 | resnext0_stage3_conv5_fwd | Convolution | [1,512,14,14] | 3488 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | volta_scudnn_winograd_128x128_ldg1_ldg4_relu_tile148t_nt_v1 | 0.00 | 0 | 0.00 | 0.00 | 0.00 | 0.00 | NaN | true | |
| 85 | resnext0_stage3_conv5_fwd | Convolution | [1,512,14,14] | 3488 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | void cudnn::winograd::generateWinogradTilesKernel<0, float, float>(cudnn::winograd::GenerateWinogradTilesParams<float, float>) | 0.00 | 0 | 0.00 | 0.00 | 0.00 | 0.00 | NaN | true | |
| 85 | resnext0_stage3_conv5_fwd | Convolution | [1,512,14,14] | 3488 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | volta_scudnn_winograd_128x128_ldg1_ldg4_relu_tile148t_nt_v1 | 0.00 | 0 | 0.00 | 0.00 | 0.00 | 0.00 | NaN | true | |
| 85 | resnext0_stage3_conv5_fwd | Convolution | [1,512,14,14] | 3488 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | void cudnn::winograd::generateWinogradTilesKernel<0, float, float>(cudnn::winograd::GenerateWinogradTilesParams<float, float>) | 0.00 | 0 | 0.00 | 0.00 | 0.00 | 0.00 | NaN | true | |
| 85 | resnext0_stage3_conv5_fwd | Convolution | [1,512,14,14] | 3488 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | volta_scudnn_winograd_128x128_ldg1_ldg4_relu_tile148t_nt_v1 | 0.00 | 0 | 0.00 | 0.00 | 0.00 | 0.00 | NaN | true | |
| 86 | resnext0_stage3_batchnorm5_fwd | BatchNorm | [1,512,14,14] | 99.333 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | void cudnn::detail::bn_fw_inf_1C11_kernel_NCHW<float, float, true, 1>(float, float, cudnnTensorStruct, float const*, cudnnTensorStruct, float*, cudnnTensorStruct, float const*, float const*, float const*, float const*, float) | 0.00 | 0 | 0.00 | 0.00 | 0.00 | 0.00 | NaN | true | |
| 87 | resnext0_stage3_relu3_fwd | Activation | [1,512,14,14] | 20.667 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | void op_generic_tensor_kernel<2, float, float, float, 256, (cudnnGenericOp_t)8, (cudnnNanPropagation_t)0, (cudnnDimOrder_t)0, 1>(cudnnTensorStruct, float*, cudnnTensorStruct, float const*, cudnnTensorStruct, float const*, float, float, float, float, dimArray, reducedDivisorArray, bool) | 0.00 | 0 | 0.00 | 0.00 | 0.00 | 0.00 | NaN | true | |
| 88 | resnext0_stage3_conv6_fwd | Convolution | [1,512,14,14] | 10685.333 | 0 | 0 | 0 | 0 | 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) | 0.00 | 0 | 0.00 | 0.00 | 0.00 | 0.00 | NaN | true | |
| 89 | resnext0_stage3_batchnorm6_fwd | BatchNorm | [1,1024,14,14] | 112.333 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | void cudnn::detail::bn_fw_inf_1C11_kernel_NCHW<float, float, true, 1>(float, float, cudnnTensorStruct, float const*, cudnnTensorStruct, float*, cudnnTensorStruct, float const*, float const*, float const*, float const*, float) | 0.00 | 0 | 0.00 | 0.00 | 0.00 | 0.00 | NaN | true | |
| 90 | add_resnext0_stage3_activation1 | add_relu | [1,1024,14,14] | 51.333 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | void mxnet::op::mxnet_op::mxnet_generic_kernel<mxnet::op::AddReluKernel, float*, float*, float*, mxnet::OpReqType>(int, float*, float*, float*, mxnet::OpReqType) | 0.00 | 0 | 0.00 | 0.00 | 0.00 | 0.00 | NaN | true | |
| 91 | resnext0_stage3_conv7_fwd | Convolution | [1,1024,14,14] | 10780.333 | 0 | 0 | 0 | 0 | 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) | 0.00 | 0 | 0.00 | 0.00 | 0.00 | 0.00 | NaN | true | |
| 92 | resnext0_stage3_batchnorm7_fwd | BatchNorm | [1,512,14,14] | 98 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | void cudnn::detail::bn_fw_inf_1C11_kernel_NCHW<float, float, true, 1>(float, float, cudnnTensorStruct, float const*, cudnnTensorStruct, float*, cudnnTensorStruct, float const*, float const*, float const*, float const*, float) | 0.00 | 0 | 0.00 | 0.00 | 0.00 | 0.00 | NaN | true | |
| 93 | resnext0_stage3_relu4_fwd | Activation | [1,512,14,14] | 21.667 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | void op_generic_tensor_kernel<2, float, float, float, 256, (cudnnGenericOp_t)8, (cudnnNanPropagation_t)0, (cudnnDimOrder_t)0, 1>(cudnnTensorStruct, float*, cudnnTensorStruct, float const*, cudnnTensorStruct, float const*, float, float, float, float, dimArray, reducedDivisorArray, bool) | 0.00 | 0 | 0.00 | 0.00 | 0.00 | 0.00 | NaN | true | |
| 94 | resnext0_stage3_conv8_fwd | Convolution | [1,512,14,14] | 3493 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | void cudnn::winograd::generateWinogradTilesKernel<0, float, float>(cudnn::winograd::GenerateWinogradTilesParams<float, float>) | 0.00 | 0 | 0.00 | 0.00 | 0.00 | 0.00 | NaN | true | |
| 94 | resnext0_stage3_conv8_fwd | Convolution | [1,512,14,14] | 3493 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | volta_scudnn_winograd_128x128_ldg1_ldg4_relu_tile148t_nt_v1 | 0.00 | 0 | 0.00 | 0.00 | 0.00 | 0.00 | NaN | true | |
| 94 | resnext0_stage3_conv8_fwd | Convolution | [1,512,14,14] | 3493 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | void cudnn::winograd::generateWinogradTilesKernel<0, float, float>(cudnn::winograd::GenerateWinogradTilesParams<float, float>) | 0.00 | 0 | 0.00 | 0.00 | 0.00 | 0.00 | NaN | true | |
| 94 | resnext0_stage3_conv8_fwd | Convolution | [1,512,14,14] | 3493 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | volta_scudnn_winograd_128x128_ldg1_ldg4_relu_tile148t_nt_v1 | 0.00 | 0 | 0.00 | 0.00 | 0.00 | 0.00 | NaN | true | |
| 94 | resnext0_stage3_conv8_fwd | Convolution | [1,512,14,14] | 3493 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | void cudnn::winograd::generateWinogradTilesKernel<0, float, float>(cudnn::winograd::GenerateWinogradTilesParams<float, float>) | 0.00 | 0 | 0.00 | 0.00 | 0.00 | 0.00 | NaN | true | |
| 94 | resnext0_stage3_conv8_fwd | Convolution | [1,512,14,14] | 3493 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | volta_scudnn_winograd_128x128_ldg1_ldg4_relu_tile148t_nt_v1 | 0.00 | 0 | 0.00 | 0.00 | 0.00 | 0.00 | NaN | true | |
| 94 | resnext0_stage3_conv8_fwd | Convolution | [1,512,14,14] | 3493 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | void cudnn::winograd::generateWinogradTilesKernel<0, float, float>(cudnn::winograd::GenerateWinogradTilesParams<float, float>) | 0.00 | 0 | 0.00 | 0.00 | 0.00 | 0.00 | NaN | true | |
| 94 | resnext0_stage3_conv8_fwd | Convolution | [1,512,14,14] | 3493 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | volta_scudnn_winograd_128x128_ldg1_ldg4_relu_tile148t_nt_v1 | 0.00 | 0 | 0.00 | 0.00 | 0.00 | 0.00 | NaN | true | |
| 94 | resnext0_stage3_conv8_fwd | Convolution | [1,512,14,14] | 3493 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | void cudnn::winograd::generateWinogradTilesKernel<0, float, float>(cudnn::winograd::GenerateWinogradTilesParams<float, float>) | 0.00 | 0 | 0.00 | 0.00 | 0.00 | 0.00 | NaN | true | |
| 94 | resnext0_stage3_conv8_fwd | Convolution | [1,512,14,14] | 3493 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | volta_scudnn_winograd_128x128_ldg1_ldg4_relu_tile148t_nt_v1 | 0.00 | 0 | 0.00 | 0.00 | 0.00 | 0.00 | NaN | true | |
| 94 | resnext0_stage3_conv8_fwd | Convolution | [1,512,14,14] | 3493 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | void cudnn::winograd::generateWinogradTilesKernel<0, float, float>(cudnn::winograd::GenerateWinogradTilesParams<float, float>) | 0.00 | 0 | 0.00 | 0.00 | 0.00 | 0.00 | NaN | true | |
| 94 | resnext0_stage3_conv8_fwd | Convolution | [1,512,14,14] | 3493 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | volta_scudnn_winograd_128x128_ldg1_ldg4_relu_tile148t_nt_v1 | 0.00 | 0 | 0.00 | 0.00 | 0.00 | 0.00 | NaN | true | |
| 94 | resnext0_stage3_conv8_fwd | Convolution | [1,512,14,14] | 3493 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | void cudnn::winograd::generateWinogradTilesKernel<0, float, float>(cudnn::winograd::GenerateWinogradTilesParams<float, float>) | 0.00 | 0 | 0.00 | 0.00 | 0.00 | 0.00 | NaN | true | |
| 94 | resnext0_stage3_conv8_fwd | Convolution | [1,512,14,14] | 3493 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | volta_scudnn_winograd_128x128_ldg1_ldg4_relu_tile148t_nt_v1 | 0.00 | 0 | 0.00 | 0.00 | 0.00 | 0.00 | NaN | true | |
| 94 | resnext0_stage3_conv8_fwd | Convolution | [1,512,14,14] | 3493 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | void cudnn::winograd::generateWinogradTilesKernel<0, float, float>(cudnn::winograd::GenerateWinogradTilesParams<float, float>) | 0.00 | 0 | 0.00 | 0.00 | 0.00 | 0.00 | NaN | true | |
| 94 | resnext0_stage3_conv8_fwd | Convolution | [1,512,14,14] | 3493 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | volta_scudnn_winograd_128x128_ldg1_ldg4_relu_tile148t_nt_v1 | 0.00 | 0 | 0.00 | 0.00 | 0.00 | 0.00 | NaN | true | |
| 94 | resnext0_stage3_conv8_fwd | Convolution | [1,512,14,14] | 3493 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | void cudnn::winograd::generateWinogradTilesKernel<0, float, float>(cudnn::winograd::GenerateWinogradTilesParams<float, float>) | 0.00 | 0 | 0.00 | 0.00 | 0.00 | 0.00 | NaN | true | |
| 94 | resnext0_stage3_conv8_fwd | Convolution | [1,512,14,14] | 3493 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | volta_scudnn_winograd_128x128_ldg1_ldg4_relu_tile148t_nt_v1 | 0.00 | 0 | 0.00 | 0.00 | 0.00 | 0.00 | NaN | true | |
| 94 | resnext0_stage3_conv8_fwd | Convolution | [1,512,14,14] | 3493 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | void cudnn::winograd::generateWinogradTilesKernel<0, float, float>(cudnn::winograd::GenerateWinogradTilesParams<float, float>) | 0.00 | 0 | 0.00 | 0.00 | 0.00 | 0.00 | NaN | true | |
| 94 | resnext0_stage3_conv8_fwd | Convolution | [1,512,14,14] | 3493 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | volta_scudnn_winograd_128x128_ldg1_ldg4_relu_tile148t_nt_v1 | 0.00 | 0 | 0.00 | 0.00 | 0.00 | 0.00 | NaN | true | |
| 94 | resnext0_stage3_conv8_fwd | Convolution | [1,512,14,14] | 3493 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | void cudnn::winograd::generateWinogradTilesKernel<0, float, float>(cudnn::winograd::GenerateWinogradTilesParams<float, float>) | 0.00 | 0 | 0.00 | 0.00 | 0.00 | 0.00 | NaN | true | |
| 94 | resnext0_stage3_conv8_fwd | Convolution | [1,512,14,14] | 3493 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | volta_scudnn_winograd_128x128_ldg1_ldg4_relu_tile148t_nt_v1 | 0.00 | 0 | 0.00 | 0.00 | 0.00 | 0.00 | NaN | true | |
| 94 | resnext0_stage3_conv8_fwd | Convolution | [1,512,14,14] | 3493 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | void cudnn::winograd::generateWinogradTilesKernel<0, float, float>(cudnn::winograd::GenerateWinogradTilesParams<float, float>) | 0.00 | 0 | 0.00 | 0.00 | 0.00 | 0.00 | NaN | true | |
| 94 | resnext0_stage3_conv8_fwd | Convolution | [1,512,14,14] | 3493 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | volta_scudnn_winograd_128x128_ldg1_ldg4_relu_tile148t_nt_v1 | 0.00 | 0 | 0.00 | 0.00 | 0.00 | 0.00 | NaN | true | |
| 94 | resnext0_stage3_conv8_fwd | Convolution | [1,512,14,14] | 3493 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | void cudnn::winograd::generateWinogradTilesKernel<0, float, float>(cudnn::winograd::GenerateWinogradTilesParams<float, float>) | 0.00 | 0 | 0.00 | 0.00 | 0.00 | 0.00 | NaN | true | |
| 94 | resnext0_stage3_conv8_fwd | Convolution | [1,512,14,14] | 3493 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | volta_scudnn_winograd_128x128_ldg1_ldg4_relu_tile148t_nt_v1 | 0.00 | 0 | 0.00 | 0.00 | 0.00 | 0.00 | NaN | true | |
| 94 | resnext0_stage3_conv8_fwd | Convolution | [1,512,14,14] | 3493 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | void cudnn::winograd::generateWinogradTilesKernel<0, float, float>(cudnn::winograd::GenerateWinogradTilesParams<float, float>) | 0.00 | 0 | 0.00 | 0.00 | 0.00 | 0.00 | NaN | true | |
| 94 | resnext0_stage3_conv8_fwd | Convolution | [1,512,14,14] | 3493 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | volta_scudnn_winograd_128x128_ldg1_ldg4_relu_tile148t_nt_v1 | 0.00 | 0 | 0.00 | 0.00 | 0.00 | 0.00 | NaN | true | |
| 94 | resnext0_stage3_conv8_fwd | Convolution | [1,512,14,14] | 3493 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | void cudnn::winograd::generateWinogradTilesKernel<0, float, float>(cudnn::winograd::GenerateWinogradTilesParams<float, float>) | 0.00 | 0 | 0.00 | 0.00 | 0.00 | 0.00 | NaN | true | |
| 94 | resnext0_stage3_conv8_fwd | Convolution | [1,512,14,14] | 3493 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | volta_scudnn_winograd_128x128_ldg1_ldg4_relu_tile148t_nt_v1 | 0.00 | 0 | 0.00 | 0.00 | 0.00 | 0.00 | NaN | true | |
| 94 | resnext0_stage3_conv8_fwd | Convolution | [1,512,14,14] | 3493 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | void cudnn::winograd::generateWinogradTilesKernel<0, float, float>(cudnn::winograd::GenerateWinogradTilesParams<float, float>) | 0.00 | 0 | 0.00 | 0.00 | 0.00 | 0.00 | NaN | true | |
| 94 | resnext0_stage3_conv8_fwd | Convolution | [1,512,14,14] | 3493 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | volta_scudnn_winograd_128x128_ldg1_ldg4_relu_tile148t_nt_v1 | 0.00 | 0 | 0.00 | 0.00 | 0.00 | 0.00 | NaN | true | |
| 94 | resnext0_stage3_conv8_fwd | Convolution | [1,512,14,14] | 3493 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | void cudnn::winograd::generateWinogradTilesKernel<0, float, float>(cudnn::winograd::GenerateWinogradTilesParams<float, float>) | 0.00 | 0 | 0.00 | 0.00 | 0.00 | 0.00 | NaN | true | |
| 94 | resnext0_stage3_conv8_fwd | Convolution | [1,512,14,14] | 3493 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | volta_scudnn_winograd_128x128_ldg1_ldg4_relu_tile148t_nt_v1 | 0.00 | 0 | 0.00 | 0.00 | 0.00 | 0.00 | NaN | true | |
| 94 | resnext0_stage3_conv8_fwd | Convolution | [1,512,14,14] | 3493 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | void cudnn::winograd::generateWinogradTilesKernel<0, float, float>(cudnn::winograd::GenerateWinogradTilesParams<float, float>) | 0.00 | 0 | 0.00 | 0.00 | 0.00 | 0.00 | NaN | true | |
| 94 | resnext0_stage3_conv8_fwd | Convolution | [1,512,14,14] | 3493 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | volta_scudnn_winograd_128x128_ldg1_ldg4_relu_tile148t_nt_v1 | 0.00 | 0 | 0.00 | 0.00 | 0.00 | 0.00 | NaN | true | |
| 94 | resnext0_stage3_conv8_fwd | Convolution | [1,512,14,14] | 3493 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | void cudnn::winograd::generateWinogradTilesKernel<0, float, float>(cudnn::winograd::GenerateWinogradTilesParams<float, float>) | 0.00 | 0 | 0.00 | 0.00 | 0.00 | 0.00 | NaN | true | |
| 94 | resnext0_stage3_conv8_fwd | Convolution | [1,512,14,14] | 3493 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | volta_scudnn_winograd_128x128_ldg1_ldg4_relu_tile148t_nt_v1 | 0.00 | 0 | 0.00 | 0.00 | 0.00 | 0.00 | NaN | true | |
| 94 | resnext0_stage3_conv8_fwd | Convolution | [1,512,14,14] | 3493 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | void cudnn::winograd::generateWinogradTilesKernel<0, float, float>(cudnn::winograd::GenerateWinogradTilesParams<float, float>) | 0.00 | 0 | 0.00 | 0.00 | 0.00 | 0.00 | NaN | true | |
| 94 | resnext0_stage3_conv8_fwd | Convolution | [1,512,14,14] | 3493 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | volta_scudnn_winograd_128x128_ldg1_ldg4_relu_tile148t_nt_v1 | 0.00 | 0 | 0.00 | 0.00 | 0.00 | 0.00 | NaN | true | |
| 94 | resnext0_stage3_conv8_fwd | Convolution | [1,512,14,14] | 3493 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | void cudnn::winograd::generateWinogradTilesKernel<0, float, float>(cudnn::winograd::GenerateWinogradTilesParams<float, float>) | 0.00 | 0 | 0.00 | 0.00 | 0.00 | 0.00 | NaN | true | |
| 94 | resnext0_stage3_conv8_fwd | Convolution | [1,512,14,14] | 3493 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | volta_scudnn_winograd_128x128_ldg1_ldg4_relu_tile148t_nt_v1 | 0.00 | 0 | 0.00 | 0.00 | 0.00 | 0.00 | NaN | true | |
| 94 | resnext0_stage3_conv8_fwd | Convolution | [1,512,14,14] | 3493 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | void cudnn::winograd::generateWinogradTilesKernel<0, float, float>(cudnn::winograd::GenerateWinogradTilesParams<float, float>) | 0.00 | 0 | 0.00 | 0.00 | 0.00 | 0.00 | NaN | true | |
| 94 | resnext0_stage3_conv8_fwd | Convolution | [1,512,14,14] | 3493 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | volta_scudnn_winograd_128x128_ldg1_ldg4_relu_tile148t_nt_v1 | 0.00 | 0 | 0.00 | 0.00 | 0.00 | 0.00 | NaN | true | |
| 94 | resnext0_stage3_conv8_fwd | Convolution | [1,512,14,14] | 3493 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | void cudnn::winograd::generateWinogradTilesKernel<0, float, float>(cudnn::winograd::GenerateWinogradTilesParams<float, float>) | 0.00 | 0 | 0.00 | 0.00 | 0.00 | 0.00 | NaN | true | |
| 94 | resnext0_stage3_conv8_fwd | Convolution | [1,512,14,14] | 3493 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | volta_scudnn_winograd_128x128_ldg1_ldg4_relu_tile148t_nt_v1 | 0.00 | 0 | 0.00 | 0.00 | 0.00 | 0.00 | NaN | true | |
| 94 | resnext0_stage3_conv8_fwd | Convolution | [1,512,14,14] | 3493 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | void cudnn::winograd::generateWinogradTilesKernel<0, float, float>(cudnn::winograd::GenerateWinogradTilesParams<float, float>) | 0.00 | 0 | 0.00 | 0.00 | 0.00 | 0.00 | NaN | true | |
| 94 | resnext0_stage3_conv8_fwd | Convolution | [1,512,14,14] | 3493 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | volta_scudnn_winograd_128x128_ldg1_ldg4_relu_tile148t_nt_v1 | 0.00 | 0 | 0.00 | 0.00 | 0.00 | 0.00 | NaN | true | |
| 94 | resnext0_stage3_conv8_fwd | Convolution | [1,512,14,14] | 3493 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | void cudnn::winograd::generateWinogradTilesKernel<0, float, float>(cudnn::winograd::GenerateWinogradTilesParams<float, float>) | 0.00 | 0 | 0.00 | 0.00 | 0.00 | 0.00 | NaN | true | |
| 94 | resnext0_stage3_conv8_fwd | Convolution | [1,512,14,14] | 3493 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | volta_scudnn_winograd_128x128_ldg1_ldg4_relu_tile148t_nt_v1 | 0.00 | 0 | 0.00 | 0.00 | 0.00 | 0.00 | NaN | true | |
| 94 | resnext0_stage3_conv8_fwd | Convolution | [1,512,14,14] | 3493 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | void cudnn::winograd::generateWinogradTilesKernel<0, float, float>(cudnn::winograd::GenerateWinogradTilesParams<float, float>) | 0.00 | 0 | 0.00 | 0.00 | 0.00 | 0.00 | NaN | true | |
| 94 | resnext0_stage3_conv8_fwd | Convolution | [1,512,14,14] | 3493 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | volta_scudnn_winograd_128x128_ldg1_ldg4_relu_tile148t_nt_v1 | 0.00 | 0 | 0.00 | 0.00 | 0.00 | 0.00 | NaN | true | |
| 94 | resnext0_stage3_conv8_fwd | Convolution | [1,512,14,14] | 3493 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | void cudnn::winograd::generateWinogradTilesKernel<0, float, float>(cudnn::winograd::GenerateWinogradTilesParams<float, float>) | 0.00 | 0 | 0.00 | 0.00 | 0.00 | 0.00 | NaN | true | |
| 94 | resnext0_stage3_conv8_fwd | Convolution | [1,512,14,14] | 3493 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | volta_scudnn_winograd_128x128_ldg1_ldg4_relu_tile148t_nt_v1 | 0.00 | 0 | 0.00 | 0.00 | 0.00 | 0.00 | NaN | true | |
| 94 | resnext0_stage3_conv8_fwd | Convolution | [1,512,14,14] | 3493 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | void cudnn::winograd::generateWinogradTilesKernel<0, float, float>(cudnn::winograd::GenerateWinogradTilesParams<float, float>) | 0.00 | 0 | 0.00 | 0.00 | 0.00 | 0.00 | NaN | true | |
| 94 | resnext0_stage3_conv8_fwd | Convolution | [1,512,14,14] | 3493 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | volta_scudnn_winograd_128x128_ldg1_ldg4_relu_tile148t_nt_v1 | 0.00 | 0 | 0.00 | 0.00 | 0.00 | 0.00 | NaN | true | |
| 94 | resnext0_stage3_conv8_fwd | Convolution | [1,512,14,14] | 3493 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | void cudnn::winograd::generateWinogradTilesKernel<0, float, float>(cudnn::winograd::GenerateWinogradTilesParams<float, float>) | 0.00 | 0 | 0.00 | 0.00 | 0.00 | 0.00 | NaN | true | |
| 94 | resnext0_stage3_conv8_fwd | Convolution | [1,512,14,14] | 3493 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | volta_scudnn_winograd_128x128_ldg1_ldg4_relu_tile148t_nt_v1 | 0.00 | 0 | 0.00 | 0.00 | 0.00 | 0.00 | NaN | true | |
| 94 | resnext0_stage3_conv8_fwd | Convolution | [1,512,14,14] | 3493 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | void cudnn::winograd::generateWinogradTilesKernel<0, float, float>(cudnn::winograd::GenerateWinogradTilesParams<float, float>) | 0.00 | 0 | 0.00 | 0.00 | 0.00 | 0.00 | NaN | true | |
| 94 | resnext0_stage3_conv8_fwd | Convolution | [1,512,14,14] | 3493 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | volta_scudnn_winograd_128x128_ldg1_ldg4_relu_tile148t_nt_v1 | 0.00 | 0 | 0.00 | 0.00 | 0.00 | 0.00 | NaN | true | |
| 94 | resnext0_stage3_conv8_fwd | Convolution | [1,512,14,14] | 3493 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | void cudnn::winograd::generateWinogradTilesKernel<0, float, float>(cudnn::winograd::GenerateWinogradTilesParams<float, float>) | 0.00 | 0 | 0.00 | 0.00 | 0.00 | 0.00 | NaN | true | |
| 94 | resnext0_stage3_conv8_fwd | Convolution | [1,512,14,14] | 3493 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | volta_scudnn_winograd_128x128_ldg1_ldg4_relu_tile148t_nt_v1 | 0.00 | 0 | 0.00 | 0.00 | 0.00 | 0.00 | NaN | true | |
| 94 | resnext0_stage3_conv8_fwd | Convolution | [1,512,14,14] | 3493 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | void cudnn::winograd::generateWinogradTilesKernel<0, float, float>(cudnn::winograd::GenerateWinogradTilesParams<float, float>) | 0.00 | 0 | 0.00 | 0.00 | 0.00 | 0.00 | NaN | true | |
| 94 | resnext0_stage3_conv8_fwd | Convolution | [1,512,14,14] | 3493 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | volta_scudnn_winograd_128x128_ldg1_ldg4_relu_tile148t_nt_v1 | 0.00 | 0 | 0.00 | 0.00 | 0.00 | 0.00 | NaN | true | |
| 95 | resnext0_stage3_batchnorm8_fwd | BatchNorm | [1,512,14,14] | 97.667 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | void cudnn::detail::bn_fw_inf_1C11_kernel_NCHW<float, float, true, 1>(float, float, cudnnTensorStruct, float const*, cudnnTensorStruct, float*, cudnnTensorStruct, float const*, float const*, float const*, float const*, float) | 0.00 | 0 | 0.00 | 0.00 | 0.00 | 0.00 | NaN | true | |
| 96 | resnext0_stage3_relu5_fwd | Activation | [1,512,14,14] | 21 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | void op_generic_tensor_kernel<2, float, float, float, 256, (cudnnGenericOp_t)8, (cudnnNanPropagation_t)0, (cudnnDimOrder_t)0, 1>(cudnnTensorStruct, float*, cudnnTensorStruct, float const*, cudnnTensorStruct, float const*, float, float, float, float, dimArray, reducedDivisorArray, bool) | 0.00 | 0 | 0.00 | 0.00 | 0.00 | 0.00 | NaN | true | |
| 97 | resnext0_stage3_conv9_fwd | Convolution | [1,512,14,14] | 10766.667 | 0 | 0 | 0 | 0 | 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) | 0.00 | 0 | 0.00 | 0.00 | 0.00 | 0.00 | NaN | true | |
| 98 | resnext0_stage3_batchnorm9_fwd | BatchNorm | [1,1024,14,14] | 118.667 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | void cudnn::detail::bn_fw_inf_1C11_kernel_NCHW<float, float, true, 1>(float, float, cudnnTensorStruct, float const*, cudnnTensorStruct, float*, cudnnTensorStruct, float const*, float const*, float const*, float const*, float) | 0.00 | 0 | 0.00 | 0.00 | 0.00 | 0.00 | NaN | true | |
| 99 | add_resnext0_stage3_activation2 | add_relu | [1,1024,14,14] | 48.333 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | void mxnet::op::mxnet_op::mxnet_generic_kernel<mxnet::op::AddReluKernel, float*, float*, float*, mxnet::OpReqType>(int, float*, float*, float*, mxnet::OpReqType) | 0.00 | 0 | 0.00 | 0.00 | 0.00 | 0.00 | NaN | true | |
| 100 | resnext0_stage3_conv10_fwd | Convolution | [1,1024,14,14] | 10907.667 | 0 | 0 | 0 | 0 | 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) | 0.00 | 0 | 0.00 | 0.00 | 0.00 | 0.00 | NaN | true | |
| 101 | resnext0_stage3_batchnorm10_fwd | BatchNorm | [1,512,14,14] | 102.333 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | void cudnn::detail::bn_fw_inf_1C11_kernel_NCHW<float, float, true, 1>(float, float, cudnnTensorStruct, float const*, cudnnTensorStruct, float*, cudnnTensorStruct, float const*, float const*, float const*, float const*, float) | 0.00 | 0 | 0.00 | 0.00 | 0.00 | 0.00 | NaN | true | |
| 102 | resnext0_stage3_relu6_fwd | Activation | [1,512,14,14] | 19.667 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | void op_generic_tensor_kernel<2, float, float, float, 256, (cudnnGenericOp_t)8, (cudnnNanPropagation_t)0, (cudnnDimOrder_t)0, 1>(cudnnTensorStruct, float*, cudnnTensorStruct, float const*, cudnnTensorStruct, float const*, float, float, float, float, dimArray, reducedDivisorArray, bool) | 0.00 | 0 | 0.00 | 0.00 | 0.00 | 0.00 | NaN | true | |
| 103 | resnext0_stage3_conv11_fwd | Convolution | [1,512,14,14] | 3426.333 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | void cudnn::winograd::generateWinogradTilesKernel<0, float, float>(cudnn::winograd::GenerateWinogradTilesParams<float, float>) | 0.00 | 0 | 0.00 | 0.00 | 0.00 | 0.00 | NaN | true | |
| 103 | resnext0_stage3_conv11_fwd | Convolution | [1,512,14,14] | 3426.333 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | volta_scudnn_winograd_128x128_ldg1_ldg4_relu_tile148t_nt_v1 | 0.00 | 0 | 0.00 | 0.00 | 0.00 | 0.00 | NaN | true | |
| 103 | resnext0_stage3_conv11_fwd | Convolution | [1,512,14,14] | 3426.333 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | void cudnn::winograd::generateWinogradTilesKernel<0, float, float>(cudnn::winograd::GenerateWinogradTilesParams<float, float>) | 0.00 | 0 | 0.00 | 0.00 | 0.00 | 0.00 | NaN | true | |
| 103 | resnext0_stage3_conv11_fwd | Convolution | [1,512,14,14] | 3426.333 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | volta_scudnn_winograd_128x128_ldg1_ldg4_relu_tile148t_nt_v1 | 0.00 | 0 | 0.00 | 0.00 | 0.00 | 0.00 | NaN | true | |
| 103 | resnext0_stage3_conv11_fwd | Convolution | [1,512,14,14] | 3426.333 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | void cudnn::winograd::generateWinogradTilesKernel<0, float, float>(cudnn::winograd::GenerateWinogradTilesParams<float, float>) | 0.00 | 0 | 0.00 | 0.00 | 0.00 | 0.00 | NaN | true | |
| 103 | resnext0_stage3_conv11_fwd | Convolution | [1,512,14,14] | 3426.333 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | volta_scudnn_winograd_128x128_ldg1_ldg4_relu_tile148t_nt_v1 | 0.00 | 0 | 0.00 | 0.00 | 0.00 | 0.00 | NaN | true | |
| 103 | resnext0_stage3_conv11_fwd | Convolution | [1,512,14,14] | 3426.333 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | void cudnn::winograd::generateWinogradTilesKernel<0, float, float>(cudnn::winograd::GenerateWinogradTilesParams<float, float>) | 0.00 | 0 | 0.00 | 0.00 | 0.00 | 0.00 | NaN | true | |
| 103 | resnext0_stage3_conv11_fwd | Convolution | [1,512,14,14] | 3426.333 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | volta_scudnn_winograd_128x128_ldg1_ldg4_relu_tile148t_nt_v1 | 0.00 | 0 | 0.00 | 0.00 | 0.00 | 0.00 | NaN | true | |
| 103 | resnext0_stage3_conv11_fwd | Convolution | [1,512,14,14] | 3426.333 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | void cudnn::winograd::generateWinogradTilesKernel<0, float, float>(cudnn::winograd::GenerateWinogradTilesParams<float, float>) | 0.00 | 0 | 0.00 | 0.00 | 0.00 | 0.00 | NaN | true | |
| 103 | resnext0_stage3_conv11_fwd | Convolution | [1,512,14,14] | 3426.333 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | volta_scudnn_winograd_128x128_ldg1_ldg4_relu_tile148t_nt_v1 | 0.00 | 0 | 0.00 | 0.00 | 0.00 | 0.00 | NaN | true | |
| 103 | resnext0_stage3_conv11_fwd | Convolution | [1,512,14,14] | 3426.333 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | void cudnn::winograd::generateWinogradTilesKernel<0, float, float>(cudnn::winograd::GenerateWinogradTilesParams<float, float>) | 0.00 | 0 | 0.00 | 0.00 | 0.00 | 0.00 | NaN | true | |
| 103 | resnext0_stage3_conv11_fwd | Convolution | [1,512,14,14] | 3426.333 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | volta_scudnn_winograd_128x128_ldg1_ldg4_relu_tile148t_nt_v1 | 0.00 | 0 | 0.00 | 0.00 | 0.00 | 0.00 | NaN | true | |
| 103 | resnext0_stage3_conv11_fwd | Convolution | [1,512,14,14] | 3426.333 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | void cudnn::winograd::generateWinogradTilesKernel<0, float, float>(cudnn::winograd::GenerateWinogradTilesParams<float, float>) | 0.00 | 0 | 0.00 | 0.00 | 0.00 | 0.00 | NaN | true | |
| 103 | resnext0_stage3_conv11_fwd | Convolution | [1,512,14,14] | 3426.333 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | volta_scudnn_winograd_128x128_ldg1_ldg4_relu_tile148t_nt_v1 | 0.00 | 0 | 0.00 | 0.00 | 0.00 | 0.00 | NaN | true | |
| 103 | resnext0_stage3_conv11_fwd | Convolution | [1,512,14,14] | 3426.333 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | void cudnn::winograd::generateWinogradTilesKernel<0, float, float>(cudnn::winograd::GenerateWinogradTilesParams<float, float>) | 0.00 | 0 | 0.00 | 0.00 | 0.00 | 0.00 | NaN | true | |
| 103 | resnext0_stage3_conv11_fwd | Convolution | [1,512,14,14] | 3426.333 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | volta_scudnn_winograd_128x128_ldg1_ldg4_relu_tile148t_nt_v1 | 0.00 | 0 | 0.00 | 0.00 | 0.00 | 0.00 | NaN | true | |
| 103 | resnext0_stage3_conv11_fwd | Convolution | [1,512,14,14] | 3426.333 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | void cudnn::winograd::generateWinogradTilesKernel<0, float, float>(cudnn::winograd::GenerateWinogradTilesParams<float, float>) | 0.00 | 0 | 0.00 | 0.00 | 0.00 | 0.00 | NaN | true | |
| 103 | resnext0_stage3_conv11_fwd | Convolution | [1,512,14,14] | 3426.333 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | volta_scudnn_winograd_128x128_ldg1_ldg4_relu_tile148t_nt_v1 | 0.00 | 0 | 0.00 | 0.00 | 0.00 | 0.00 | NaN | true | |
| 103 | resnext0_stage3_conv11_fwd | Convolution | [1,512,14,14] | 3426.333 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | void cudnn::winograd::generateWinogradTilesKernel<0, float, float>(cudnn::winograd::GenerateWinogradTilesParams<float, float>) | 0.00 | 0 | 0.00 | 0.00 | 0.00 | 0.00 | NaN | true | |
| 103 | resnext0_stage3_conv11_fwd | Convolution | [1,512,14,14] | 3426.333 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | volta_scudnn_winograd_128x128_ldg1_ldg4_relu_tile148t_nt_v1 | 0.00 | 0 | 0.00 | 0.00 | 0.00 | 0.00 | NaN | true | |
| 103 | resnext0_stage3_conv11_fwd | Convolution | [1,512,14,14] | 3426.333 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | void cudnn::winograd::generateWinogradTilesKernel<0, float, float>(cudnn::winograd::GenerateWinogradTilesParams<float, float>) | 0.00 | 0 | 0.00 | 0.00 | 0.00 | 0.00 | NaN | true | |
| 103 | resnext0_stage3_conv11_fwd | Convolution | [1,512,14,14] | 3426.333 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | volta_scudnn_winograd_128x128_ldg1_ldg4_relu_tile148t_nt_v1 | 0.00 | 0 | 0.00 | 0.00 | 0.00 | 0.00 | NaN | true | |
| 103 | resnext0_stage3_conv11_fwd | Convolution | [1,512,14,14] | 3426.333 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | void cudnn::winograd::generateWinogradTilesKernel<0, float, float>(cudnn::winograd::GenerateWinogradTilesParams<float, float>) | 0.00 | 0 | 0.00 | 0.00 | 0.00 | 0.00 | NaN | true | |
| 103 | resnext0_stage3_conv11_fwd | Convolution | [1,512,14,14] | 3426.333 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | volta_scudnn_winograd_128x128_ldg1_ldg4_relu_tile148t_nt_v1 | 0.00 | 0 | 0.00 | 0.00 | 0.00 | 0.00 | NaN | true | |
| 103 | resnext0_stage3_conv11_fwd | Convolution | [1,512,14,14] | 3426.333 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | void cudnn::winograd::generateWinogradTilesKernel<0, float, float>(cudnn::winograd::GenerateWinogradTilesParams<float, float>) | 0.00 | 0 | 0.00 | 0.00 | 0.00 | 0.00 | NaN | true | |
| 103 | resnext0_stage3_conv11_fwd | Convolution | [1,512,14,14] | 3426.333 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | volta_scudnn_winograd_128x128_ldg1_ldg4_relu_tile148t_nt_v1 | 0.00 | 0 | 0.00 | 0.00 | 0.00 | 0.00 | NaN | true | |
| 103 | resnext0_stage3_conv11_fwd | Convolution | [1,512,14,14] | 3426.333 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | void cudnn::winograd::generateWinogradTilesKernel<0, float, float>(cudnn::winograd::GenerateWinogradTilesParams<float, float>) | 0.00 | 0 | 0.00 | 0.00 | 0.00 | 0.00 | NaN | true | |
| 103 | resnext0_stage3_conv11_fwd | Convolution | [1,512,14,14] | 3426.333 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | volta_scudnn_winograd_128x128_ldg1_ldg4_relu_tile148t_nt_v1 | 0.00 | 0 | 0.00 | 0.00 | 0.00 | 0.00 | NaN | true | |
| 103 | resnext0_stage3_conv11_fwd | Convolution | [1,512,14,14] | 3426.333 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | void cudnn::winograd::generateWinogradTilesKernel<0, float, float>(cudnn::winograd::GenerateWinogradTilesParams<float, float>) | 0.00 | 0 | 0.00 | 0.00 | 0.00 | 0.00 | NaN | true | |
| 103 | resnext0_stage3_conv11_fwd | Convolution | [1,512,14,14] | 3426.333 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | volta_scudnn_winograd_128x128_ldg1_ldg4_relu_tile148t_nt_v1 | 0.00 | 0 | 0.00 | 0.00 | 0.00 | 0.00 | NaN | true | |
| 103 | resnext0_stage3_conv11_fwd | Convolution | [1,512,14,14] | 3426.333 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | void cudnn::winograd::generateWinogradTilesKernel<0, float, float>(cudnn::winograd::GenerateWinogradTilesParams<float, float>) | 0.00 | 0 | 0.00 | 0.00 | 0.00 | 0.00 | NaN | true | |
| 103 | resnext0_stage3_conv11_fwd | Convolution | [1,512,14,14] | 3426.333 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | volta_scudnn_winograd_128x128_ldg1_ldg4_relu_tile148t_nt_v1 | 0.00 | 0 | 0.00 | 0.00 | 0.00 | 0.00 | NaN | true | |
| 103 | resnext0_stage3_conv11_fwd | Convolution | [1,512,14,14] | 3426.333 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | void cudnn::winograd::generateWinogradTilesKernel<0, float, float>(cudnn::winograd::GenerateWinogradTilesParams<float, float>) | 0.00 | 0 | 0.00 | 0.00 | 0.00 | 0.00 | NaN | true | |
| 103 | resnext0_stage3_conv11_fwd | Convolution | [1,512,14,14] | 3426.333 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | volta_scudnn_winograd_128x128_ldg1_ldg4_relu_tile148t_nt_v1 | 0.00 | 0 | 0.00 | 0.00 | 0.00 | 0.00 | NaN | true | |
| 103 | resnext0_stage3_conv11_fwd | Convolution | [1,512,14,14] | 3426.333 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | void cudnn::winograd::generateWinogradTilesKernel<0, float, float>(cudnn::winograd::GenerateWinogradTilesParams<float, float>) | 0.00 | 0 | 0.00 | 0.00 | 0.00 | 0.00 | NaN | true | |
| 103 | resnext0_stage3_conv11_fwd | Convolution | [1,512,14,14] | 3426.333 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | volta_scudnn_winograd_128x128_ldg1_ldg4_relu_tile148t_nt_v1 | 0.00 | 0 | 0.00 | 0.00 | 0.00 | 0.00 | NaN | true | |
| 103 | resnext0_stage3_conv11_fwd | Convolution | [1,512,14,14] | 3426.333 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | void cudnn::winograd::generateWinogradTilesKernel<0, float, float>(cudnn::winograd::GenerateWinogradTilesParams<float, float>) | 0.00 | 0 | 0.00 | 0.00 | 0.00 | 0.00 | NaN | true | |
| 103 | resnext0_stage3_conv11_fwd | Convolution | [1,512,14,14] | 3426.333 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | volta_scudnn_winograd_128x128_ldg1_ldg4_relu_tile148t_nt_v1 | 0.00 | 0 | 0.00 | 0.00 | 0.00 | 0.00 | NaN | true | |
| 103 | resnext0_stage3_conv11_fwd | Convolution | [1,512,14,14] | 3426.333 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | void cudnn::winograd::generateWinogradTilesKernel<0, float, float>(cudnn::winograd::GenerateWinogradTilesParams<float, float>) | 0.00 | 0 | 0.00 | 0.00 | 0.00 | 0.00 | NaN | true | |
| 103 | resnext0_stage3_conv11_fwd | Convolution | [1,512,14,14] | 3426.333 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | volta_scudnn_winograd_128x128_ldg1_ldg4_relu_tile148t_nt_v1 | 0.00 | 0 | 0.00 | 0.00 | 0.00 | 0.00 | NaN | true | |
| 103 | resnext0_stage3_conv11_fwd | Convolution | [1,512,14,14] | 3426.333 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | void cudnn::winograd::generateWinogradTilesKernel<0, float, float>(cudnn::winograd::GenerateWinogradTilesParams<float, float>) | 0.00 | 0 | 0.00 | 0.00 | 0.00 | 0.00 | NaN | true | |
| 103 | resnext0_stage3_conv11_fwd | Convolution | [1,512,14,14] | 3426.333 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | volta_scudnn_winograd_128x128_ldg1_ldg4_relu_tile148t_nt_v1 | 0.00 | 0 | 0.00 | 0.00 | 0.00 | 0.00 | NaN | true | |
| 103 | resnext0_stage3_conv11_fwd | Convolution | [1,512,14,14] | 3426.333 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | void cudnn::winograd::generateWinogradTilesKernel<0, float, float>(cudnn::winograd::GenerateWinogradTilesParams<float, float>) | 0.00 | 0 | 0.00 | 0.00 | 0.00 | 0.00 | NaN | true | |
| 103 | resnext0_stage3_conv11_fwd | Convolution | [1,512,14,14] | 3426.333 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | volta_scudnn_winograd_128x128_ldg1_ldg4_relu_tile148t_nt_v1 | 0.00 | 0 | 0.00 | 0.00 | 0.00 | 0.00 | NaN | true | |
| 103 | resnext0_stage3_conv11_fwd | Convolution | [1,512,14,14] | 3426.333 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | void cudnn::winograd::generateWinogradTilesKernel<0, float, float>(cudnn::winograd::GenerateWinogradTilesParams<float, float>) | 0.00 | 0 | 0.00 | 0.00 | 0.00 | 0.00 | NaN | true | |
| 103 | resnext0_stage3_conv11_fwd | Convolution | [1,512,14,14] | 3426.333 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | volta_scudnn_winograd_128x128_ldg1_ldg4_relu_tile148t_nt_v1 | 0.00 | 0 | 0.00 | 0.00 | 0.00 | 0.00 | NaN | true | |
| 103 | resnext0_stage3_conv11_fwd | Convolution | [1,512,14,14] | 3426.333 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | void cudnn::winograd::generateWinogradTilesKernel<0, float, float>(cudnn::winograd::GenerateWinogradTilesParams<float, float>) | 0.00 | 0 | 0.00 | 0.00 | 0.00 | 0.00 | NaN | true | |
| 103 | resnext0_stage3_conv11_fwd | Convolution | [1,512,14,14] | 3426.333 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | volta_scudnn_winograd_128x128_ldg1_ldg4_relu_tile148t_nt_v1 | 0.00 | 0 | 0.00 | 0.00 | 0.00 | 0.00 | NaN | true | |
| 103 | resnext0_stage3_conv11_fwd | Convolution | [1,512,14,14] | 3426.333 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | void cudnn::winograd::generateWinogradTilesKernel<0, float, float>(cudnn::winograd::GenerateWinogradTilesParams<float, float>) | 0.00 | 0 | 0.00 | 0.00 | 0.00 | 0.00 | NaN | true | |
| 103 | resnext0_stage3_conv11_fwd | Convolution | [1,512,14,14] | 3426.333 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | volta_scudnn_winograd_128x128_ldg1_ldg4_relu_tile148t_nt_v1 | 0.00 | 0 | 0.00 | 0.00 | 0.00 | 0.00 | NaN | true | |
| 103 | resnext0_stage3_conv11_fwd | Convolution | [1,512,14,14] | 3426.333 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | void cudnn::winograd::generateWinogradTilesKernel<0, float, float>(cudnn::winograd::GenerateWinogradTilesParams<float, float>) | 0.00 | 0 | 0.00 | 0.00 | 0.00 | 0.00 | NaN | true | |
| 103 | resnext0_stage3_conv11_fwd | Convolution | [1,512,14,14] | 3426.333 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | volta_scudnn_winograd_128x128_ldg1_ldg4_relu_tile148t_nt_v1 | 0.00 | 0 | 0.00 | 0.00 | 0.00 | 0.00 | NaN | true | |
| 103 | resnext0_stage3_conv11_fwd | Convolution | [1,512,14,14] | 3426.333 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | void cudnn::winograd::generateWinogradTilesKernel<0, float, float>(cudnn::winograd::GenerateWinogradTilesParams<float, float>) | 0.00 | 0 | 0.00 | 0.00 | 0.00 | 0.00 | NaN | true | |
| 103 | resnext0_stage3_conv11_fwd | Convolution | [1,512,14,14] | 3426.333 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | volta_scudnn_winograd_128x128_ldg1_ldg4_relu_tile148t_nt_v1 | 0.00 | 0 | 0.00 | 0.00 | 0.00 | 0.00 | NaN | true | |
| 103 | resnext0_stage3_conv11_fwd | Convolution | [1,512,14,14] | 3426.333 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | void cudnn::winograd::generateWinogradTilesKernel<0, float, float>(cudnn::winograd::GenerateWinogradTilesParams<float, float>) | 0.00 | 0 | 0.00 | 0.00 | 0.00 | 0.00 | NaN | true | |
| 103 | resnext0_stage3_conv11_fwd | Convolution | [1,512,14,14] | 3426.333 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | volta_scudnn_winograd_128x128_ldg1_ldg4_relu_tile148t_nt_v1 | 0.00 | 0 | 0.00 | 0.00 | 0.00 | 0.00 | NaN | true | |
| 103 | resnext0_stage3_conv11_fwd | Convolution | [1,512,14,14] | 3426.333 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | void cudnn::winograd::generateWinogradTilesKernel<0, float, float>(cudnn::winograd::GenerateWinogradTilesParams<float, float>) | 0.00 | 0 | 0.00 | 0.00 | 0.00 | 0.00 | NaN | true | |
| 103 | resnext0_stage3_conv11_fwd | Convolution | [1,512,14,14] | 3426.333 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | volta_scudnn_winograd_128x128_ldg1_ldg4_relu_tile148t_nt_v1 | 0.00 | 0 | 0.00 | 0.00 | 0.00 | 0.00 | NaN | true | |
| 103 | resnext0_stage3_conv11_fwd | Convolution | [1,512,14,14] | 3426.333 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | void cudnn::winograd::generateWinogradTilesKernel<0, float, float>(cudnn::winograd::GenerateWinogradTilesParams<float, float>) | 0.00 | 0 | 0.00 | 0.00 | 0.00 | 0.00 | NaN | true | |
| 103 | resnext0_stage3_conv11_fwd | Convolution | [1,512,14,14] | 3426.333 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | volta_scudnn_winograd_128x128_ldg1_ldg4_relu_tile148t_nt_v1 | 0.00 | 0 | 0.00 | 0.00 | 0.00 | 0.00 | NaN | true | |
| 103 | resnext0_stage3_conv11_fwd | Convolution | [1,512,14,14] | 3426.333 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | void cudnn::winograd::generateWinogradTilesKernel<0, float, float>(cudnn::winograd::GenerateWinogradTilesParams<float, float>) | 0.00 | 0 | 0.00 | 0.00 | 0.00 | 0.00 | NaN | true | |
| 103 | resnext0_stage3_conv11_fwd | Convolution | [1,512,14,14] | 3426.333 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | volta_scudnn_winograd_128x128_ldg1_ldg4_relu_tile148t_nt_v1 | 0.00 | 0 | 0.00 | 0.00 | 0.00 | 0.00 | NaN | true | |
| 103 | resnext0_stage3_conv11_fwd | Convolution | [1,512,14,14] | 3426.333 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | void cudnn::winograd::generateWinogradTilesKernel<0, float, float>(cudnn::winograd::GenerateWinogradTilesParams<float, float>) | 0.00 | 0 | 0.00 | 0.00 | 0.00 | 0.00 | NaN | true | |
| 103 | resnext0_stage3_conv11_fwd | Convolution | [1,512,14,14] | 3426.333 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | volta_scudnn_winograd_128x128_ldg1_ldg4_relu_tile148t_nt_v1 | 0.00 | 0 | 0.00 | 0.00 | 0.00 | 0.00 | NaN | true | |
| 104 | resnext0_stage3_batchnorm11_fwd | BatchNorm | [1,512,14,14] | 84 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | void cudnn::detail::bn_fw_inf_1C11_kernel_NCHW<float, float, true, 1>(float, float, cudnnTensorStruct, float const*, cudnnTensorStruct, float*, cudnnTensorStruct, float const*, float const*, float const*, float const*, float) | 0.00 | 0 | 0.00 | 0.00 | 0.00 | 0.00 | NaN | true | |
| 105 | resnext0_stage3_relu7_fwd | Activation | [1,512,14,14] | 21 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | void op_generic_tensor_kernel<2, float, float, float, 256, (cudnnGenericOp_t)8, (cudnnNanPropagation_t)0, (cudnnDimOrder_t)0, 1>(cudnnTensorStruct, float*, cudnnTensorStruct, float const*, cudnnTensorStruct, float const*, float, float, float, float, dimArray, reducedDivisorArray, bool) | 0.00 | 0 | 0.00 | 0.00 | 0.00 | 0.00 | NaN | true | |
| 106 | resnext0_stage3_conv12_fwd | Convolution | [1,512,14,14] | 10548.333 | 0 | 0 | 0 | 0 | 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) | 0.00 | 0 | 0.00 | 0.00 | 0.00 | 0.00 | NaN | true | |
| 107 | resnext0_stage3_batchnorm12_fwd | BatchNorm | [1,1024,14,14] | 110 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | void cudnn::detail::bn_fw_inf_1C11_kernel_NCHW<float, float, true, 1>(float, float, cudnnTensorStruct, float const*, cudnnTensorStruct, float*, cudnnTensorStruct, float const*, float const*, float const*, float const*, float) | 0.00 | 0 | 0.00 | 0.00 | 0.00 | 0.00 | NaN | true | |
| 108 | add_resnext0_stage3_activation3 | add_relu | [1,1024,14,14] | 49 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | void mxnet::op::mxnet_op::mxnet_generic_kernel<mxnet::op::AddReluKernel, float*, float*, float*, mxnet::OpReqType>(int, float*, float*, float*, mxnet::OpReqType) | 0.00 | 0 | 0.00 | 0.00 | 0.00 | 0.00 | NaN | true | |
| 109 | resnext0_stage3_conv13_fwd | Convolution | [1,1024,14,14] | 10828 | 0 | 0 | 0 | 0 | 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) | 0.00 | 0 | 0.00 | 0.00 | 0.00 | 0.00 | NaN | true | |
| 110 | resnext0_stage3_batchnorm13_fwd | BatchNorm | [1,512,14,14] | 98 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | void cudnn::detail::bn_fw_inf_1C11_kernel_NCHW<float, float, true, 1>(float, float, cudnnTensorStruct, float const*, cudnnTensorStruct, float*, cudnnTensorStruct, float const*, float const*, float const*, float const*, float) | 0.00 | 0 | 0.00 | 0.00 | 0.00 | 0.00 | NaN | true | |
| 111 | resnext0_stage3_relu8_fwd | Activation | [1,512,14,14] | 21.333 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | void op_generic_tensor_kernel<2, float, float, float, 256, (cudnnGenericOp_t)8, (cudnnNanPropagation_t)0, (cudnnDimOrder_t)0, 1>(cudnnTensorStruct, float*, cudnnTensorStruct, float const*, cudnnTensorStruct, float const*, float, float, float, float, dimArray, reducedDivisorArray, bool) | 0.00 | 0 | 0.00 | 0.00 | 0.00 | 0.00 | NaN | true | |
| 112 | resnext0_stage3_conv14_fwd | Convolution | [1,512,14,14] | 3473 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | void cudnn::winograd::generateWinogradTilesKernel<0, float, float>(cudnn::winograd::GenerateWinogradTilesParams<float, float>) | 0.00 | 0 | 0.00 | 0.00 | 0.00 | 0.00 | NaN | true | |
| 112 | resnext0_stage3_conv14_fwd | Convolution | [1,512,14,14] | 3473 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | volta_scudnn_winograd_128x128_ldg1_ldg4_relu_tile148t_nt_v1 | 0.00 | 0 | 0.00 | 0.00 | 0.00 | 0.00 | NaN | true | |
| 112 | resnext0_stage3_conv14_fwd | Convolution | [1,512,14,14] | 3473 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | void cudnn::winograd::generateWinogradTilesKernel<0, float, float>(cudnn::winograd::GenerateWinogradTilesParams<float, float>) | 0.00 | 0 | 0.00 | 0.00 | 0.00 | 0.00 | NaN | true | |
| 112 | resnext0_stage3_conv14_fwd | Convolution | [1,512,14,14] | 3473 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | volta_scudnn_winograd_128x128_ldg1_ldg4_relu_tile148t_nt_v1 | 0.00 | 0 | 0.00 | 0.00 | 0.00 | 0.00 | NaN | true | |
| 112 | resnext0_stage3_conv14_fwd | Convolution | [1,512,14,14] | 3473 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | void cudnn::winograd::generateWinogradTilesKernel<0, float, float>(cudnn::winograd::GenerateWinogradTilesParams<float, float>) | 0.00 | 0 | 0.00 | 0.00 | 0.00 | 0.00 | NaN | true | |
| 112 | resnext0_stage3_conv14_fwd | Convolution | [1,512,14,14] | 3473 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | volta_scudnn_winograd_128x128_ldg1_ldg4_relu_tile148t_nt_v1 | 0.00 | 0 | 0.00 | 0.00 | 0.00 | 0.00 | NaN | true | |
| 112 | resnext0_stage3_conv14_fwd | Convolution | [1,512,14,14] | 3473 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | void cudnn::winograd::generateWinogradTilesKernel<0, float, float>(cudnn::winograd::GenerateWinogradTilesParams<float, float>) | 0.00 | 0 | 0.00 | 0.00 | 0.00 | 0.00 | NaN | true | |
| 112 | resnext0_stage3_conv14_fwd | Convolution | [1,512,14,14] | 3473 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | volta_scudnn_winograd_128x128_ldg1_ldg4_relu_tile148t_nt_v1 | 0.00 | 0 | 0.00 | 0.00 | 0.00 | 0.00 | NaN | true | |
| 112 | resnext0_stage3_conv14_fwd | Convolution | [1,512,14,14] | 3473 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | void cudnn::winograd::generateWinogradTilesKernel<0, float, float>(cudnn::winograd::GenerateWinogradTilesParams<float, float>) | 0.00 | 0 | 0.00 | 0.00 | 0.00 | 0.00 | NaN | true | |
| 112 | resnext0_stage3_conv14_fwd | Convolution | [1,512,14,14] | 3473 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | volta_scudnn_winograd_128x128_ldg1_ldg4_relu_tile148t_nt_v1 | 0.00 | 0 | 0.00 | 0.00 | 0.00 | 0.00 | NaN | true | |
| 112 | resnext0_stage3_conv14_fwd | Convolution | [1,512,14,14] | 3473 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | void cudnn::winograd::generateWinogradTilesKernel<0, float, float>(cudnn::winograd::GenerateWinogradTilesParams<float, float>) | 0.00 | 0 | 0.00 | 0.00 | 0.00 | 0.00 | NaN | true | |
| 112 | resnext0_stage3_conv14_fwd | Convolution | [1,512,14,14] | 3473 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | volta_scudnn_winograd_128x128_ldg1_ldg4_relu_tile148t_nt_v1 | 0.00 | 0 | 0.00 | 0.00 | 0.00 | 0.00 | NaN | true | |
| 112 | resnext0_stage3_conv14_fwd | Convolution | [1,512,14,14] | 3473 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | void cudnn::winograd::generateWinogradTilesKernel<0, float, float>(cudnn::winograd::GenerateWinogradTilesParams<float, float>) | 0.00 | 0 | 0.00 | 0.00 | 0.00 | 0.00 | NaN | true | |
| 112 | resnext0_stage3_conv14_fwd | Convolution | [1,512,14,14] | 3473 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | volta_scudnn_winograd_128x128_ldg1_ldg4_relu_tile148t_nt_v1 | 0.00 | 0 | 0.00 | 0.00 | 0.00 | 0.00 | NaN | true | |
| 112 | resnext0_stage3_conv14_fwd | Convolution | [1,512,14,14] | 3473 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | void cudnn::winograd::generateWinogradTilesKernel<0, float, float>(cudnn::winograd::GenerateWinogradTilesParams<float, float>) | 0.00 | 0 | 0.00 | 0.00 | 0.00 | 0.00 | NaN | true | |
| 112 | resnext0_stage3_conv14_fwd | Convolution | [1,512,14,14] | 3473 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | volta_scudnn_winograd_128x128_ldg1_ldg4_relu_tile148t_nt_v1 | 0.00 | 0 | 0.00 | 0.00 | 0.00 | 0.00 | NaN | true | |
| 112 | resnext0_stage3_conv14_fwd | Convolution | [1,512,14,14] | 3473 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | void cudnn::winograd::generateWinogradTilesKernel<0, float, float>(cudnn::winograd::GenerateWinogradTilesParams<float, float>) | 0.00 | 0 | 0.00 | 0.00 | 0.00 | 0.00 | NaN | true | |
| 112 | resnext0_stage3_conv14_fwd | Convolution | [1,512,14,14] | 3473 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | volta_scudnn_winograd_128x128_ldg1_ldg4_relu_tile148t_nt_v1 | 0.00 | 0 | 0.00 | 0.00 | 0.00 | 0.00 | NaN | true | |
| 112 | resnext0_stage3_conv14_fwd | Convolution | [1,512,14,14] | 3473 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | void cudnn::winograd::generateWinogradTilesKernel<0, float, float>(cudnn::winograd::GenerateWinogradTilesParams<float, float>) | 0.00 | 0 | 0.00 | 0.00 | 0.00 | 0.00 | NaN | true | |
| 112 | resnext0_stage3_conv14_fwd | Convolution | [1,512,14,14] | 3473 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | volta_scudnn_winograd_128x128_ldg1_ldg4_relu_tile148t_nt_v1 | 0.00 | 0 | 0.00 | 0.00 | 0.00 | 0.00 | NaN | true | |
| 112 | resnext0_stage3_conv14_fwd | Convolution | [1,512,14,14] | 3473 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | void cudnn::winograd::generateWinogradTilesKernel<0, float, float>(cudnn::winograd::GenerateWinogradTilesParams<float, float>) | 0.00 | 0 | 0.00 | 0.00 | 0.00 | 0.00 | NaN | true | |
| 112 | resnext0_stage3_conv14_fwd | Convolution | [1,512,14,14] | 3473 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | volta_scudnn_winograd_128x128_ldg1_ldg4_relu_tile148t_nt_v1 | 0.00 | 0 | 0.00 | 0.00 | 0.00 | 0.00 | NaN | true | |
| 112 | resnext0_stage3_conv14_fwd | Convolution | [1,512,14,14] | 3473 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | void cudnn::winograd::generateWinogradTilesKernel<0, float, float>(cudnn::winograd::GenerateWinogradTilesParams<float, float>) | 0.00 | 0 | 0.00 | 0.00 | 0.00 | 0.00 | NaN | true | |
| 112 | resnext0_stage3_conv14_fwd | Convolution | [1,512,14,14] | 3473 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | volta_scudnn_winograd_128x128_ldg1_ldg4_relu_tile148t_nt_v1 | 0.00 | 0 | 0.00 | 0.00 | 0.00 | 0.00 | NaN | true | |
| 112 | resnext0_stage3_conv14_fwd | Convolution | [1,512,14,14] | 3473 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | void cudnn::winograd::generateWinogradTilesKernel<0, float, float>(cudnn::winograd::GenerateWinogradTilesParams<float, float>) | 0.00 | 0 | 0.00 | 0.00 | 0.00 | 0.00 | NaN | true | |
| 112 | resnext0_stage3_conv14_fwd | Convolution | [1,512,14,14] | 3473 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | volta_scudnn_winograd_128x128_ldg1_ldg4_relu_tile148t_nt_v1 | 0.00 | 0 | 0.00 | 0.00 | 0.00 | 0.00 | NaN | true | |
| 112 | resnext0_stage3_conv14_fwd | Convolution | [1,512,14,14] | 3473 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | void cudnn::winograd::generateWinogradTilesKernel<0, float, float>(cudnn::winograd::GenerateWinogradTilesParams<float, float>) | 0.00 | 0 | 0.00 | 0.00 | 0.00 | 0.00 | NaN | true | |
| 112 | resnext0_stage3_conv14_fwd | Convolution | [1,512,14,14] | 3473 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | volta_scudnn_winograd_128x128_ldg1_ldg4_relu_tile148t_nt_v1 | 0.00 | 0 | 0.00 | 0.00 | 0.00 | 0.00 | NaN | true | |
| 112 | resnext0_stage3_conv14_fwd | Convolution | [1,512,14,14] | 3473 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | void cudnn::winograd::generateWinogradTilesKernel<0, float, float>(cudnn::winograd::GenerateWinogradTilesParams<float, float>) | 0.00 | 0 | 0.00 | 0.00 | 0.00 | 0.00 | NaN | true | |
| 112 | resnext0_stage3_conv14_fwd | Convolution | [1,512,14,14] | 3473 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | volta_scudnn_winograd_128x128_ldg1_ldg4_relu_tile148t_nt_v1 | 0.00 | 0 | 0.00 | 0.00 | 0.00 | 0.00 | NaN | true | |
| 112 | resnext0_stage3_conv14_fwd | Convolution | [1,512,14,14] | 3473 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | void cudnn::winograd::generateWinogradTilesKernel<0, float, float>(cudnn::winograd::GenerateWinogradTilesParams<float, float>) | 0.00 | 0 | 0.00 | 0.00 | 0.00 | 0.00 | NaN | true | |
| 112 | resnext0_stage3_conv14_fwd | Convolution | [1,512,14,14] | 3473 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | volta_scudnn_winograd_128x128_ldg1_ldg4_relu_tile148t_nt_v1 | 0.00 | 0 | 0.00 | 0.00 | 0.00 | 0.00 | NaN | true | |
| 112 | resnext0_stage3_conv14_fwd | Convolution | [1,512,14,14] | 3473 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | void cudnn::winograd::generateWinogradTilesKernel<0, float, float>(cudnn::winograd::GenerateWinogradTilesParams<float, float>) | 0.00 | 0 | 0.00 | 0.00 | 0.00 | 0.00 | NaN | true | |
| 112 | resnext0_stage3_conv14_fwd | Convolution | [1,512,14,14] | 3473 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | volta_scudnn_winograd_128x128_ldg1_ldg4_relu_tile148t_nt_v1 | 0.00 | 0 | 0.00 | 0.00 | 0.00 | 0.00 | NaN | true | |
| 112 | resnext0_stage3_conv14_fwd | Convolution | [1,512,14,14] | 3473 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | void cudnn::winograd::generateWinogradTilesKernel<0, float, float>(cudnn::winograd::GenerateWinogradTilesParams<float, float>) | 0.00 | 0 | 0.00 | 0.00 | 0.00 | 0.00 | NaN | true | |
| 112 | resnext0_stage3_conv14_fwd | Convolution | [1,512,14,14] | 3473 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | volta_scudnn_winograd_128x128_ldg1_ldg4_relu_tile148t_nt_v1 | 0.00 | 0 | 0.00 | 0.00 | 0.00 | 0.00 | NaN | true | |
| 112 | resnext0_stage3_conv14_fwd | Convolution | [1,512,14,14] | 3473 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | void cudnn::winograd::generateWinogradTilesKernel<0, float, float>(cudnn::winograd::GenerateWinogradTilesParams<float, float>) | 0.00 | 0 | 0.00 | 0.00 | 0.00 | 0.00 | NaN | true | |
| 112 | resnext0_stage3_conv14_fwd | Convolution | [1,512,14,14] | 3473 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | volta_scudnn_winograd_128x128_ldg1_ldg4_relu_tile148t_nt_v1 | 0.00 | 0 | 0.00 | 0.00 | 0.00 | 0.00 | NaN | true | |
| 112 | resnext0_stage3_conv14_fwd | Convolution | [1,512,14,14] | 3473 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | void cudnn::winograd::generateWinogradTilesKernel<0, float, float>(cudnn::winograd::GenerateWinogradTilesParams<float, float>) | 0.00 | 0 | 0.00 | 0.00 | 0.00 | 0.00 | NaN | true | |
| 112 | resnext0_stage3_conv14_fwd | Convolution | [1,512,14,14] | 3473 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | volta_scudnn_winograd_128x128_ldg1_ldg4_relu_tile148t_nt_v1 | 0.00 | 0 | 0.00 | 0.00 | 0.00 | 0.00 | NaN | true | |
| 112 | resnext0_stage3_conv14_fwd | Convolution | [1,512,14,14] | 3473 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | void cudnn::winograd::generateWinogradTilesKernel<0, float, float>(cudnn::winograd::GenerateWinogradTilesParams<float, float>) | 0.00 | 0 | 0.00 | 0.00 | 0.00 | 0.00 | NaN | true | |
| 112 | resnext0_stage3_conv14_fwd | Convolution | [1,512,14,14] | 3473 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | volta_scudnn_winograd_128x128_ldg1_ldg4_relu_tile148t_nt_v1 | 0.00 | 0 | 0.00 | 0.00 | 0.00 | 0.00 | NaN | true | |
| 112 | resnext0_stage3_conv14_fwd | Convolution | [1,512,14,14] | 3473 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | void cudnn::winograd::generateWinogradTilesKernel<0, float, float>(cudnn::winograd::GenerateWinogradTilesParams<float, float>) | 0.00 | 0 | 0.00 | 0.00 | 0.00 | 0.00 | NaN | true | |
| 112 | resnext0_stage3_conv14_fwd | Convolution | [1,512,14,14] | 3473 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | volta_scudnn_winograd_128x128_ldg1_ldg4_relu_tile148t_nt_v1 | 0.00 | 0 | 0.00 | 0.00 | 0.00 | 0.00 | NaN | true | |
| 112 | resnext0_stage3_conv14_fwd | Convolution | [1,512,14,14] | 3473 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | void cudnn::winograd::generateWinogradTilesKernel<0, float, float>(cudnn::winograd::GenerateWinogradTilesParams<float, float>) | 0.00 | 0 | 0.00 | 0.00 | 0.00 | 0.00 | NaN | true | |
| 112 | resnext0_stage3_conv14_fwd | Convolution | [1,512,14,14] | 3473 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | volta_scudnn_winograd_128x128_ldg1_ldg4_relu_tile148t_nt_v1 | 0.00 | 0 | 0.00 | 0.00 | 0.00 | 0.00 | NaN | true | |
| 112 | resnext0_stage3_conv14_fwd | Convolution | [1,512,14,14] | 3473 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | void cudnn::winograd::generateWinogradTilesKernel<0, float, float>(cudnn::winograd::GenerateWinogradTilesParams<float, float>) | 0.00 | 0 | 0.00 | 0.00 | 0.00 | 0.00 | NaN | true | |
| 112 | resnext0_stage3_conv14_fwd | Convolution | [1,512,14,14] | 3473 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | volta_scudnn_winograd_128x128_ldg1_ldg4_relu_tile148t_nt_v1 | 0.00 | 0 | 0.00 | 0.00 | 0.00 | 0.00 | NaN | true | |
| 112 | resnext0_stage3_conv14_fwd | Convolution | [1,512,14,14] | 3473 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | void cudnn::winograd::generateWinogradTilesKernel<0, float, float>(cudnn::winograd::GenerateWinogradTilesParams<float, float>) | 0.00 | 0 | 0.00 | 0.00 | 0.00 | 0.00 | NaN | true | |
| 112 | resnext0_stage3_conv14_fwd | Convolution | [1,512,14,14] | 3473 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | volta_scudnn_winograd_128x128_ldg1_ldg4_relu_tile148t_nt_v1 | 0.00 | 0 | 0.00 | 0.00 | 0.00 | 0.00 | NaN | true | |
| 112 | resnext0_stage3_conv14_fwd | Convolution | [1,512,14,14] | 3473 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | void cudnn::winograd::generateWinogradTilesKernel<0, float, float>(cudnn::winograd::GenerateWinogradTilesParams<float, float>) | 0.00 | 0 | 0.00 | 0.00 | 0.00 | 0.00 | NaN | true | |
| 112 | resnext0_stage3_conv14_fwd | Convolution | [1,512,14,14] | 3473 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | volta_scudnn_winograd_128x128_ldg1_ldg4_relu_tile148t_nt_v1 | 0.00 | 0 | 0.00 | 0.00 | 0.00 | 0.00 | NaN | true | |
| 112 | resnext0_stage3_conv14_fwd | Convolution | [1,512,14,14] | 3473 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | void cudnn::winograd::generateWinogradTilesKernel<0, float, float>(cudnn::winograd::GenerateWinogradTilesParams<float, float>) | 0.00 | 0 | 0.00 | 0.00 | 0.00 | 0.00 | NaN | true | |
| 112 | resnext0_stage3_conv14_fwd | Convolution | [1,512,14,14] | 3473 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | volta_scudnn_winograd_128x128_ldg1_ldg4_relu_tile148t_nt_v1 | 0.00 | 0 | 0.00 | 0.00 | 0.00 | 0.00 | NaN | true | |
| 112 | resnext0_stage3_conv14_fwd | Convolution | [1,512,14,14] | 3473 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | void cudnn::winograd::generateWinogradTilesKernel<0, float, float>(cudnn::winograd::GenerateWinogradTilesParams<float, float>) | 0.00 | 0 | 0.00 | 0.00 | 0.00 | 0.00 | NaN | true | |
| 112 | resnext0_stage3_conv14_fwd | Convolution | [1,512,14,14] | 3473 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | volta_scudnn_winograd_128x128_ldg1_ldg4_relu_tile148t_nt_v1 | 0.00 | 0 | 0.00 | 0.00 | 0.00 | 0.00 | NaN | true | |
| 112 | resnext0_stage3_conv14_fwd | Convolution | [1,512,14,14] | 3473 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | void cudnn::winograd::generateWinogradTilesKernel<0, float, float>(cudnn::winograd::GenerateWinogradTilesParams<float, float>) | 0.00 | 0 | 0.00 | 0.00 | 0.00 | 0.00 | NaN | true | |
| 112 | resnext0_stage3_conv14_fwd | Convolution | [1,512,14,14] | 3473 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | volta_scudnn_winograd_128x128_ldg1_ldg4_relu_tile148t_nt_v1 | 0.00 | 0 | 0.00 | 0.00 | 0.00 | 0.00 | NaN | true | |
| 112 | resnext0_stage3_conv14_fwd | Convolution | [1,512,14,14] | 3473 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | void cudnn::winograd::generateWinogradTilesKernel<0, float, float>(cudnn::winograd::GenerateWinogradTilesParams<float, float>) | 0.00 | 0 | 0.00 | 0.00 | 0.00 | 0.00 | NaN | true | |
| 112 | resnext0_stage3_conv14_fwd | Convolution | [1,512,14,14] | 3473 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | volta_scudnn_winograd_128x128_ldg1_ldg4_relu_tile148t_nt_v1 | 0.00 | 0 | 0.00 | 0.00 | 0.00 | 0.00 | NaN | true | |
| 112 | resnext0_stage3_conv14_fwd | Convolution | [1,512,14,14] | 3473 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | void cudnn::winograd::generateWinogradTilesKernel<0, float, float>(cudnn::winograd::GenerateWinogradTilesParams<float, float>) | 0.00 | 0 | 0.00 | 0.00 | 0.00 | 0.00 | NaN | true | |
| 112 | resnext0_stage3_conv14_fwd | Convolution | [1,512,14,14] | 3473 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | volta_scudnn_winograd_128x128_ldg1_ldg4_relu_tile148t_nt_v1 | 0.00 | 0 | 0.00 | 0.00 | 0.00 | 0.00 | NaN | true | |
| 112 | resnext0_stage3_conv14_fwd | Convolution | [1,512,14,14] | 3473 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | void cudnn::winograd::generateWinogradTilesKernel<0, float, float>(cudnn::winograd::GenerateWinogradTilesParams<float, float>) | 0.00 | 0 | 0.00 | 0.00 | 0.00 | 0.00 | NaN | true | |
| 112 | resnext0_stage3_conv14_fwd | Convolution | [1,512,14,14] | 3473 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | volta_scudnn_winograd_128x128_ldg1_ldg4_relu_tile148t_nt_v1 | 0.00 | 0 | 0.00 | 0.00 | 0.00 | 0.00 | NaN | true | |
| 113 | resnext0_stage3_batchnorm14_fwd | BatchNorm | [1,512,14,14] | 91.667 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | void cudnn::detail::bn_fw_inf_1C11_kernel_NCHW<float, float, true, 1>(float, float, cudnnTensorStruct, float const*, cudnnTensorStruct, float*, cudnnTensorStruct, float const*, float const*, float const*, float const*, float) | 0.00 | 0 | 0.00 | 0.00 | 0.00 | 0.00 | NaN | true | |
| 114 | resnext0_stage3_relu9_fwd | Activation | [1,512,14,14] | 21 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | void op_generic_tensor_kernel<2, float, float, float, 256, (cudnnGenericOp_t)8, (cudnnNanPropagation_t)0, (cudnnDimOrder_t)0, 1>(cudnnTensorStruct, float*, cudnnTensorStruct, float const*, cudnnTensorStruct, float const*, float, float, float, float, dimArray, reducedDivisorArray, bool) | 0.00 | 0 | 0.00 | 0.00 | 0.00 | 0.00 | NaN | true | |
| 115 | resnext0_stage3_conv15_fwd | Convolution | [1,512,14,14] | 10702 | 0 | 0 | 0 | 0 | 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) | 0.00 | 0 | 0.00 | 0.00 | 0.00 | 0.00 | NaN | true | |
| 116 | resnext0_stage3_batchnorm15_fwd | BatchNorm | [1,1024,14,14] | 112.333 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | void cudnn::detail::bn_fw_inf_1C11_kernel_NCHW<float, float, true, 1>(float, float, cudnnTensorStruct, float const*, cudnnTensorStruct, float*, cudnnTensorStruct, float const*, float const*, float const*, float const*, float) | 0.00 | 0 | 0.00 | 0.00 | 0.00 | 0.00 | NaN | true | |
| 117 | add_resnext0_stage3_activation4 | add_relu | [1,1024,14,14] | 52 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | void mxnet::op::mxnet_op::mxnet_generic_kernel<mxnet::op::AddReluKernel, float*, float*, float*, mxnet::OpReqType>(int, float*, float*, float*, mxnet::OpReqType) | 0.00 | 0 | 0.00 | 0.00 | 0.00 | 0.00 | NaN | true | |
| 118 | resnext0_stage3_conv16_fwd | Convolution | [1,1024,14,14] | 11281.333 | 0 | 0 | 0 | 0 | 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) | 0.00 | 0 | 0.00 | 0.00 | 0.00 | 0.00 | NaN | true | |
| 119 | resnext0_stage3_batchnorm16_fwd | BatchNorm | [1,512,14,14] | 98.333 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | void cudnn::detail::bn_fw_inf_1C11_kernel_NCHW<float, float, true, 1>(float, float, cudnnTensorStruct, float const*, cudnnTensorStruct, float*, cudnnTensorStruct, float const*, float const*, float const*, float const*, float) | 0.00 | 0 | 0.00 | 0.00 | 0.00 | 0.00 | NaN | true | |
| 120 | resnext0_stage3_relu10_fwd | Activation | [1,512,14,14] | 21 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | void op_generic_tensor_kernel<2, float, float, float, 256, (cudnnGenericOp_t)8, (cudnnNanPropagation_t)0, (cudnnDimOrder_t)0, 1>(cudnnTensorStruct, float*, cudnnTensorStruct, float const*, cudnnTensorStruct, float const*, float, float, float, float, dimArray, reducedDivisorArray, bool) | 0.00 | 0 | 0.00 | 0.00 | 0.00 | 0.00 | NaN | true | |
| 121 | resnext0_stage3_conv17_fwd | Convolution | [1,512,14,14] | 3483 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | void cudnn::winograd::generateWinogradTilesKernel<0, float, float>(cudnn::winograd::GenerateWinogradTilesParams<float, float>) | 0.00 | 0 | 0.00 | 0.00 | 0.00 | 0.00 | NaN | true | |
| 121 | resnext0_stage3_conv17_fwd | Convolution | [1,512,14,14] | 3483 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | volta_scudnn_winograd_128x128_ldg1_ldg4_relu_tile148t_nt_v1 | 0.00 | 0 | 0.00 | 0.00 | 0.00 | 0.00 | NaN | true | |
| 121 | resnext0_stage3_conv17_fwd | Convolution | [1,512,14,14] | 3483 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | void cudnn::winograd::generateWinogradTilesKernel<0, float, float>(cudnn::winograd::GenerateWinogradTilesParams<float, float>) | 0.00 | 0 | 0.00 | 0.00 | 0.00 | 0.00 | NaN | true | |
| 121 | resnext0_stage3_conv17_fwd | Convolution | [1,512,14,14] | 3483 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | volta_scudnn_winograd_128x128_ldg1_ldg4_relu_tile148t_nt_v1 | 0.00 | 0 | 0.00 | 0.00 | 0.00 | 0.00 | NaN | true | |
| 121 | resnext0_stage3_conv17_fwd | Convolution | [1,512,14,14] | 3483 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | void cudnn::winograd::generateWinogradTilesKernel<0, float, float>(cudnn::winograd::GenerateWinogradTilesParams<float, float>) | 0.00 | 0 | 0.00 | 0.00 | 0.00 | 0.00 | NaN | true | |
| 121 | resnext0_stage3_conv17_fwd | Convolution | [1,512,14,14] | 3483 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | volta_scudnn_winograd_128x128_ldg1_ldg4_relu_tile148t_nt_v1 | 0.00 | 0 | 0.00 | 0.00 | 0.00 | 0.00 | NaN | true | |
| 121 | resnext0_stage3_conv17_fwd | Convolution | [1,512,14,14] | 3483 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | void cudnn::winograd::generateWinogradTilesKernel<0, float, float>(cudnn::winograd::GenerateWinogradTilesParams<float, float>) | 0.00 | 0 | 0.00 | 0.00 | 0.00 | 0.00 | NaN | true | |
| 121 | resnext0_stage3_conv17_fwd | Convolution | [1,512,14,14] | 3483 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | volta_scudnn_winograd_128x128_ldg1_ldg4_relu_tile148t_nt_v1 | 0.00 | 0 | 0.00 | 0.00 | 0.00 | 0.00 | NaN | true | |
| 121 | resnext0_stage3_conv17_fwd | Convolution | [1,512,14,14] | 3483 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | void cudnn::winograd::generateWinogradTilesKernel<0, float, float>(cudnn::winograd::GenerateWinogradTilesParams<float, float>) | 0.00 | 0 | 0.00 | 0.00 | 0.00 | 0.00 | NaN | true | |
| 121 | resnext0_stage3_conv17_fwd | Convolution | [1,512,14,14] | 3483 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | volta_scudnn_winograd_128x128_ldg1_ldg4_relu_tile148t_nt_v1 | 0.00 | 0 | 0.00 | 0.00 | 0.00 | 0.00 | NaN | true | |
| 121 | resnext0_stage3_conv17_fwd | Convolution | [1,512,14,14] | 3483 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | void cudnn::winograd::generateWinogradTilesKernel<0, float, float>(cudnn::winograd::GenerateWinogradTilesParams<float, float>) | 0.00 | 0 | 0.00 | 0.00 | 0.00 | 0.00 | NaN | true | |
| 121 | resnext0_stage3_conv17_fwd | Convolution | [1,512,14,14] | 3483 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | volta_scudnn_winograd_128x128_ldg1_ldg4_relu_tile148t_nt_v1 | 0.00 | 0 | 0.00 | 0.00 | 0.00 | 0.00 | NaN | true | |
| 121 | resnext0_stage3_conv17_fwd | Convolution | [1,512,14,14] | 3483 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | void cudnn::winograd::generateWinogradTilesKernel<0, float, float>(cudnn::winograd::GenerateWinogradTilesParams<float, float>) | 0.00 | 0 | 0.00 | 0.00 | 0.00 | 0.00 | NaN | true | |
| 121 | resnext0_stage3_conv17_fwd | Convolution | [1,512,14,14] | 3483 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | volta_scudnn_winograd_128x128_ldg1_ldg4_relu_tile148t_nt_v1 | 0.00 | 0 | 0.00 | 0.00 | 0.00 | 0.00 | NaN | true | |
| 121 | resnext0_stage3_conv17_fwd | Convolution | [1,512,14,14] | 3483 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | void cudnn::winograd::generateWinogradTilesKernel<0, float, float>(cudnn::winograd::GenerateWinogradTilesParams<float, float>) | 0.00 | 0 | 0.00 | 0.00 | 0.00 | 0.00 | NaN | true | |
| 121 | resnext0_stage3_conv17_fwd | Convolution | [1,512,14,14] | 3483 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | volta_scudnn_winograd_128x128_ldg1_ldg4_relu_tile148t_nt_v1 | 0.00 | 0 | 0.00 | 0.00 | 0.00 | 0.00 | NaN | true | |
| 121 | resnext0_stage3_conv17_fwd | Convolution | [1,512,14,14] | 3483 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | void cudnn::winograd::generateWinogradTilesKernel<0, float, float>(cudnn::winograd::GenerateWinogradTilesParams<float, float>) | 0.00 | 0 | 0.00 | 0.00 | 0.00 | 0.00 | NaN | true | |
| 121 | resnext0_stage3_conv17_fwd | Convolution | [1,512,14,14] | 3483 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | volta_scudnn_winograd_128x128_ldg1_ldg4_relu_tile148t_nt_v1 | 0.00 | 0 | 0.00 | 0.00 | 0.00 | 0.00 | NaN | true | |
| 121 | resnext0_stage3_conv17_fwd | Convolution | [1,512,14,14] | 3483 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | void cudnn::winograd::generateWinogradTilesKernel<0, float, float>(cudnn::winograd::GenerateWinogradTilesParams<float, float>) | 0.00 | 0 | 0.00 | 0.00 | 0.00 | 0.00 | NaN | true | |
| 121 | resnext0_stage3_conv17_fwd | Convolution | [1,512,14,14] | 3483 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | volta_scudnn_winograd_128x128_ldg1_ldg4_relu_tile148t_nt_v1 | 0.00 | 0 | 0.00 | 0.00 | 0.00 | 0.00 | NaN | true | |
| 121 | resnext0_stage3_conv17_fwd | Convolution | [1,512,14,14] | 3483 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | void cudnn::winograd::generateWinogradTilesKernel<0, float, float>(cudnn::winograd::GenerateWinogradTilesParams<float, float>) | 0.00 | 0 | 0.00 | 0.00 | 0.00 | 0.00 | NaN | true | |
| 121 | resnext0_stage3_conv17_fwd | Convolution | [1,512,14,14] | 3483 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | volta_scudnn_winograd_128x128_ldg1_ldg4_relu_tile148t_nt_v1 | 0.00 | 0 | 0.00 | 0.00 | 0.00 | 0.00 | NaN | true | |
| 121 | resnext0_stage3_conv17_fwd | Convolution | [1,512,14,14] | 3483 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | void cudnn::winograd::generateWinogradTilesKernel<0, float, float>(cudnn::winograd::GenerateWinogradTilesParams<float, float>) | 0.00 | 0 | 0.00 | 0.00 | 0.00 | 0.00 | NaN | true | |
| 121 | resnext0_stage3_conv17_fwd | Convolution | [1,512,14,14] | 3483 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | volta_scudnn_winograd_128x128_ldg1_ldg4_relu_tile148t_nt_v1 | 0.00 | 0 | 0.00 | 0.00 | 0.00 | 0.00 | NaN | true | |
| 121 | resnext0_stage3_conv17_fwd | Convolution | [1,512,14,14] | 3483 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | void cudnn::winograd::generateWinogradTilesKernel<0, float, float>(cudnn::winograd::GenerateWinogradTilesParams<float, float>) | 0.00 | 0 | 0.00 | 0.00 | 0.00 | 0.00 | NaN | true | |
| 121 | resnext0_stage3_conv17_fwd | Convolution | [1,512,14,14] | 3483 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | volta_scudnn_winograd_128x128_ldg1_ldg4_relu_tile148t_nt_v1 | 0.00 | 0 | 0.00 | 0.00 | 0.00 | 0.00 | NaN | true | |
| 121 | resnext0_stage3_conv17_fwd | Convolution | [1,512,14,14] | 3483 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | void cudnn::winograd::generateWinogradTilesKernel<0, float, float>(cudnn::winograd::GenerateWinogradTilesParams<float, float>) | 0.00 | 0 | 0.00 | 0.00 | 0.00 | 0.00 | NaN | true | |
| 121 | resnext0_stage3_conv17_fwd | Convolution | [1,512,14,14] | 3483 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | volta_scudnn_winograd_128x128_ldg1_ldg4_relu_tile148t_nt_v1 | 0.00 | 0 | 0.00 | 0.00 | 0.00 | 0.00 | NaN | true | |
| 121 | resnext0_stage3_conv17_fwd | Convolution | [1,512,14,14] | 3483 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | void cudnn::winograd::generateWinogradTilesKernel<0, float, float>(cudnn::winograd::GenerateWinogradTilesParams<float, float>) | 0.00 | 0 | 0.00 | 0.00 | 0.00 | 0.00 | NaN | true | |
| 121 | resnext0_stage3_conv17_fwd | Convolution | [1,512,14,14] | 3483 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | volta_scudnn_winograd_128x128_ldg1_ldg4_relu_tile148t_nt_v1 | 0.00 | 0 | 0.00 | 0.00 | 0.00 | 0.00 | NaN | true | |
| 121 | resnext0_stage3_conv17_fwd | Convolution | [1,512,14,14] | 3483 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | void cudnn::winograd::generateWinogradTilesKernel<0, float, float>(cudnn::winograd::GenerateWinogradTilesParams<float, float>) | 0.00 | 0 | 0.00 | 0.00 | 0.00 | 0.00 | NaN | true | |
| 121 | resnext0_stage3_conv17_fwd | Convolution | [1,512,14,14] | 3483 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | volta_scudnn_winograd_128x128_ldg1_ldg4_relu_tile148t_nt_v1 | 0.00 | 0 | 0.00 | 0.00 | 0.00 | 0.00 | NaN | true | |
| 121 | resnext0_stage3_conv17_fwd | Convolution | [1,512,14,14] | 3483 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | void cudnn::winograd::generateWinogradTilesKernel<0, float, float>(cudnn::winograd::GenerateWinogradTilesParams<float, float>) | 0.00 | 0 | 0.00 | 0.00 | 0.00 | 0.00 | NaN | true | |
| 121 | resnext0_stage3_conv17_fwd | Convolution | [1,512,14,14] | 3483 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | volta_scudnn_winograd_128x128_ldg1_ldg4_relu_tile148t_nt_v1 | 0.00 | 0 | 0.00 | 0.00 | 0.00 | 0.00 | NaN | true | |
| 121 | resnext0_stage3_conv17_fwd | Convolution | [1,512,14,14] | 3483 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | void cudnn::winograd::generateWinogradTilesKernel<0, float, float>(cudnn::winograd::GenerateWinogradTilesParams<float, float>) | 0.00 | 0 | 0.00 | 0.00 | 0.00 | 0.00 | NaN | true | |
| 121 | resnext0_stage3_conv17_fwd | Convolution | [1,512,14,14] | 3483 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | volta_scudnn_winograd_128x128_ldg1_ldg4_relu_tile148t_nt_v1 | 0.00 | 0 | 0.00 | 0.00 | 0.00 | 0.00 | NaN | true | |
| 121 | resnext0_stage3_conv17_fwd | Convolution | [1,512,14,14] | 3483 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | void cudnn::winograd::generateWinogradTilesKernel<0, float, float>(cudnn::winograd::GenerateWinogradTilesParams<float, float>) | 0.00 | 0 | 0.00 | 0.00 | 0.00 | 0.00 | NaN | true | |
| 121 | resnext0_stage3_conv17_fwd | Convolution | [1,512,14,14] | 3483 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | volta_scudnn_winograd_128x128_ldg1_ldg4_relu_tile148t_nt_v1 | 0.00 | 0 | 0.00 | 0.00 | 0.00 | 0.00 | NaN | true | |
| 121 | resnext0_stage3_conv17_fwd | Convolution | [1,512,14,14] | 3483 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | void cudnn::winograd::generateWinogradTilesKernel<0, float, float>(cudnn::winograd::GenerateWinogradTilesParams<float, float>) | 0.00 | 0 | 0.00 | 0.00 | 0.00 | 0.00 | NaN | true | |
| 121 | resnext0_stage3_conv17_fwd | Convolution | [1,512,14,14] | 3483 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | volta_scudnn_winograd_128x128_ldg1_ldg4_relu_tile148t_nt_v1 | 0.00 | 0 | 0.00 | 0.00 | 0.00 | 0.00 | NaN | true | |
| 121 | resnext0_stage3_conv17_fwd | Convolution | [1,512,14,14] | 3483 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | void cudnn::winograd::generateWinogradTilesKernel<0, float, float>(cudnn::winograd::GenerateWinogradTilesParams<float, float>) | 0.00 | 0 | 0.00 | 0.00 | 0.00 | 0.00 | NaN | true | |
| 121 | resnext0_stage3_conv17_fwd | Convolution | [1,512,14,14] | 3483 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | volta_scudnn_winograd_128x128_ldg1_ldg4_relu_tile148t_nt_v1 | 0.00 | 0 | 0.00 | 0.00 | 0.00 | 0.00 | NaN | true | |
| 121 | resnext0_stage3_conv17_fwd | Convolution | [1,512,14,14] | 3483 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | void cudnn::winograd::generateWinogradTilesKernel<0, float, float>(cudnn::winograd::GenerateWinogradTilesParams<float, float>) | 0.00 | 0 | 0.00 | 0.00 | 0.00 | 0.00 | NaN | true | |
| 121 | resnext0_stage3_conv17_fwd | Convolution | [1,512,14,14] | 3483 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | volta_scudnn_winograd_128x128_ldg1_ldg4_relu_tile148t_nt_v1 | 0.00 | 0 | 0.00 | 0.00 | 0.00 | 0.00 | NaN | true | |
| 121 | resnext0_stage3_conv17_fwd | Convolution | [1,512,14,14] | 3483 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | void cudnn::winograd::generateWinogradTilesKernel<0, float, float>(cudnn::winograd::GenerateWinogradTilesParams<float, float>) | 0.00 | 0 | 0.00 | 0.00 | 0.00 | 0.00 | NaN | true | |
| 121 | resnext0_stage3_conv17_fwd | Convolution | [1,512,14,14] | 3483 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | volta_scudnn_winograd_128x128_ldg1_ldg4_relu_tile148t_nt_v1 | 0.00 | 0 | 0.00 | 0.00 | 0.00 | 0.00 | NaN | true | |
| 121 | resnext0_stage3_conv17_fwd | Convolution | [1,512,14,14] | 3483 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | void cudnn::winograd::generateWinogradTilesKernel<0, float, float>(cudnn::winograd::GenerateWinogradTilesParams<float, float>) | 0.00 | 0 | 0.00 | 0.00 | 0.00 | 0.00 | NaN | true | |
| 121 | resnext0_stage3_conv17_fwd | Convolution | [1,512,14,14] | 3483 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | volta_scudnn_winograd_128x128_ldg1_ldg4_relu_tile148t_nt_v1 | 0.00 | 0 | 0.00 | 0.00 | 0.00 | 0.00 | NaN | true | |
| 121 | resnext0_stage3_conv17_fwd | Convolution | [1,512,14,14] | 3483 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | void cudnn::winograd::generateWinogradTilesKernel<0, float, float>(cudnn::winograd::GenerateWinogradTilesParams<float, float>) | 0.00 | 0 | 0.00 | 0.00 | 0.00 | 0.00 | NaN | true | |
| 121 | resnext0_stage3_conv17_fwd | Convolution | [1,512,14,14] | 3483 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | volta_scudnn_winograd_128x128_ldg1_ldg4_relu_tile148t_nt_v1 | 0.00 | 0 | 0.00 | 0.00 | 0.00 | 0.00 | NaN | true | |
| 121 | resnext0_stage3_conv17_fwd | Convolution | [1,512,14,14] | 3483 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | void cudnn::winograd::generateWinogradTilesKernel<0, float, float>(cudnn::winograd::GenerateWinogradTilesParams<float, float>) | 0.00 | 0 | 0.00 | 0.00 | 0.00 | 0.00 | NaN | true | |
| 121 | resnext0_stage3_conv17_fwd | Convolution | [1,512,14,14] | 3483 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | volta_scudnn_winograd_128x128_ldg1_ldg4_relu_tile148t_nt_v1 | 0.00 | 0 | 0.00 | 0.00 | 0.00 | 0.00 | NaN | true | |
| 121 | resnext0_stage3_conv17_fwd | Convolution | [1,512,14,14] | 3483 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | void cudnn::winograd::generateWinogradTilesKernel<0, float, float>(cudnn::winograd::GenerateWinogradTilesParams<float, float>) | 0.00 | 0 | 0.00 | 0.00 | 0.00 | 0.00 | NaN | true | |
| 121 | resnext0_stage3_conv17_fwd | Convolution | [1,512,14,14] | 3483 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | volta_scudnn_winograd_128x128_ldg1_ldg4_relu_tile148t_nt_v1 | 0.00 | 0 | 0.00 | 0.00 | 0.00 | 0.00 | NaN | true | |
| 121 | resnext0_stage3_conv17_fwd | Convolution | [1,512,14,14] | 3483 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | void cudnn::winograd::generateWinogradTilesKernel<0, float, float>(cudnn::winograd::GenerateWinogradTilesParams<float, float>) | 0.00 | 0 | 0.00 | 0.00 | 0.00 | 0.00 | NaN | true | |
| 121 | resnext0_stage3_conv17_fwd | Convolution | [1,512,14,14] | 3483 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | volta_scudnn_winograd_128x128_ldg1_ldg4_relu_tile148t_nt_v1 | 0.00 | 0 | 0.00 | 0.00 | 0.00 | 0.00 | NaN | true | |
| 121 | resnext0_stage3_conv17_fwd | Convolution | [1,512,14,14] | 3483 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | void cudnn::winograd::generateWinogradTilesKernel<0, float, float>(cudnn::winograd::GenerateWinogradTilesParams<float, float>) | 0.00 | 0 | 0.00 | 0.00 | 0.00 | 0.00 | NaN | true | |
| 121 | resnext0_stage3_conv17_fwd | Convolution | [1,512,14,14] | 3483 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | volta_scudnn_winograd_128x128_ldg1_ldg4_relu_tile148t_nt_v1 | 0.00 | 0 | 0.00 | 0.00 | 0.00 | 0.00 | NaN | true | |
| 121 | resnext0_stage3_conv17_fwd | Convolution | [1,512,14,14] | 3483 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | void cudnn::winograd::generateWinogradTilesKernel<0, float, float>(cudnn::winograd::GenerateWinogradTilesParams<float, float>) | 0.00 | 0 | 0.00 | 0.00 | 0.00 | 0.00 | NaN | true | |
| 121 | resnext0_stage3_conv17_fwd | Convolution | [1,512,14,14] | 3483 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | volta_scudnn_winograd_128x128_ldg1_ldg4_relu_tile148t_nt_v1 | 0.00 | 0 | 0.00 | 0.00 | 0.00 | 0.00 | NaN | true | |
| 121 | resnext0_stage3_conv17_fwd | Convolution | [1,512,14,14] | 3483 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | void cudnn::winograd::generateWinogradTilesKernel<0, float, float>(cudnn::winograd::GenerateWinogradTilesParams<float, float>) | 0.00 | 0 | 0.00 | 0.00 | 0.00 | 0.00 | NaN | true | |
| 121 | resnext0_stage3_conv17_fwd | Convolution | [1,512,14,14] | 3483 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | volta_scudnn_winograd_128x128_ldg1_ldg4_relu_tile148t_nt_v1 | 0.00 | 0 | 0.00 | 0.00 | 0.00 | 0.00 | NaN | true | |
| 121 | resnext0_stage3_conv17_fwd | Convolution | [1,512,14,14] | 3483 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | void cudnn::winograd::generateWinogradTilesKernel<0, float, float>(cudnn::winograd::GenerateWinogradTilesParams<float, float>) | 0.00 | 0 | 0.00 | 0.00 | 0.00 | 0.00 | NaN | true | |
| 121 | resnext0_stage3_conv17_fwd | Convolution | [1,512,14,14] | 3483 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | volta_scudnn_winograd_128x128_ldg1_ldg4_relu_tile148t_nt_v1 | 0.00 | 0 | 0.00 | 0.00 | 0.00 | 0.00 | NaN | true | |
| 122 | resnext0_stage3_batchnorm17_fwd | BatchNorm | [1,512,14,14] | 57.333 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | void cudnn::detail::bn_fw_inf_1C11_kernel_NCHW<float, float, true, 1>(float, float, cudnnTensorStruct, float const*, cudnnTensorStruct, float*, cudnnTensorStruct, float const*, float const*, float const*, float const*, float) | 0.00 | 0 | 0.00 | 0.00 | 0.00 | 0.00 | NaN | true | |
| 123 | resnext0_stage3_relu11_fwd | Activation | [1,512,14,14] | 21.333 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | void op_generic_tensor_kernel<2, float, float, float, 256, (cudnnGenericOp_t)8, (cudnnNanPropagation_t)0, (cudnnDimOrder_t)0, 1>(cudnnTensorStruct, float*, cudnnTensorStruct, float const*, cudnnTensorStruct, float const*, float, float, float, float, dimArray, reducedDivisorArray, bool) | 0.00 | 0 | 0.00 | 0.00 | 0.00 | 0.00 | NaN | true | |
| 124 | resnext0_stage3_conv18_fwd | Convolution | [1,512,14,14] | 10951 | 0 | 0 | 0 | 0 | 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) | 0.00 | 0 | 0.00 | 0.00 | 0.00 | 0.00 | NaN | true | |
| 125 | resnext0_stage3_batchnorm18_fwd | BatchNorm | [1,1024,14,14] | 124.667 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | void cudnn::detail::bn_fw_inf_1C11_kernel_NCHW<float, float, true, 1>(float, float, cudnnTensorStruct, float const*, cudnnTensorStruct, float*, cudnnTensorStruct, float const*, float const*, float const*, float const*, float) | 0.00 | 0 | 0.00 | 0.00 | 0.00 | 0.00 | NaN | true | |
| 126 | add_resnext0_stage3_activation5 | add_relu | [1,1024,14,14] | 55 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | void mxnet::op::mxnet_op::mxnet_generic_kernel<mxnet::op::AddReluKernel, float*, float*, float*, mxnet::OpReqType>(int, float*, float*, float*, mxnet::OpReqType) | 0.00 | 0 | 0.00 | 0.00 | 0.00 | 0.00 | NaN | true | |
| 127 | resnext0_stage4_conv0_fwd | Convolution | [1,1024,14,14] | 21011 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | cudnn::gemm::computeOffsetsKernel(cudnn::gemm::ComputeOffsetsParams) | 0.00 | 0 | 0.00 | 0.00 | 0.00 | 0.00 | NaN | true | |
| 127 | resnext0_stage4_conv0_fwd | Convolution | [1,1024,14,14] | 21011 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | volta_scudnn_128x64_relu_interior_nn_v1 | 0.00 | 0 | 0.00 | 0.00 | 0.00 | 0.00 | NaN | true | |
| 128 | resnext0_stage4_batchnorm0_fwd | BatchNorm | [1,1024,14,14] | 108.333 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | void cudnn::detail::bn_fw_inf_1C11_kernel_NCHW<float, float, true, 1>(float, float, cudnnTensorStruct, float const*, cudnnTensorStruct, float*, cudnnTensorStruct, float const*, float const*, float const*, float const*, float) | 0.00 | 0 | 0.00 | 0.00 | 0.00 | 0.00 | NaN | true | |
| 129 | resnext0_stage4_relu0_fwd | Activation | [1,1024,14,14] | 38.333 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | void op_generic_tensor_kernel<2, float, float, float, 256, (cudnnGenericOp_t)8, (cudnnNanPropagation_t)0, (cudnnDimOrder_t)0, 1>(cudnnTensorStruct, float*, cudnnTensorStruct, float const*, cudnnTensorStruct, float const*, float, float, float, float, dimArray, reducedDivisorArray, bool) | 0.00 | 0 | 0.00 | 0.00 | 0.00 | 0.00 | NaN | true | |
| 130 | resnext0_stage4_conv1_fwd | Convolution | [1,1024,14,14] | 3063.667 | 0 | 0 | 0 | 0 | 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) | 0.00 | 0 | 0.00 | 0.00 | 0.00 | 0.00 | NaN | true | |
| 130 | resnext0_stage4_conv1_fwd | Convolution | [1,1024,14,14] | 3063.667 | 0 | 0 | 0 | 0 | 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) | 0.00 | 0 | 0.00 | 0.00 | 0.00 | 0.00 | NaN | true | |
| 130 | resnext0_stage4_conv1_fwd | Convolution | [1,1024,14,14] | 3063.667 | 0 | 0 | 0 | 0 | 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) | 0.00 | 0 | 0.00 | 0.00 | 0.00 | 0.00 | NaN | true | |
| 130 | resnext0_stage4_conv1_fwd | Convolution | [1,1024,14,14] | 3063.667 | 0 | 0 | 0 | 0 | 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) | 0.00 | 0 | 0.00 | 0.00 | 0.00 | 0.00 | NaN | true | |
| 130 | resnext0_stage4_conv1_fwd | Convolution | [1,1024,14,14] | 3063.667 | 0 | 0 | 0 | 0 | 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) | 0.00 | 0 | 0.00 | 0.00 | 0.00 | 0.00 | NaN | true | |
| 130 | resnext0_stage4_conv1_fwd | Convolution | [1,1024,14,14] | 3063.667 | 0 | 0 | 0 | 0 | 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) | 0.00 | 0 | 0.00 | 0.00 | 0.00 | 0.00 | NaN | true | |
| 130 | resnext0_stage4_conv1_fwd | Convolution | [1,1024,14,14] | 3063.667 | 0 | 0 | 0 | 0 | 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) | 0.00 | 0 | 0.00 | 0.00 | 0.00 | 0.00 | NaN | true | |
| 130 | resnext0_stage4_conv1_fwd | Convolution | [1,1024,14,14] | 3063.667 | 0 | 0 | 0 | 0 | 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) | 0.00 | 0 | 0.00 | 0.00 | 0.00 | 0.00 | NaN | true | |
| 130 | resnext0_stage4_conv1_fwd | Convolution | [1,1024,14,14] | 3063.667 | 0 | 0 | 0 | 0 | 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) | 0.00 | 0 | 0.00 | 0.00 | 0.00 | 0.00 | NaN | true | |
| 130 | resnext0_stage4_conv1_fwd | Convolution | [1,1024,14,14] | 3063.667 | 0 | 0 | 0 | 0 | 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) | 0.00 | 0 | 0.00 | 0.00 | 0.00 | 0.00 | NaN | true | |
| 130 | resnext0_stage4_conv1_fwd | Convolution | [1,1024,14,14] | 3063.667 | 0 | 0 | 0 | 0 | 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) | 0.00 | 0 | 0.00 | 0.00 | 0.00 | 0.00 | NaN | true | |
| 130 | resnext0_stage4_conv1_fwd | Convolution | [1,1024,14,14] | 3063.667 | 0 | 0 | 0 | 0 | 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) | 0.00 | 0 | 0.00 | 0.00 | 0.00 | 0.00 | NaN | true | |
| 130 | resnext0_stage4_conv1_fwd | Convolution | [1,1024,14,14] | 3063.667 | 0 | 0 | 0 | 0 | 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) | 0.00 | 0 | 0.00 | 0.00 | 0.00 | 0.00 | NaN | true | |
| 130 | resnext0_stage4_conv1_fwd | Convolution | [1,1024,14,14] | 3063.667 | 0 | 0 | 0 | 0 | 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) | 0.00 | 0 | 0.00 | 0.00 | 0.00 | 0.00 | NaN | true | |
| 130 | resnext0_stage4_conv1_fwd | Convolution | [1,1024,14,14] | 3063.667 | 0 | 0 | 0 | 0 | 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) | 0.00 | 0 | 0.00 | 0.00 | 0.00 | 0.00 | NaN | true | |
| 130 | resnext0_stage4_conv1_fwd | Convolution | [1,1024,14,14] | 3063.667 | 0 | 0 | 0 | 0 | 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) | 0.00 | 0 | 0.00 | 0.00 | 0.00 | 0.00 | NaN | true | |
| 130 | resnext0_stage4_conv1_fwd | Convolution | [1,1024,14,14] | 3063.667 | 0 | 0 | 0 | 0 | 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) | 0.00 | 0 | 0.00 | 0.00 | 0.00 | 0.00 | NaN | true | |
| 130 | resnext0_stage4_conv1_fwd | Convolution | [1,1024,14,14] | 3063.667 | 0 | 0 | 0 | 0 | 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) | 0.00 | 0 | 0.00 | 0.00 | 0.00 | 0.00 | NaN | true | |
| 130 | resnext0_stage4_conv1_fwd | Convolution | [1,1024,14,14] | 3063.667 | 0 | 0 | 0 | 0 | 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) | 0.00 | 0 | 0.00 | 0.00 | 0.00 | 0.00 | NaN | true | |
| 130 | resnext0_stage4_conv1_fwd | Convolution | [1,1024,14,14] | 3063.667 | 0 | 0 | 0 | 0 | 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) | 0.00 | 0 | 0.00 | 0.00 | 0.00 | 0.00 | NaN | true | |
| 130 | resnext0_stage4_conv1_fwd | Convolution | [1,1024,14,14] | 3063.667 | 0 | 0 | 0 | 0 | 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) | 0.00 | 0 | 0.00 | 0.00 | 0.00 | 0.00 | NaN | true | |
| 130 | resnext0_stage4_conv1_fwd | Convolution | [1,1024,14,14] | 3063.667 | 0 | 0 | 0 | 0 | 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) | 0.00 | 0 | 0.00 | 0.00 | 0.00 | 0.00 | NaN | true | |
| 130 | resnext0_stage4_conv1_fwd | Convolution | [1,1024,14,14] | 3063.667 | 0 | 0 | 0 | 0 | 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) | 0.00 | 0 | 0.00 | 0.00 | 0.00 | 0.00 | NaN | true | |
| 130 | resnext0_stage4_conv1_fwd | Convolution | [1,1024,14,14] | 3063.667 | 0 | 0 | 0 | 0 | 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) | 0.00 | 0 | 0.00 | 0.00 | 0.00 | 0.00 | NaN | true | |
| 130 | resnext0_stage4_conv1_fwd | Convolution | [1,1024,14,14] | 3063.667 | 0 | 0 | 0 | 0 | 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) | 0.00 | 0 | 0.00 | 0.00 | 0.00 | 0.00 | NaN | true | |
| 130 | resnext0_stage4_conv1_fwd | Convolution | [1,1024,14,14] | 3063.667 | 0 | 0 | 0 | 0 | 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) | 0.00 | 0 | 0.00 | 0.00 | 0.00 | 0.00 | NaN | true | |
| 130 | resnext0_stage4_conv1_fwd | Convolution | [1,1024,14,14] | 3063.667 | 0 | 0 | 0 | 0 | 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) | 0.00 | 0 | 0.00 | 0.00 | 0.00 | 0.00 | NaN | true | |
| 130 | resnext0_stage4_conv1_fwd | Convolution | [1,1024,14,14] | 3063.667 | 0 | 0 | 0 | 0 | 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) | 0.00 | 0 | 0.00 | 0.00 | 0.00 | 0.00 | NaN | true | |
| 130 | resnext0_stage4_conv1_fwd | Convolution | [1,1024,14,14] | 3063.667 | 0 | 0 | 0 | 0 | 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) | 0.00 | 0 | 0.00 | 0.00 | 0.00 | 0.00 | NaN | true | |
| 130 | resnext0_stage4_conv1_fwd | Convolution | [1,1024,14,14] | 3063.667 | 0 | 0 | 0 | 0 | 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) | 0.00 | 0 | 0.00 | 0.00 | 0.00 | 0.00 | NaN | true | |
| 130 | resnext0_stage4_conv1_fwd | Convolution | [1,1024,14,14] | 3063.667 | 0 | 0 | 0 | 0 | 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) | 0.00 | 0 | 0.00 | 0.00 | 0.00 | 0.00 | NaN | true | |
| 130 | resnext0_stage4_conv1_fwd | Convolution | [1,1024,14,14] | 3063.667 | 0 | 0 | 0 | 0 | 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) | 0.00 | 0 | 0.00 | 0.00 | 0.00 | 0.00 | NaN | true | |
| 131 | resnext0_stage4_batchnorm1_fwd | BatchNorm | [1,1024,7,7] | 95 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | void cudnn::detail::bn_fw_inf_1C11_kernel_NCHW<float, float, true, 1>(float, float, cudnnTensorStruct, float const*, cudnnTensorStruct, float*, cudnnTensorStruct, float const*, float const*, float const*, float const*, float) | 0.00 | 0 | 0.00 | 0.00 | 0.00 | 0.00 | NaN | true | |
| 132 | resnext0_stage4_relu1_fwd | Activation | [1,1024,7,7] | 12 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | void op_generic_tensor_kernel<2, float, float, float, 256, (cudnnGenericOp_t)8, (cudnnNanPropagation_t)0, (cudnnDimOrder_t)0, 1>(cudnnTensorStruct, float*, cudnnTensorStruct, float const*, cudnnTensorStruct, float const*, float, float, float, float, dimArray, reducedDivisorArray, bool) | 0.00 | 0 | 0.00 | 0.00 | 0.00 | 0.00 | NaN | true | |
| 133 | resnext0_stage4_conv2_fwd | Convolution | [1,1024,7,7] | 13197 | 0 | 0 | 0 | 0 | 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) | 0.00 | 0 | 0.00 | 0.00 | 0.00 | 0.00 | NaN | true | |
| 134 | resnext0_stage4_batchnorm2_fwd | BatchNorm | [1,2048,7,7] | 114.667 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | void cudnn::detail::bn_fw_inf_1C11_kernel_NCHW<float, float, true, 1>(float, float, cudnnTensorStruct, float const*, cudnnTensorStruct, float*, cudnnTensorStruct, float const*, float const*, float const*, float const*, float) | 0.00 | 0 | 0.00 | 0.00 | 0.00 | 0.00 | NaN | true | |
| 135 | resnext0_stage4_conv3_fwd | Convolution | [1,1024,14,14] | 12975.333 | 0 | 0 | 0 | 0 | 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) | 0.00 | 0 | 0.00 | 0.00 | 0.00 | 0.00 | NaN | true | |
| 136 | resnext0_stage4_batchnorm3_fwd | BatchNorm | [1,2048,7,7] | 111.667 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | void cudnn::detail::bn_fw_inf_1C11_kernel_NCHW<float, float, true, 1>(float, float, cudnnTensorStruct, float const*, cudnnTensorStruct, float*, cudnnTensorStruct, float const*, float const*, float const*, float const*, float) | 0.00 | 0 | 0.00 | 0.00 | 0.00 | 0.00 | NaN | true | |
| 137 | add_resnext0_stage4_activation0 | add_relu | [1,2048,7,7] | 28 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | void mxnet::op::mxnet_op::mxnet_generic_kernel<mxnet::op::AddReluKernel, float*, float*, float*, mxnet::OpReqType>(int, float*, float*, float*, mxnet::OpReqType) | 0.00 | 0 | 0.00 | 0.00 | 0.00 | 0.00 | NaN | true | |
| 138 | resnext0_stage4_conv4_fwd | Convolution | [1,2048,7,7] | 13073.667 | 0 | 0 | 0 | 0 | 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) | 0.00 | 0 | 0.00 | 0.00 | 0.00 | 0.00 | NaN | true | |
| 139 | resnext0_stage4_batchnorm4_fwd | BatchNorm | [1,1024,7,7] | 106 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | void cudnn::detail::bn_fw_inf_1C11_kernel_NCHW<float, float, true, 1>(float, float, cudnnTensorStruct, float const*, cudnnTensorStruct, float*, cudnnTensorStruct, float const*, float const*, float const*, float const*, float) | 0.00 | 0 | 0.00 | 0.00 | 0.00 | 0.00 | NaN | true | |
| 140 | resnext0_stage4_relu2_fwd | Activation | [1,1024,7,7] | 14.333 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | void op_generic_tensor_kernel<2, float, float, float, 256, (cudnnGenericOp_t)8, (cudnnNanPropagation_t)0, (cudnnDimOrder_t)0, 1>(cudnnTensorStruct, float*, cudnnTensorStruct, float const*, cudnnTensorStruct, float const*, float, float, float, float, dimArray, reducedDivisorArray, bool) | 0.00 | 0 | 0.00 | 0.00 | 0.00 | 0.00 | NaN | true | |
| 141 | resnext0_stage4_conv5_fwd | Convolution | [1,1024,7,7] | 3050.667 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | void cudnn::winograd::generateWinogradTilesKernel<0, float, float>(cudnn::winograd::GenerateWinogradTilesParams<float, float>) | 0.00 | 0 | 0.00 | 0.00 | 0.00 | 0.00 | NaN | true | |
| 141 | resnext0_stage4_conv5_fwd | Convolution | [1,1024,7,7] | 3050.667 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | volta_scudnn_winograd_128x128_ldg1_ldg4_relu_tile148t_nt_v1 | 0.00 | 0 | 0.00 | 0.00 | 0.00 | 0.00 | NaN | true | |
| 141 | resnext0_stage4_conv5_fwd | Convolution | [1,1024,7,7] | 3050.667 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | void cudnn::winograd::generateWinogradTilesKernel<0, float, float>(cudnn::winograd::GenerateWinogradTilesParams<float, float>) | 0.00 | 0 | 0.00 | 0.00 | 0.00 | 0.00 | NaN | true | |
| 141 | resnext0_stage4_conv5_fwd | Convolution | [1,1024,7,7] | 3050.667 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | volta_scudnn_winograd_128x128_ldg1_ldg4_relu_tile148t_nt_v1 | 0.00 | 0 | 0.00 | 0.00 | 0.00 | 0.00 | NaN | true | |
| 141 | resnext0_stage4_conv5_fwd | Convolution | [1,1024,7,7] | 3050.667 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | void cudnn::winograd::generateWinogradTilesKernel<0, float, float>(cudnn::winograd::GenerateWinogradTilesParams<float, float>) | 0.00 | 0 | 0.00 | 0.00 | 0.00 | 0.00 | NaN | true | |
| 141 | resnext0_stage4_conv5_fwd | Convolution | [1,1024,7,7] | 3050.667 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | volta_scudnn_winograd_128x128_ldg1_ldg4_relu_tile148t_nt_v1 | 0.00 | 0 | 0.00 | 0.00 | 0.00 | 0.00 | NaN | true | |
| 141 | resnext0_stage4_conv5_fwd | Convolution | [1,1024,7,7] | 3050.667 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | void cudnn::winograd::generateWinogradTilesKernel<0, float, float>(cudnn::winograd::GenerateWinogradTilesParams<float, float>) | 0.00 | 0 | 0.00 | 0.00 | 0.00 | 0.00 | NaN | true | |
| 141 | resnext0_stage4_conv5_fwd | Convolution | [1,1024,7,7] | 3050.667 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | volta_scudnn_winograd_128x128_ldg1_ldg4_relu_tile148t_nt_v1 | 0.00 | 0 | 0.00 | 0.00 | 0.00 | 0.00 | NaN | true | |
| 141 | resnext0_stage4_conv5_fwd | Convolution | [1,1024,7,7] | 3050.667 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | void cudnn::winograd::generateWinogradTilesKernel<0, float, float>(cudnn::winograd::GenerateWinogradTilesParams<float, float>) | 0.00 | 0 | 0.00 | 0.00 | 0.00 | 0.00 | NaN | true | |
| 141 | resnext0_stage4_conv5_fwd | Convolution | [1,1024,7,7] | 3050.667 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | volta_scudnn_winograd_128x128_ldg1_ldg4_relu_tile148t_nt_v1 | 0.00 | 0 | 0.00 | 0.00 | 0.00 | 0.00 | NaN | true | |
| 141 | resnext0_stage4_conv5_fwd | Convolution | [1,1024,7,7] | 3050.667 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | void cudnn::winograd::generateWinogradTilesKernel<0, float, float>(cudnn::winograd::GenerateWinogradTilesParams<float, float>) | 0.00 | 0 | 0.00 | 0.00 | 0.00 | 0.00 | NaN | true | |
| 141 | resnext0_stage4_conv5_fwd | Convolution | [1,1024,7,7] | 3050.667 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | volta_scudnn_winograd_128x128_ldg1_ldg4_relu_tile148t_nt_v1 | 0.00 | 0 | 0.00 | 0.00 | 0.00 | 0.00 | NaN | true | |
| 141 | resnext0_stage4_conv5_fwd | Convolution | [1,1024,7,7] | 3050.667 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | void cudnn::winograd::generateWinogradTilesKernel<0, float, float>(cudnn::winograd::GenerateWinogradTilesParams<float, float>) | 0.00 | 0 | 0.00 | 0.00 | 0.00 | 0.00 | NaN | true | |
| 141 | resnext0_stage4_conv5_fwd | Convolution | [1,1024,7,7] | 3050.667 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | volta_scudnn_winograd_128x128_ldg1_ldg4_relu_tile148t_nt_v1 | 0.00 | 0 | 0.00 | 0.00 | 0.00 | 0.00 | NaN | true | |
| 141 | resnext0_stage4_conv5_fwd | Convolution | [1,1024,7,7] | 3050.667 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | void cudnn::winograd::generateWinogradTilesKernel<0, float, float>(cudnn::winograd::GenerateWinogradTilesParams<float, float>) | 0.00 | 0 | 0.00 | 0.00 | 0.00 | 0.00 | NaN | true | |
| 141 | resnext0_stage4_conv5_fwd | Convolution | [1,1024,7,7] | 3050.667 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | volta_scudnn_winograd_128x128_ldg1_ldg4_relu_tile148t_nt_v1 | 0.00 | 0 | 0.00 | 0.00 | 0.00 | 0.00 | NaN | true | |
| 141 | resnext0_stage4_conv5_fwd | Convolution | [1,1024,7,7] | 3050.667 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | void cudnn::winograd::generateWinogradTilesKernel<0, float, float>(cudnn::winograd::GenerateWinogradTilesParams<float, float>) | 0.00 | 0 | 0.00 | 0.00 | 0.00 | 0.00 | NaN | true | |
| 141 | resnext0_stage4_conv5_fwd | Convolution | [1,1024,7,7] | 3050.667 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | volta_scudnn_winograd_128x128_ldg1_ldg4_relu_tile148t_nt_v1 | 0.00 | 0 | 0.00 | 0.00 | 0.00 | 0.00 | NaN | true | |
| 141 | resnext0_stage4_conv5_fwd | Convolution | [1,1024,7,7] | 3050.667 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | void cudnn::winograd::generateWinogradTilesKernel<0, float, float>(cudnn::winograd::GenerateWinogradTilesParams<float, float>) | 0.00 | 0 | 0.00 | 0.00 | 0.00 | 0.00 | NaN | true | |
| 141 | resnext0_stage4_conv5_fwd | Convolution | [1,1024,7,7] | 3050.667 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | volta_scudnn_winograd_128x128_ldg1_ldg4_relu_tile148t_nt_v1 | 0.00 | 0 | 0.00 | 0.00 | 0.00 | 0.00 | NaN | true | |
| 141 | resnext0_stage4_conv5_fwd | Convolution | [1,1024,7,7] | 3050.667 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | void cudnn::winograd::generateWinogradTilesKernel<0, float, float>(cudnn::winograd::GenerateWinogradTilesParams<float, float>) | 0.00 | 0 | 0.00 | 0.00 | 0.00 | 0.00 | NaN | true | |
| 141 | resnext0_stage4_conv5_fwd | Convolution | [1,1024,7,7] | 3050.667 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | volta_scudnn_winograd_128x128_ldg1_ldg4_relu_tile148t_nt_v1 | 0.00 | 0 | 0.00 | 0.00 | 0.00 | 0.00 | NaN | true | |
| 141 | resnext0_stage4_conv5_fwd | Convolution | [1,1024,7,7] | 3050.667 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | void cudnn::winograd::generateWinogradTilesKernel<0, float, float>(cudnn::winograd::GenerateWinogradTilesParams<float, float>) | 0.00 | 0 | 0.00 | 0.00 | 0.00 | 0.00 | NaN | true | |
| 141 | resnext0_stage4_conv5_fwd | Convolution | [1,1024,7,7] | 3050.667 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | volta_scudnn_winograd_128x128_ldg1_ldg4_relu_tile148t_nt_v1 | 0.00 | 0 | 0.00 | 0.00 | 0.00 | 0.00 | NaN | true | |
| 141 | resnext0_stage4_conv5_fwd | Convolution | [1,1024,7,7] | 3050.667 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | void cudnn::winograd::generateWinogradTilesKernel<0, float, float>(cudnn::winograd::GenerateWinogradTilesParams<float, float>) | 0.00 | 0 | 0.00 | 0.00 | 0.00 | 0.00 | NaN | true | |
| 141 | resnext0_stage4_conv5_fwd | Convolution | [1,1024,7,7] | 3050.667 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | volta_scudnn_winograd_128x128_ldg1_ldg4_relu_tile148t_nt_v1 | 0.00 | 0 | 0.00 | 0.00 | 0.00 | 0.00 | NaN | true | |
| 141 | resnext0_stage4_conv5_fwd | Convolution | [1,1024,7,7] | 3050.667 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | void cudnn::winograd::generateWinogradTilesKernel<0, float, float>(cudnn::winograd::GenerateWinogradTilesParams<float, float>) | 0.00 | 0 | 0.00 | 0.00 | 0.00 | 0.00 | NaN | true | |
| 141 | resnext0_stage4_conv5_fwd | Convolution | [1,1024,7,7] | 3050.667 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | volta_scudnn_winograd_128x128_ldg1_ldg4_relu_tile148t_nt_v1 | 0.00 | 0 | 0.00 | 0.00 | 0.00 | 0.00 | NaN | true | |
| 141 | resnext0_stage4_conv5_fwd | Convolution | [1,1024,7,7] | 3050.667 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | void cudnn::winograd::generateWinogradTilesKernel<0, float, float>(cudnn::winograd::GenerateWinogradTilesParams<float, float>) | 0.00 | 0 | 0.00 | 0.00 | 0.00 | 0.00 | NaN | true | |
| 141 | resnext0_stage4_conv5_fwd | Convolution | [1,1024,7,7] | 3050.667 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | volta_scudnn_winograd_128x128_ldg1_ldg4_relu_tile148t_nt_v1 | 0.00 | 0 | 0.00 | 0.00 | 0.00 | 0.00 | NaN | true | |
| 141 | resnext0_stage4_conv5_fwd | Convolution | [1,1024,7,7] | 3050.667 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | void cudnn::winograd::generateWinogradTilesKernel<0, float, float>(cudnn::winograd::GenerateWinogradTilesParams<float, float>) | 0.00 | 0 | 0.00 | 0.00 | 0.00 | 0.00 | NaN | true | |
| 141 | resnext0_stage4_conv5_fwd | Convolution | [1,1024,7,7] | 3050.667 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | volta_scudnn_winograd_128x128_ldg1_ldg4_relu_tile148t_nt_v1 | 0.00 | 0 | 0.00 | 0.00 | 0.00 | 0.00 | NaN | true | |
| 141 | resnext0_stage4_conv5_fwd | Convolution | [1,1024,7,7] | 3050.667 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | void cudnn::winograd::generateWinogradTilesKernel<0, float, float>(cudnn::winograd::GenerateWinogradTilesParams<float, float>) | 0.00 | 0 | 0.00 | 0.00 | 0.00 | 0.00 | NaN | true | |
| 141 | resnext0_stage4_conv5_fwd | Convolution | [1,1024,7,7] | 3050.667 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | volta_scudnn_winograd_128x128_ldg1_ldg4_relu_tile148t_nt_v1 | 0.00 | 0 | 0.00 | 0.00 | 0.00 | 0.00 | NaN | true | |
| 141 | resnext0_stage4_conv5_fwd | Convolution | [1,1024,7,7] | 3050.667 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | void cudnn::winograd::generateWinogradTilesKernel<0, float, float>(cudnn::winograd::GenerateWinogradTilesParams<float, float>) | 0.00 | 0 | 0.00 | 0.00 | 0.00 | 0.00 | NaN | true | |
| 141 | resnext0_stage4_conv5_fwd | Convolution | [1,1024,7,7] | 3050.667 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | volta_scudnn_winograd_128x128_ldg1_ldg4_relu_tile148t_nt_v1 | 0.00 | 0 | 0.00 | 0.00 | 0.00 | 0.00 | NaN | true | |
| 141 | resnext0_stage4_conv5_fwd | Convolution | [1,1024,7,7] | 3050.667 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | void cudnn::winograd::generateWinogradTilesKernel<0, float, float>(cudnn::winograd::GenerateWinogradTilesParams<float, float>) | 0.00 | 0 | 0.00 | 0.00 | 0.00 | 0.00 | NaN | true | |
| 141 | resnext0_stage4_conv5_fwd | Convolution | [1,1024,7,7] | 3050.667 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | volta_scudnn_winograd_128x128_ldg1_ldg4_relu_tile148t_nt_v1 | 0.00 | 0 | 0.00 | 0.00 | 0.00 | 0.00 | NaN | true | |
| 141 | resnext0_stage4_conv5_fwd | Convolution | [1,1024,7,7] | 3050.667 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | void cudnn::winograd::generateWinogradTilesKernel<0, float, float>(cudnn::winograd::GenerateWinogradTilesParams<float, float>) | 0.00 | 0 | 0.00 | 0.00 | 0.00 | 0.00 | NaN | true | |
| 141 | resnext0_stage4_conv5_fwd | Convolution | [1,1024,7,7] | 3050.667 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | volta_scudnn_winograd_128x128_ldg1_ldg4_relu_tile148t_nt_v1 | 0.00 | 0 | 0.00 | 0.00 | 0.00 | 0.00 | NaN | true | |
| 141 | resnext0_stage4_conv5_fwd | Convolution | [1,1024,7,7] | 3050.667 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | void cudnn::winograd::generateWinogradTilesKernel<0, float, float>(cudnn::winograd::GenerateWinogradTilesParams<float, float>) | 0.00 | 0 | 0.00 | 0.00 | 0.00 | 0.00 | NaN | true | |
| 141 | resnext0_stage4_conv5_fwd | Convolution | [1,1024,7,7] | 3050.667 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | volta_scudnn_winograd_128x128_ldg1_ldg4_relu_tile148t_nt_v1 | 0.00 | 0 | 0.00 | 0.00 | 0.00 | 0.00 | NaN | true | |
| 141 | resnext0_stage4_conv5_fwd | Convolution | [1,1024,7,7] | 3050.667 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | void cudnn::winograd::generateWinogradTilesKernel<0, float, float>(cudnn::winograd::GenerateWinogradTilesParams<float, float>) | 0.00 | 0 | 0.00 | 0.00 | 0.00 | 0.00 | NaN | true | |
| 141 | resnext0_stage4_conv5_fwd | Convolution | [1,1024,7,7] | 3050.667 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | volta_scudnn_winograd_128x128_ldg1_ldg4_relu_tile148t_nt_v1 | 0.00 | 0 | 0.00 | 0.00 | 0.00 | 0.00 | NaN | true | |
| 141 | resnext0_stage4_conv5_fwd | Convolution | [1,1024,7,7] | 3050.667 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | void cudnn::winograd::generateWinogradTilesKernel<0, float, float>(cudnn::winograd::GenerateWinogradTilesParams<float, float>) | 0.00 | 0 | 0.00 | 0.00 | 0.00 | 0.00 | NaN | true | |
| 141 | resnext0_stage4_conv5_fwd | Convolution | [1,1024,7,7] | 3050.667 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | volta_scudnn_winograd_128x128_ldg1_ldg4_relu_tile148t_nt_v1 | 0.00 | 0 | 0.00 | 0.00 | 0.00 | 0.00 | NaN | true | |
| 141 | resnext0_stage4_conv5_fwd | Convolution | [1,1024,7,7] | 3050.667 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | void cudnn::winograd::generateWinogradTilesKernel<0, float, float>(cudnn::winograd::GenerateWinogradTilesParams<float, float>) | 0.00 | 0 | 0.00 | 0.00 | 0.00 | 0.00 | NaN | true | |
| 141 | resnext0_stage4_conv5_fwd | Convolution | [1,1024,7,7] | 3050.667 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | volta_scudnn_winograd_128x128_ldg1_ldg4_relu_tile148t_nt_v1 | 0.00 | 0 | 0.00 | 0.00 | 0.00 | 0.00 | NaN | true | |
| 141 | resnext0_stage4_conv5_fwd | Convolution | [1,1024,7,7] | 3050.667 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | void cudnn::winograd::generateWinogradTilesKernel<0, float, float>(cudnn::winograd::GenerateWinogradTilesParams<float, float>) | 0.00 | 0 | 0.00 | 0.00 | 0.00 | 0.00 | NaN | true | |
| 141 | resnext0_stage4_conv5_fwd | Convolution | [1,1024,7,7] | 3050.667 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | volta_scudnn_winograd_128x128_ldg1_ldg4_relu_tile148t_nt_v1 | 0.00 | 0 | 0.00 | 0.00 | 0.00 | 0.00 | NaN | true | |
| 141 | resnext0_stage4_conv5_fwd | Convolution | [1,1024,7,7] | 3050.667 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | void cudnn::winograd::generateWinogradTilesKernel<0, float, float>(cudnn::winograd::GenerateWinogradTilesParams<float, float>) | 0.00 | 0 | 0.00 | 0.00 | 0.00 | 0.00 | NaN | true | |
| 141 | resnext0_stage4_conv5_fwd | Convolution | [1,1024,7,7] | 3050.667 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | volta_scudnn_winograd_128x128_ldg1_ldg4_relu_tile148t_nt_v1 | 0.00 | 0 | 0.00 | 0.00 | 0.00 | 0.00 | NaN | true | |
| 141 | resnext0_stage4_conv5_fwd | Convolution | [1,1024,7,7] | 3050.667 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | void cudnn::winograd::generateWinogradTilesKernel<0, float, float>(cudnn::winograd::GenerateWinogradTilesParams<float, float>) | 0.00 | 0 | 0.00 | 0.00 | 0.00 | 0.00 | NaN | true | |
| 141 | resnext0_stage4_conv5_fwd | Convolution | [1,1024,7,7] | 3050.667 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | volta_scudnn_winograd_128x128_ldg1_ldg4_relu_tile148t_nt_v1 | 0.00 | 0 | 0.00 | 0.00 | 0.00 | 0.00 | NaN | true | |
| 141 | resnext0_stage4_conv5_fwd | Convolution | [1,1024,7,7] | 3050.667 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | void cudnn::winograd::generateWinogradTilesKernel<0, float, float>(cudnn::winograd::GenerateWinogradTilesParams<float, float>) | 0.00 | 0 | 0.00 | 0.00 | 0.00 | 0.00 | NaN | true | |
| 141 | resnext0_stage4_conv5_fwd | Convolution | [1,1024,7,7] | 3050.667 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | volta_scudnn_winograd_128x128_ldg1_ldg4_relu_tile148t_nt_v1 | 0.00 | 0 | 0.00 | 0.00 | 0.00 | 0.00 | NaN | true | |
| 141 | resnext0_stage4_conv5_fwd | Convolution | [1,1024,7,7] | 3050.667 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | void cudnn::winograd::generateWinogradTilesKernel<0, float, float>(cudnn::winograd::GenerateWinogradTilesParams<float, float>) | 0.00 | 0 | 0.00 | 0.00 | 0.00 | 0.00 | NaN | true | |
| 141 | resnext0_stage4_conv5_fwd | Convolution | [1,1024,7,7] | 3050.667 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | volta_scudnn_winograd_128x128_ldg1_ldg4_relu_tile148t_nt_v1 | 0.00 | 0 | 0.00 | 0.00 | 0.00 | 0.00 | NaN | true | |
| 141 | resnext0_stage4_conv5_fwd | Convolution | [1,1024,7,7] | 3050.667 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | void cudnn::winograd::generateWinogradTilesKernel<0, float, float>(cudnn::winograd::GenerateWinogradTilesParams<float, float>) | 0.00 | 0 | 0.00 | 0.00 | 0.00 | 0.00 | NaN | true | |
| 141 | resnext0_stage4_conv5_fwd | Convolution | [1,1024,7,7] | 3050.667 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | volta_scudnn_winograd_128x128_ldg1_ldg4_relu_tile148t_nt_v1 | 0.00 | 0 | 0.00 | 0.00 | 0.00 | 0.00 | NaN | true | |
| 141 | resnext0_stage4_conv5_fwd | Convolution | [1,1024,7,7] | 3050.667 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | void cudnn::winograd::generateWinogradTilesKernel<0, float, float>(cudnn::winograd::GenerateWinogradTilesParams<float, float>) | 0.00 | 0 | 0.00 | 0.00 | 0.00 | 0.00 | NaN | true | |
| 141 | resnext0_stage4_conv5_fwd | Convolution | [1,1024,7,7] | 3050.667 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | volta_scudnn_winograd_128x128_ldg1_ldg4_relu_tile148t_nt_v1 | 0.00 | 0 | 0.00 | 0.00 | 0.00 | 0.00 | NaN | true | |
| 141 | resnext0_stage4_conv5_fwd | Convolution | [1,1024,7,7] | 3050.667 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | void cudnn::winograd::generateWinogradTilesKernel<0, float, float>(cudnn::winograd::GenerateWinogradTilesParams<float, float>) | 0.00 | 0 | 0.00 | 0.00 | 0.00 | 0.00 | NaN | true | |
| 141 | resnext0_stage4_conv5_fwd | Convolution | [1,1024,7,7] | 3050.667 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | volta_scudnn_winograd_128x128_ldg1_ldg4_relu_tile148t_nt_v1 | 0.00 | 0 | 0.00 | 0.00 | 0.00 | 0.00 | NaN | true | |
| 142 | resnext0_stage4_batchnorm5_fwd | BatchNorm | [1,1024,7,7] | 27.667 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | void cudnn::detail::bn_fw_inf_1C11_kernel_NCHW<float, float, true, 1>(float, float, cudnnTensorStruct, float const*, cudnnTensorStruct, float*, cudnnTensorStruct, float const*, float const*, float const*, float const*, float) | 0.00 | 0 | 0.00 | 0.00 | 0.00 | 0.00 | NaN | true | |
| 143 | resnext0_stage4_relu3_fwd | Activation | [1,1024,7,7] | 12.667 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | void op_generic_tensor_kernel<2, float, float, float, 256, (cudnnGenericOp_t)8, (cudnnNanPropagation_t)0, (cudnnDimOrder_t)0, 1>(cudnnTensorStruct, float*, cudnnTensorStruct, float const*, cudnnTensorStruct, float const*, float, float, float, float, dimArray, reducedDivisorArray, bool) | 0.00 | 0 | 0.00 | 0.00 | 0.00 | 0.00 | NaN | true | |
| 144 | resnext0_stage4_conv6_fwd | Convolution | [1,1024,7,7] | 12946.667 | 0 | 0 | 0 | 0 | 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) | 0.00 | 0 | 0.00 | 0.00 | 0.00 | 0.00 | NaN | true | |
| 145 | resnext0_stage4_batchnorm6_fwd | BatchNorm | [1,2048,7,7] | 107.333 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | void cudnn::detail::bn_fw_inf_1C11_kernel_NCHW<float, float, true, 1>(float, float, cudnnTensorStruct, float const*, cudnnTensorStruct, float*, cudnnTensorStruct, float const*, float const*, float const*, float const*, float) | 0.00 | 0 | 0.00 | 0.00 | 0.00 | 0.00 | NaN | true | |
| 146 | add_resnext0_stage4_activation1 | add_relu | [1,2048,7,7] | 27.667 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | void mxnet::op::mxnet_op::mxnet_generic_kernel<mxnet::op::AddReluKernel, float*, float*, float*, mxnet::OpReqType>(int, float*, float*, float*, mxnet::OpReqType) | 0.00 | 0 | 0.00 | 0.00 | 0.00 | 0.00 | NaN | true | |
| 147 | resnext0_stage4_conv7_fwd | Convolution | [1,2048,7,7] | 13031.667 | 0 | 0 | 0 | 0 | 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) | 0.00 | 0 | 0.00 | 0.00 | 0.00 | 0.00 | NaN | true | |
| 148 | resnext0_stage4_batchnorm7_fwd | BatchNorm | [1,1024,7,7] | 103 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | void cudnn::detail::bn_fw_inf_1C11_kernel_NCHW<float, float, true, 1>(float, float, cudnnTensorStruct, float const*, cudnnTensorStruct, float*, cudnnTensorStruct, float const*, float const*, float const*, float const*, float) | 0.00 | 0 | 0.00 | 0.00 | 0.00 | 0.00 | NaN | true | |
| 149 | resnext0_stage4_relu4_fwd | Activation | [1,1024,7,7] | 12 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | void op_generic_tensor_kernel<2, float, float, float, 256, (cudnnGenericOp_t)8, (cudnnNanPropagation_t)0, (cudnnDimOrder_t)0, 1>(cudnnTensorStruct, float*, cudnnTensorStruct, float const*, cudnnTensorStruct, float const*, float, float, float, float, dimArray, reducedDivisorArray, bool) | 0.00 | 0 | 0.00 | 0.00 | 0.00 | 0.00 | NaN | true | |
| 150 | resnext0_stage4_conv8_fwd | Convolution | [1,1024,7,7] | 3042 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | void cudnn::winograd::generateWinogradTilesKernel<0, float, float>(cudnn::winograd::GenerateWinogradTilesParams<float, float>) | 0.00 | 0 | 0.00 | 0.00 | 0.00 | 0.00 | NaN | true | |
| 150 | resnext0_stage4_conv8_fwd | Convolution | [1,1024,7,7] | 3042 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | volta_scudnn_winograd_128x128_ldg1_ldg4_relu_tile148t_nt_v1 | 0.00 | 0 | 0.00 | 0.00 | 0.00 | 0.00 | NaN | true | |
| 150 | resnext0_stage4_conv8_fwd | Convolution | [1,1024,7,7] | 3042 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | void cudnn::winograd::generateWinogradTilesKernel<0, float, float>(cudnn::winograd::GenerateWinogradTilesParams<float, float>) | 0.00 | 0 | 0.00 | 0.00 | 0.00 | 0.00 | NaN | true | |
| 150 | resnext0_stage4_conv8_fwd | Convolution | [1,1024,7,7] | 3042 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | volta_scudnn_winograd_128x128_ldg1_ldg4_relu_tile148t_nt_v1 | 0.00 | 0 | 0.00 | 0.00 | 0.00 | 0.00 | NaN | true | |
| 150 | resnext0_stage4_conv8_fwd | Convolution | [1,1024,7,7] | 3042 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | void cudnn::winograd::generateWinogradTilesKernel<0, float, float>(cudnn::winograd::GenerateWinogradTilesParams<float, float>) | 0.00 | 0 | 0.00 | 0.00 | 0.00 | 0.00 | NaN | true | |
| 150 | resnext0_stage4_conv8_fwd | Convolution | [1,1024,7,7] | 3042 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | volta_scudnn_winograd_128x128_ldg1_ldg4_relu_tile148t_nt_v1 | 0.00 | 0 | 0.00 | 0.00 | 0.00 | 0.00 | NaN | true | |
| 150 | resnext0_stage4_conv8_fwd | Convolution | [1,1024,7,7] | 3042 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | void cudnn::winograd::generateWinogradTilesKernel<0, float, float>(cudnn::winograd::GenerateWinogradTilesParams<float, float>) | 0.00 | 0 | 0.00 | 0.00 | 0.00 | 0.00 | NaN | true | |
| 150 | resnext0_stage4_conv8_fwd | Convolution | [1,1024,7,7] | 3042 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | volta_scudnn_winograd_128x128_ldg1_ldg4_relu_tile148t_nt_v1 | 0.00 | 0 | 0.00 | 0.00 | 0.00 | 0.00 | NaN | true | |
| 150 | resnext0_stage4_conv8_fwd | Convolution | [1,1024,7,7] | 3042 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | void cudnn::winograd::generateWinogradTilesKernel<0, float, float>(cudnn::winograd::GenerateWinogradTilesParams<float, float>) | 0.00 | 0 | 0.00 | 0.00 | 0.00 | 0.00 | NaN | true | |
| 150 | resnext0_stage4_conv8_fwd | Convolution | [1,1024,7,7] | 3042 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | volta_scudnn_winograd_128x128_ldg1_ldg4_relu_tile148t_nt_v1 | 0.00 | 0 | 0.00 | 0.00 | 0.00 | 0.00 | NaN | true | |
| 150 | resnext0_stage4_conv8_fwd | Convolution | [1,1024,7,7] | 3042 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | void cudnn::winograd::generateWinogradTilesKernel<0, float, float>(cudnn::winograd::GenerateWinogradTilesParams<float, float>) | 0.00 | 0 | 0.00 | 0.00 | 0.00 | 0.00 | NaN | true | |
| 150 | resnext0_stage4_conv8_fwd | Convolution | [1,1024,7,7] | 3042 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | volta_scudnn_winograd_128x128_ldg1_ldg4_relu_tile148t_nt_v1 | 0.00 | 0 | 0.00 | 0.00 | 0.00 | 0.00 | NaN | true | |
| 150 | resnext0_stage4_conv8_fwd | Convolution | [1,1024,7,7] | 3042 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | void cudnn::winograd::generateWinogradTilesKernel<0, float, float>(cudnn::winograd::GenerateWinogradTilesParams<float, float>) | 0.00 | 0 | 0.00 | 0.00 | 0.00 | 0.00 | NaN | true | |
| 150 | resnext0_stage4_conv8_fwd | Convolution | [1,1024,7,7] | 3042 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | volta_scudnn_winograd_128x128_ldg1_ldg4_relu_tile148t_nt_v1 | 0.00 | 0 | 0.00 | 0.00 | 0.00 | 0.00 | NaN | true | |
| 150 | resnext0_stage4_conv8_fwd | Convolution | [1,1024,7,7] | 3042 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | void cudnn::winograd::generateWinogradTilesKernel<0, float, float>(cudnn::winograd::GenerateWinogradTilesParams<float, float>) | 0.00 | 0 | 0.00 | 0.00 | 0.00 | 0.00 | NaN | true | |
| 150 | resnext0_stage4_conv8_fwd | Convolution | [1,1024,7,7] | 3042 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | volta_scudnn_winograd_128x128_ldg1_ldg4_relu_tile148t_nt_v1 | 0.00 | 0 | 0.00 | 0.00 | 0.00 | 0.00 | NaN | true | |
| 150 | resnext0_stage4_conv8_fwd | Convolution | [1,1024,7,7] | 3042 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | void cudnn::winograd::generateWinogradTilesKernel<0, float, float>(cudnn::winograd::GenerateWinogradTilesParams<float, float>) | 0.00 | 0 | 0.00 | 0.00 | 0.00 | 0.00 | NaN | true | |
| 150 | resnext0_stage4_conv8_fwd | Convolution | [1,1024,7,7] | 3042 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | volta_scudnn_winograd_128x128_ldg1_ldg4_relu_tile148t_nt_v1 | 0.00 | 0 | 0.00 | 0.00 | 0.00 | 0.00 | NaN | true | |
| 150 | resnext0_stage4_conv8_fwd | Convolution | [1,1024,7,7] | 3042 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | void cudnn::winograd::generateWinogradTilesKernel<0, float, float>(cudnn::winograd::GenerateWinogradTilesParams<float, float>) | 0.00 | 0 | 0.00 | 0.00 | 0.00 | 0.00 | NaN | true | |
| 150 | resnext0_stage4_conv8_fwd | Convolution | [1,1024,7,7] | 3042 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | volta_scudnn_winograd_128x128_ldg1_ldg4_relu_tile148t_nt_v1 | 0.00 | 0 | 0.00 | 0.00 | 0.00 | 0.00 | NaN | true | |
| 150 | resnext0_stage4_conv8_fwd | Convolution | [1,1024,7,7] | 3042 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | void cudnn::winograd::generateWinogradTilesKernel<0, float, float>(cudnn::winograd::GenerateWinogradTilesParams<float, float>) | 0.00 | 0 | 0.00 | 0.00 | 0.00 | 0.00 | NaN | true | |
| 150 | resnext0_stage4_conv8_fwd | Convolution | [1,1024,7,7] | 3042 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | volta_scudnn_winograd_128x128_ldg1_ldg4_relu_tile148t_nt_v1 | 0.00 | 0 | 0.00 | 0.00 | 0.00 | 0.00 | NaN | true | |
| 150 | resnext0_stage4_conv8_fwd | Convolution | [1,1024,7,7] | 3042 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | void cudnn::winograd::generateWinogradTilesKernel<0, float, float>(cudnn::winograd::GenerateWinogradTilesParams<float, float>) | 0.00 | 0 | 0.00 | 0.00 | 0.00 | 0.00 | NaN | true | |
| 150 | resnext0_stage4_conv8_fwd | Convolution | [1,1024,7,7] | 3042 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | volta_scudnn_winograd_128x128_ldg1_ldg4_relu_tile148t_nt_v1 | 0.00 | 0 | 0.00 | 0.00 | 0.00 | 0.00 | NaN | true | |
| 150 | resnext0_stage4_conv8_fwd | Convolution | [1,1024,7,7] | 3042 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | void cudnn::winograd::generateWinogradTilesKernel<0, float, float>(cudnn::winograd::GenerateWinogradTilesParams<float, float>) | 0.00 | 0 | 0.00 | 0.00 | 0.00 | 0.00 | NaN | true | |
| 150 | resnext0_stage4_conv8_fwd | Convolution | [1,1024,7,7] | 3042 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | volta_scudnn_winograd_128x128_ldg1_ldg4_relu_tile148t_nt_v1 | 0.00 | 0 | 0.00 | 0.00 | 0.00 | 0.00 | NaN | true | |
| 150 | resnext0_stage4_conv8_fwd | Convolution | [1,1024,7,7] | 3042 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | void cudnn::winograd::generateWinogradTilesKernel<0, float, float>(cudnn::winograd::GenerateWinogradTilesParams<float, float>) | 0.00 | 0 | 0.00 | 0.00 | 0.00 | 0.00 | NaN | true | |
| 150 | resnext0_stage4_conv8_fwd | Convolution | [1,1024,7,7] | 3042 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | volta_scudnn_winograd_128x128_ldg1_ldg4_relu_tile148t_nt_v1 | 0.00 | 0 | 0.00 | 0.00 | 0.00 | 0.00 | NaN | true | |
| 150 | resnext0_stage4_conv8_fwd | Convolution | [1,1024,7,7] | 3042 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | void cudnn::winograd::generateWinogradTilesKernel<0, float, float>(cudnn::winograd::GenerateWinogradTilesParams<float, float>) | 0.00 | 0 | 0.00 | 0.00 | 0.00 | 0.00 | NaN | true | |
| 150 | resnext0_stage4_conv8_fwd | Convolution | [1,1024,7,7] | 3042 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | volta_scudnn_winograd_128x128_ldg1_ldg4_relu_tile148t_nt_v1 | 0.00 | 0 | 0.00 | 0.00 | 0.00 | 0.00 | NaN | true | |
| 150 | resnext0_stage4_conv8_fwd | Convolution | [1,1024,7,7] | 3042 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | void cudnn::winograd::generateWinogradTilesKernel<0, float, float>(cudnn::winograd::GenerateWinogradTilesParams<float, float>) | 0.00 | 0 | 0.00 | 0.00 | 0.00 | 0.00 | NaN | true | |
| 150 | resnext0_stage4_conv8_fwd | Convolution | [1,1024,7,7] | 3042 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | volta_scudnn_winograd_128x128_ldg1_ldg4_relu_tile148t_nt_v1 | 0.00 | 0 | 0.00 | 0.00 | 0.00 | 0.00 | NaN | true | |
| 150 | resnext0_stage4_conv8_fwd | Convolution | [1,1024,7,7] | 3042 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | void cudnn::winograd::generateWinogradTilesKernel<0, float, float>(cudnn::winograd::GenerateWinogradTilesParams<float, float>) | 0.00 | 0 | 0.00 | 0.00 | 0.00 | 0.00 | NaN | true | |
| 150 | resnext0_stage4_conv8_fwd | Convolution | [1,1024,7,7] | 3042 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | volta_scudnn_winograd_128x128_ldg1_ldg4_relu_tile148t_nt_v1 | 0.00 | 0 | 0.00 | 0.00 | 0.00 | 0.00 | NaN | true | |
| 150 | resnext0_stage4_conv8_fwd | Convolution | [1,1024,7,7] | 3042 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | void cudnn::winograd::generateWinogradTilesKernel<0, float, float>(cudnn::winograd::GenerateWinogradTilesParams<float, float>) | 0.00 | 0 | 0.00 | 0.00 | 0.00 | 0.00 | NaN | true | |
| 150 | resnext0_stage4_conv8_fwd | Convolution | [1,1024,7,7] | 3042 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | volta_scudnn_winograd_128x128_ldg1_ldg4_relu_tile148t_nt_v1 | 0.00 | 0 | 0.00 | 0.00 | 0.00 | 0.00 | NaN | true | |
| 150 | resnext0_stage4_conv8_fwd | Convolution | [1,1024,7,7] | 3042 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | void cudnn::winograd::generateWinogradTilesKernel<0, float, float>(cudnn::winograd::GenerateWinogradTilesParams<float, float>) | 0.00 | 0 | 0.00 | 0.00 | 0.00 | 0.00 | NaN | true | |
| 150 | resnext0_stage4_conv8_fwd | Convolution | [1,1024,7,7] | 3042 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | volta_scudnn_winograd_128x128_ldg1_ldg4_relu_tile148t_nt_v1 | 0.00 | 0 | 0.00 | 0.00 | 0.00 | 0.00 | NaN | true | |
| 150 | resnext0_stage4_conv8_fwd | Convolution | [1,1024,7,7] | 3042 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | void cudnn::winograd::generateWinogradTilesKernel<0, float, float>(cudnn::winograd::GenerateWinogradTilesParams<float, float>) | 0.00 | 0 | 0.00 | 0.00 | 0.00 | 0.00 | NaN | true | |
| 150 | resnext0_stage4_conv8_fwd | Convolution | [1,1024,7,7] | 3042 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | volta_scudnn_winograd_128x128_ldg1_ldg4_relu_tile148t_nt_v1 | 0.00 | 0 | 0.00 | 0.00 | 0.00 | 0.00 | NaN | true | |
| 150 | resnext0_stage4_conv8_fwd | Convolution | [1,1024,7,7] | 3042 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | void cudnn::winograd::generateWinogradTilesKernel<0, float, float>(cudnn::winograd::GenerateWinogradTilesParams<float, float>) | 0.00 | 0 | 0.00 | 0.00 | 0.00 | 0.00 | NaN | true | |
| 150 | resnext0_stage4_conv8_fwd | Convolution | [1,1024,7,7] | 3042 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | volta_scudnn_winograd_128x128_ldg1_ldg4_relu_tile148t_nt_v1 | 0.00 | 0 | 0.00 | 0.00 | 0.00 | 0.00 | NaN | true | |
| 150 | resnext0_stage4_conv8_fwd | Convolution | [1,1024,7,7] | 3042 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | void cudnn::winograd::generateWinogradTilesKernel<0, float, float>(cudnn::winograd::GenerateWinogradTilesParams<float, float>) | 0.00 | 0 | 0.00 | 0.00 | 0.00 | 0.00 | NaN | true | |
| 150 | resnext0_stage4_conv8_fwd | Convolution | [1,1024,7,7] | 3042 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | volta_scudnn_winograd_128x128_ldg1_ldg4_relu_tile148t_nt_v1 | 0.00 | 0 | 0.00 | 0.00 | 0.00 | 0.00 | NaN | true | |
| 150 | resnext0_stage4_conv8_fwd | Convolution | [1,1024,7,7] | 3042 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | void cudnn::winograd::generateWinogradTilesKernel<0, float, float>(cudnn::winograd::GenerateWinogradTilesParams<float, float>) | 0.00 | 0 | 0.00 | 0.00 | 0.00 | 0.00 | NaN | true | |
| 150 | resnext0_stage4_conv8_fwd | Convolution | [1,1024,7,7] | 3042 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | volta_scudnn_winograd_128x128_ldg1_ldg4_relu_tile148t_nt_v1 | 0.00 | 0 | 0.00 | 0.00 | 0.00 | 0.00 | NaN | true | |
| 150 | resnext0_stage4_conv8_fwd | Convolution | [1,1024,7,7] | 3042 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | void cudnn::winograd::generateWinogradTilesKernel<0, float, float>(cudnn::winograd::GenerateWinogradTilesParams<float, float>) | 0.00 | 0 | 0.00 | 0.00 | 0.00 | 0.00 | NaN | true | |
| 150 | resnext0_stage4_conv8_fwd | Convolution | [1,1024,7,7] | 3042 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | volta_scudnn_winograd_128x128_ldg1_ldg4_relu_tile148t_nt_v1 | 0.00 | 0 | 0.00 | 0.00 | 0.00 | 0.00 | NaN | true | |
| 150 | resnext0_stage4_conv8_fwd | Convolution | [1,1024,7,7] | 3042 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | void cudnn::winograd::generateWinogradTilesKernel<0, float, float>(cudnn::winograd::GenerateWinogradTilesParams<float, float>) | 0.00 | 0 | 0.00 | 0.00 | 0.00 | 0.00 | NaN | true | |
| 150 | resnext0_stage4_conv8_fwd | Convolution | [1,1024,7,7] | 3042 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | volta_scudnn_winograd_128x128_ldg1_ldg4_relu_tile148t_nt_v1 | 0.00 | 0 | 0.00 | 0.00 | 0.00 | 0.00 | NaN | true | |
| 150 | resnext0_stage4_conv8_fwd | Convolution | [1,1024,7,7] | 3042 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | void cudnn::winograd::generateWinogradTilesKernel<0, float, float>(cudnn::winograd::GenerateWinogradTilesParams<float, float>) | 0.00 | 0 | 0.00 | 0.00 | 0.00 | 0.00 | NaN | true | |
| 150 | resnext0_stage4_conv8_fwd | Convolution | [1,1024,7,7] | 3042 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | volta_scudnn_winograd_128x128_ldg1_ldg4_relu_tile148t_nt_v1 | 0.00 | 0 | 0.00 | 0.00 | 0.00 | 0.00 | NaN | true | |
| 150 | resnext0_stage4_conv8_fwd | Convolution | [1,1024,7,7] | 3042 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | void cudnn::winograd::generateWinogradTilesKernel<0, float, float>(cudnn::winograd::GenerateWinogradTilesParams<float, float>) | 0.00 | 0 | 0.00 | 0.00 | 0.00 | 0.00 | NaN | true | |
| 150 | resnext0_stage4_conv8_fwd | Convolution | [1,1024,7,7] | 3042 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | volta_scudnn_winograd_128x128_ldg1_ldg4_relu_tile148t_nt_v1 | 0.00 | 0 | 0.00 | 0.00 | 0.00 | 0.00 | NaN | true | |
| 150 | resnext0_stage4_conv8_fwd | Convolution | [1,1024,7,7] | 3042 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | void cudnn::winograd::generateWinogradTilesKernel<0, float, float>(cudnn::winograd::GenerateWinogradTilesParams<float, float>) | 0.00 | 0 | 0.00 | 0.00 | 0.00 | 0.00 | NaN | true | |
| 150 | resnext0_stage4_conv8_fwd | Convolution | [1,1024,7,7] | 3042 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | volta_scudnn_winograd_128x128_ldg1_ldg4_relu_tile148t_nt_v1 | 0.00 | 0 | 0.00 | 0.00 | 0.00 | 0.00 | NaN | true | |
| 150 | resnext0_stage4_conv8_fwd | Convolution | [1,1024,7,7] | 3042 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | void cudnn::winograd::generateWinogradTilesKernel<0, float, float>(cudnn::winograd::GenerateWinogradTilesParams<float, float>) | 0.00 | 0 | 0.00 | 0.00 | 0.00 | 0.00 | NaN | true | |
| 150 | resnext0_stage4_conv8_fwd | Convolution | [1,1024,7,7] | 3042 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | volta_scudnn_winograd_128x128_ldg1_ldg4_relu_tile148t_nt_v1 | 0.00 | 0 | 0.00 | 0.00 | 0.00 | 0.00 | NaN | true | |
| 150 | resnext0_stage4_conv8_fwd | Convolution | [1,1024,7,7] | 3042 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | void cudnn::winograd::generateWinogradTilesKernel<0, float, float>(cudnn::winograd::GenerateWinogradTilesParams<float, float>) | 0.00 | 0 | 0.00 | 0.00 | 0.00 | 0.00 | NaN | true | |
| 150 | resnext0_stage4_conv8_fwd | Convolution | [1,1024,7,7] | 3042 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | volta_scudnn_winograd_128x128_ldg1_ldg4_relu_tile148t_nt_v1 | 0.00 | 0 | 0.00 | 0.00 | 0.00 | 0.00 | NaN | true | |
| 150 | resnext0_stage4_conv8_fwd | Convolution | [1,1024,7,7] | 3042 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | void cudnn::winograd::generateWinogradTilesKernel<0, float, float>(cudnn::winograd::GenerateWinogradTilesParams<float, float>) | 0.00 | 0 | 0.00 | 0.00 | 0.00 | 0.00 | NaN | true | |
| 150 | resnext0_stage4_conv8_fwd | Convolution | [1,1024,7,7] | 3042 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | volta_scudnn_winograd_128x128_ldg1_ldg4_relu_tile148t_nt_v1 | 0.00 | 0 | 0.00 | 0.00 | 0.00 | 0.00 | NaN | true | |
| 150 | resnext0_stage4_conv8_fwd | Convolution | [1,1024,7,7] | 3042 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | void cudnn::winograd::generateWinogradTilesKernel<0, float, float>(cudnn::winograd::GenerateWinogradTilesParams<float, float>) | 0.00 | 0 | 0.00 | 0.00 | 0.00 | 0.00 | NaN | true | |
| 150 | resnext0_stage4_conv8_fwd | Convolution | [1,1024,7,7] | 3042 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | volta_scudnn_winograd_128x128_ldg1_ldg4_relu_tile148t_nt_v1 | 0.00 | 0 | 0.00 | 0.00 | 0.00 | 0.00 | NaN | true | |
| 151 | resnext0_stage4_batchnorm8_fwd | BatchNorm | [1,1024,7,7] | 30.667 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | void cudnn::detail::bn_fw_inf_1C11_kernel_NCHW<float, float, true, 1>(float, float, cudnnTensorStruct, float const*, cudnnTensorStruct, float*, cudnnTensorStruct, float const*, float const*, float const*, float const*, float) | 0.00 | 0 | 0.00 | 0.00 | 0.00 | 0.00 | NaN | true | |
| 152 | resnext0_stage4_relu5_fwd | Activation | [1,1024,7,7] | 12.333 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | void op_generic_tensor_kernel<2, float, float, float, 256, (cudnnGenericOp_t)8, (cudnnNanPropagation_t)0, (cudnnDimOrder_t)0, 1>(cudnnTensorStruct, float*, cudnnTensorStruct, float const*, cudnnTensorStruct, float const*, float, float, float, float, dimArray, reducedDivisorArray, bool) | 0.00 | 0 | 0.00 | 0.00 | 0.00 | 0.00 | NaN | true | |
| 153 | resnext0_stage4_conv9_fwd | Convolution | [1,1024,7,7] | 13024.333 | 0 | 0 | 0 | 0 | 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) | 0.00 | 0 | 0.00 | 0.00 | 0.00 | 0.00 | NaN | true | |
| 154 | resnext0_stage4_batchnorm9_fwd | BatchNorm | [1,2048,7,7] | 118 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | void cudnn::detail::bn_fw_inf_1C11_kernel_NCHW<float, float, true, 1>(float, float, cudnnTensorStruct, float const*, cudnnTensorStruct, float*, cudnnTensorStruct, float const*, float const*, float const*, float const*, float) | 0.00 | 0 | 0.00 | 0.00 | 0.00 | 0.00 | NaN | true | |
| 155 | add_resnext0_stage4_activation2 | add_relu | [1,2048,7,7] | 26.667 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | void mxnet::op::mxnet_op::mxnet_generic_kernel<mxnet::op::AddReluKernel, float*, float*, float*, mxnet::OpReqType>(int, float*, float*, float*, mxnet::OpReqType) | 0.00 | 0 | 0.00 | 0.00 | 0.00 | 0.00 | NaN | true | |
| 156 | resnext0_pool1_fwd | Pooling | [1,2048,7,7] | 155 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | void cudnn::detail::pooling_fw_4d_kernel<float, float, cudnn::detail::averpooling_func<float>, 1, false>(cudnnTensorStruct, float const*, cudnnTensorStruct, float*, cudnnPoolingStruct, float, float, int, cudnn::reduced_divisor, cudnn::reduced_divisor) | 0.00 | 0 | 0.00 | 0.00 | 0.00 | 0.00 | NaN | true | |
| 157 | resnext0_dense0_fwd | FullyConnected | [1,2048,1,1] | 2093.667 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | void gemv2T_kernel_val<int, int, float, float, float, 128, 16, 4, 4, false, cublasGemvParams<cublasGemvTensorStridedBatched<float const>, cublasGemvTensorStridedBatched<float>, float> >(cublasGemvParams<cublasGemvTensorStridedBatched<float const>, cublasGemvTensorStridedBatched<float>, float>, float, float) | 0.00 | 0 | 0.00 | 0.00 | 0.00 | 0.00 | NaN | true | |
| 157 | resnext0_dense0_fwd | FullyConnected | [1,2048,1,1] | 2093.667 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | void mshadow::cuda::MapPlanKernel<mshadow::sv::plusto, 8, mshadow::expr::Plan<mshadow::Tensor<mshadow::gpu, 2, float>, float>, mshadow::expr::Plan<mshadow::expr::Broadcast1DExp<mshadow::Tensor<mshadow::gpu, 1, float>, float, 2, 1>, float> >(mshadow::expr::Plan<mshadow::Tensor<mshadow::gpu, 2, float>, float>, int, mshadow::Shape<2>, mshadow::expr::Plan<mshadow::expr::Broadcast1DExp<mshadow::Tensor<mshadow::gpu, 1, float>, float, 2, 1>, float>) | 0.00 | 0 | 0.00 | 0.00 | 0.00 | 0.00 | NaN | true |
Showing 1 to 803 of 803 entries