MCPcopy Create free account

hub / github.com/Maratyszcza/NNPACK / functions

Functions873 in github.com/Maratyszcza/NNPACK

↓ 2 callersFunctionnnp_relu_input_gradient
src/relu-input-gradient.c:35
↓ 2 callersFunctionnnp_softmax_output
src/softmax-output.c:50
↓ 2 callersFunctionnnp_softmax_output__reference
src/ref/softmax-output.c:49
↓ 2 callersMethodoutputSize
test/testers/winograd.h:44
↓ 2 callersFunctionpool1
* VGG model A pool1 layer: * channels = 64 * input size = 224x224 * implicit padding = 0 * pooling size = 2x2 *
test/models/vgg-a.h:262
↓ 2 callersFunctionpool2
* VGG model A pool2 layer: * channels = 128 * input size = 112x112 * implicit padding = 0 * pooling size = 2x2 *
test/models/vgg-a.h:279
↓ 2 callersFunctionpool3
* VGG model A pool3 layer: * channels = 256 * input size = 56x56 * implicit padding = 0 * pooling size = 2x2 * p
test/models/vgg-a.h:296
↓ 2 callersFunctionpsimd_fft16_dualreal_f32
src/psimd/fft/dualreal.h:56
↓ 2 callersFunctionpsimd_fft4_aos_f32
src/psimd/fft/aos.h:12
↓ 2 callersFunctionpsimd_fft8_aos_f32
src/psimd/fft/aos.h:185
↓ 2 callersFunctionpsimd_fft8_dualreal_f32
src/psimd/fft/dualreal.h:9
↓ 2 callersFunctionpsimd_ifft4_aos_f32
src/psimd/fft/aos.h:125
↓ 2 callersFunctionpsimd_ifft8_aos_f32
src/psimd/fft/aos.h:420
↓ 2 callersFunctionscalar_fft16_dualreal
src/scalar/fft/dualreal.h:51
↓ 2 callersFunctionscalar_fft16_real
src/scalar/fft/real.h:49
↓ 2 callersFunctionscalar_fft4_aos
src/scalar/fft/aos.h:10
↓ 2 callersFunctionscalar_fft8_aos
src/scalar/fft/aos.h:183
↓ 2 callersFunctionscalar_fft8_dualreal
src/scalar/fft/dualreal.h:7
↓ 2 callersFunctionscalar_fft8_real
src/scalar/fft/real.h:7
↓ 2 callersFunctionscalar_ifft4_aos
src/scalar/fft/aos.h:123
↓ 2 callersFunctionscalar_ifft8_aos
src/scalar/fft/aos.h:418
↓ 2 callersFunctionscaled_exp_minus_c
(reg_n, reg_x, reg_y, ymm_scale, ymm_c)
src/x86_64-fma/softmax.py:188
↓ 2 callersFunctionscaled_exp_minus_c__scalar
src/psimd/softmax.c:92
↓ 2 callersFunctionscaled_exp_minus_c__scalar
src/scalar/softmax.c:25
↓ 2 callersFunctionsum_exp_minus_c__psimd
src/psimd/softmax.c:62
↓ 2 callersFunctionsum_exp_minus_c__scalar
src/psimd/softmax.c:54
↓ 2 callersFunctionsum_exp_minus_c__scalar
src/scalar/softmax.c:17
↓ 2 callersFunctionulpf
src/x86_64-fma/exp.c:67
↓ 2 callersFunctionvalidate_relu_arguments
include/nnpack/validation.h:142
↓ 1 callersFunctionaverage
bench/median.c:29
↓ 1 callersFunctionaverage_profile
bench/median.c:33
↓ 1 callersFunctionbenchmark_batch_transform
bench/transform.c:124
↓ 1 callersFunctionbenchmark_convolution
bench/convolution.c:37
↓ 1 callersFunctionbenchmark_fully_connected
bench/fully-connected.c:22
↓ 1 callersFunctionbenchmark_gemm
bench/gemm.c:31
↓ 1 callersFunctionbenchmark_pooling
bench/pooling.c:16
↓ 1 callersFunctionbenchmark_relu
bench/relu.c:23
↓ 1 callersFunctioncompute_direct_convolution_inference
src/convolution-inference.c:953
↓ 1 callersFunctioncompute_fast_convolution_inference
src/convolution-inference.c:470
↓ 1 callersFunctioncompute_fast_convolution_input_gradient
src/convolution-input-gradient.c:238
↓ 1 callersFunctioncompute_fast_convolution_kernel_gradient
src/convolution-kernel-gradient.c:222
↓ 1 callersFunctioncompute_fast_convolution_output
src/convolution-output.c:235
↓ 1 callersFunctioncompute_fully_connected_output
src/fully-connected-output.c:140
↓ 1 callersFunctioncompute_gemm_convolution_inference
src/convolution-inference.c:736
↓ 1 callersFunctionconvolution
(mode, batch_size, input_channels, output_channels, image_size, kernel_size, padding, algorithm, transform_str
benchmark.py:13
↓ 1 callersFunctionfully_connected
(mode, batch_size, input_channels, output_channels, threads=None, verbose=False, use_selldr=False)
benchmark.py:57
↓ 1 callersFunctionget_x86_cpu_info
bench/perf_counter.c:22
↓ 1 callersFunctioninit_static_hwinfo
src/init.c:61
↓ 1 callersFunctioninit_static_ios_hwinfo
src/init.c:84
↓ 1 callersFunctioninit_x86_hwinfo
src/init.c:21
↓ 1 callersFunctioninplace_scaled_exp_minus_c__psimd
src/psimd/softmax.c:98
↓ 1 callersFunctionmain
(args)
configure.py:16
↓ 1 callersFunctionneon_grad_reluq_f32
include/nnpack/activations.h:32
↓ 1 callersFunctionnnp_conv1x1_only_2x4__psimd
src/psimd/blas/conv1x1.c:14
↓ 1 callersFunctionnnp_conv1x1_only_2x4__scalar
src/scalar/blas/conv1x1.c:5
↓ 1 callersFunctionnnp_conv1x1_only_4x4__neon
src/neon/blas/conv1x1.c:8
↓ 1 callersFunctionnnp_convolution_input_gradient__reference
src/ref/convolution-input-gradient.c:56
↓ 1 callersFunctionnnp_convolution_kernel_gradient__reference
src/ref/convolution-kernel.c:71
↓ 1 callersFunctionnnp_deinitialize
src/init.c:613
↓ 1 callersFunctionnnp_fully_connected_output_f16f32__reference
src/ref/fully-connected-output.c:75
↓ 1 callersFunctionnnp_h4gemm_only_3x3__neonhp
src/neon/blas/h4gemm.c:8
↓ 1 callersFunctionnnp_h4gemm_upto_3x3__neonhp
src/neon/blas/h4gemm.c:71
↓ 1 callersFunctionnnp_iwt8x8_3x3_fp16_with_offset__neonhp
src/neon/2d-winograd-8x8-3x3-fp16.c:13
↓ 1 callersFunctionnnp_iwt8x8_3x3_with_offset__neon
src/neon/2d-winograd-8x8-3x3.c:13
↓ 1 callersFunctionnnp_max_pooling_output__reference
src/ref/max-pooling-output.c:51
↓ 1 callersFunctionnnp_owt8x8_3x3__neon
src/neon/2d-winograd-8x8-3x3.c:186
↓ 1 callersFunctionnnp_owt8x8_3x3_fp16__neonhp
src/neon/2d-winograd-8x8-3x3-fp16.c:136
↓ 1 callersFunctionnnp_relu_input_gradient__reference
src/ref/relu-input-gradient.c:29
↓ 1 callersFunctionnnp_s2gemm_only_2x2__scalar
src/scalar/blas/s2gemm.c:5
↓ 1 callersFunctionnnp_s4gemm_only_3x3__neon
src/neon/blas/s4gemm.c:8
↓ 1 callersFunctionnnp_s4gemm_only_3x4__psimd
src/psimd/blas/s4gemm.c:7
↓ 1 callersFunctionnnp_s4gemm_upto_3x3__neon
src/neon/blas/s4gemm.c:67
↓ 1 callersFunctionnnp_sgemm_only_4x3__scalar
src/scalar/blas/sgemm.c:7
↓ 1 callersFunctionnnp_sgemm_only_4x8__psimd
src/psimd/blas/sgemm.c:9
↓ 1 callersFunctionnnp_sgemm_only_6x8__neon
src/neon/blas/sgemm.c:8
↓ 1 callersFunctionoutplace_scaled_exp_minus_c__psimd
src/psimd/softmax.c:125
↓ 1 callersMethodoutputHeight
test/testers/pooling.h:165
↓ 1 callersMethodoutputHeight
test/testers/convolution.h:168
↓ 1 callersMethodoutputWidth
test/testers/pooling.h:171
↓ 1 callersMethodoutputWidth
test/testers/convolution.h:172
↓ 1 callersFunctionparse_options
bench/convolution.c:253
↓ 1 callersFunctionparse_options
bench/ugemm.c:111
↓ 1 callersFunctionparse_options
bench/gemm.c:120
↓ 1 callersFunctionparse_options
bench/pooling.c:85
↓ 1 callersFunctionparse_options
bench/relu.c:96
↓ 1 callersFunctionparse_options
bench/transform.c:359
↓ 1 callersFunctionparse_options
bench/fully-connected.c:126
↓ 1 callersFunctionperf_event_open
bench/perf_counter.c:630
↓ 1 callersFunctionpool4
* VGG model A pool4 layer: * channels = 512 * input size = 28x28 * implicit padding = 0 * pooling size = 2x2 * p
test/models/vgg-a.h:313
↓ 1 callersFunctionpool5
* VGG model A pool5 layer: * channels = 512 * input size = 14x14 * implicit padding = 0 * pooling size = 2x2 * p
test/models/vgg-a.h:330
↓ 1 callersFunctionprofile_batch_fft
bench/transform.c:250
↓ 1 callersFunctionprofile_gemm
bench/ugemm.c:24
↓ 1 callersFunctionpsimd_grad_relu_f32
include/nnpack/activations.h:19
↓ 1 callersFunctionround_up_by_power_of_2
include/nnpack/utils.h:29
↓ 1 callersFunctionselect_algorithm
src/convolution-inference.c:1022
↓ 1 callersFunctionswitch
Add bias */
src/convolution-inference.c:883
↓ 1 callersFunctiontranspose8x3
(xmm_rows)
src/x86_64-fma/winograd/o6x6k3x3.py:329
↓ 1 callersFunctiontranspose8x8
(ymm_rows)
src/x86_64-fma/winograd/o6x6k3x3.py:225
↓ 1 callersFunctionvalidate_pooling_arguments
include/nnpack/validation.h:102
↓ 1 callersFunctionvalidate_softmax_arguments
include/nnpack/validation.h:164
← previousnext →201–300 of 873, ranked by callers