Files
Raptor/validation/operations/validation_results.csv
T

17 KiB

1OperationResultCompileHost memCores memCoresXbarsLatencyPowerEnergy
2add/after_gemmPASS0.072 s0.01 MiB0.01 MiB540.007784 ms104.703618 mW815012.960000 pJ
3add/basicPASS0.049 s0.00 MiB0.00 MiB100.000323 ms78.222910 mW25266.000000 pJ
4add/broadcast_rowPASS0.048 s0.00 MiB0.00 MiB100.000323 ms78.222910 mW25266.000000 pJ
5add/channel_broadcast_1024PASS0.051 s0.02 MiB0.01 MiB100.006913 ms78.118038 mW540030.000000 pJ
6add/leading_dimension_broadcastPASS0.051 s0.00 MiB0.00 MiB100.000323 ms78.222910 mW25266.000000 pJ
7concat/channel_axisPASS0.051 s0.00 MiB0.00 MiB100.000457 ms78.157549 mW35718.000000 pJ
8concat/negative_axisPASS0.061 s0.00 MiB0.00 MiB100.001043 ms78.092042 mW81450.000000 pJ
9concat/three_inputs_channel_axisPASS0.054 s0.00 MiB0.00 MiB100.000644 ms78.149068 mW50328.000000 pJ
10conv/batch_2PASS0.059 s0.00 MiB0.00 MiB220.013694 ms82.623885 mW1131451.480000 pJ
11conv/batch_4_pointwisePASS0.056 s0.00 MiB0.01 MiB540.003932 ms116.078576 mW456420.960000 pJ
12conv/depthwise_1024_channelsPASS0.081 s0.19 MiB0.38 MiB1291280.220751 ms178.454307 mW39393966.720000 pJ
13conv/depthwise_groupedPASS0.059 s0.01 MiB0.00 MiB540.006024 ms108.326521 mW652558.960000 pJ
14conv/dilated_3x3PASS0.059 s0.00 MiB0.00 MiB330.004045 ms110.234541 mW445898.720000 pJ
15conv/dynamicPASS0.059 s0.00 MiB0.00 MiB500.001835 ms92.281199 mW169336.000000 pJ
16conv/explicit_paddingPASS0.058 s0.00 MiB0.00 MiB440.004327 ms115.794768 mW501043.960000 pJ
17conv/grouped_many_groupsPASS0.505 s0.05 MiB0.09 MiB65640.181845 ms142.210104 mW25860196.360000 pJ
18conv/grouped_two_groupsPASS0.061 s0.00 MiB0.00 MiB320.005360 ms101.459418 mW543822.480000 pJ
19conv/huge_pointwise_1024PASS0.643 s0.01 MiB0.01 MiB1640.028261 ms133.488743 mW3772525.360000 pJ
20conv/huge_pointwise_1024_dynamicPASS0.081 s8.04 MiB12.61 MiB16802.627964 ms169.518697 mW445489032.000000 pJ
21conv/kernel_3x3PASS0.056 s0.00 MiB0.00 MiB330.003091 ms115.862414 mW358130.720000 pJ
22conv/kernel_equals_input_spatialPASS0.060 s0.00 MiB0.00 MiB120.008443 ms83.863849 mW708062.480000 pJ
23conv/large_input_channels_1x1PASS0.088 s0.01 MiB0.01 MiB180.017167 ms89.416900 mW1535019.920000 pJ
24conv/large_output_channels_1x1PASS0.087 s0.00 MiB0.01 MiB180.004964 ms117.628106 mW583905.920000 pJ
25conv/large_spatialPASS0.055 s0.00 MiB0.01 MiB660.004096 ms129.015000 mW528445.440000 pJ
26conv/multi_channelPASS0.056 s0.00 MiB0.00 MiB330.005148 ms106.453520 mW548022.720000 pJ
27conv/non_square_kernel_1x3PASS0.056 s0.00 MiB0.00 MiB550.004029 ms123.600943 mW497988.200000 pJ
28conv/non_square_kernel_3x1PASS0.058 s0.00 MiB0.00 MiB330.005526 ms105.464843 mW582798.720000 pJ
29conv/non_uniform_stridePASS0.062 s0.00 MiB0.00 MiB440.005808 ms110.081433 mW639352.960000 pJ
30conv/pointwise_1x1PASS0.057 s0.00 MiB0.00 MiB440.004539 ms114.835858 mW521239.960000 pJ
31conv/pointwise_tiled_chainPASS0.771 s0.01 MiB0.02 MiB2800.084437 ms102.289307 mW8637002.200000 pJ
32conv/real_asymmetric_paddingPASS0.055 s0.00 MiB0.00 MiB440.005232 ms111.870214 mW585304.960000 pJ
33conv/relu_conv_storePASS0.076 s0.02 MiB0.10 MiB32320.064390 ms243.916397 mW15705776.800000 pJ
34conv/same_lower_3x3PASS0.057 s0.00 MiB0.00 MiB550.004700 ms119.232170 mW560391.200000 pJ
35conv/same_padding_3x3PASS0.060 s0.00 MiB0.00 MiB550.004700 ms119.232170 mW560391.200000 pJ
36conv/simplePASS0.055 s0.00 MiB0.00 MiB220.003148 ms94.665972 mW298008.480000 pJ
37conv/stride_2PASS0.053 s0.00 MiB0.00 MiB220.002827 ms96.393873 mW272505.480000 pJ
38conv/with_bias_3x3PASS0.055 s0.00 MiB0.00 MiB330.004898 ms107.176546 mW524950.720000 pJ
39conv/with_constantPASS0.073 s0.00 MiB0.00 MiB330.004273 ms109.362677 mW467306.720000 pJ
40conv/without_kernel_shape_attrPASS0.057 s0.00 MiB0.00 MiB330.003091 ms115.862414 mW358130.720000 pJ
41conv/yolo11n_depthwise_headPASS0.509 s4.82 MiB15.92 MiB1607204.105780 ms492.826315 mW2023436428.000010 pJ
42conv/yolo11n_heavyPASS0.488 s4.82 MiB20.66 MiB1608006.275385 ms418.028199 mW2623287892.000010 pJ
43conv/yolo11n_stemPASS1.730 s12.86 MiB31.38 MiB1684889.799204 ms361.075380 mW3538251304.000010 pJ
44div/after_gemmPASS0.059 s0.01 MiB0.01 MiB540.007784 ms104.703618 mW815012.960000 pJ
45div/basicPASS0.048 s0.00 MiB0.00 MiB100.000323 ms78.222910 mW25266.000000 pJ
46div/channel_broadcast_1024PASS0.051 s0.02 MiB0.01 MiB100.006913 ms78.118038 mW540030.000000 pJ
47div/leading_dimension_broadcastPASS0.063 s0.00 MiB0.00 MiB100.000323 ms78.222910 mW25266.000000 pJ
48div/runtime_scalar_rhsPASS0.057 s0.02 MiB0.01 MiB100.006913 ms78.118038 mW540030.000000 pJ
49div/scalar_constantPASS0.049 s0.00 MiB0.00 MiB100.000323 ms78.222910 mW25266.000000 pJ
50gather/3d_input_axis1PASS0.050 s0.00 MiB0.00 MiB100.000589 ms78.081494 mW45990.000000 pJ
51gather/axis0_matrix_indicesPASS0.050 s0.00 MiB0.00 MiB100.000697 ms78.068867 mW54414.000000 pJ
52gather/axis1PASS0.053 s0.00 MiB0.00 MiB100.000801 ms78.059925 mW62526.000000 pJ
53gather/negative_axisPASS0.053 s0.00 MiB0.00 MiB100.001437 ms78.033403 mW112134.000000 pJ
54gather/negative_indicesPASS0.052 s0.00 MiB0.00 MiB100.000376 ms78.127660 mW29376.000000 pJ
55gemm/alpha_betaPASS0.056 s0.01 MiB0.01 MiB540.007456 ms105.272125 mW784908.960000 pJ
56gemm/bias_rank2_broadcastPASS0.056 s0.00 MiB0.01 MiB540.007072 ms105.979208 mW749484.960000 pJ
57gemm/dynamicPASS0.054 s0.00 MiB0.00 MiB500.002421 ms91.480793 mW221475.000000 pJ
58gemm/dynamic_alphaPASS0.056 s0.00 MiB0.00 MiB500.003262 ms91.415696 mW298198.000000 pJ
59gemm/dynamic_betaPASS0.059 s0.00 MiB0.00 MiB500.004365 ms91.316151 mW398595.000000 pJ
60gemm/dynamic_biasPASS0.056 s0.00 MiB0.00 MiB500.002665 ms91.445779 mW243703.000000 pJ
61gemm/dynamic_bias_alpha_betaPASS0.053 s0.00 MiB0.00 MiB500.005629 ms91.279268 mW513811.000000 pJ
62gemm/dynamic_transBPASS0.054 s0.00 MiB0.00 MiB500.001301 ms91.378171 mW118883.000000 pJ
63gemm/huge_1024PASS0.157 s0.01 MiB0.10 MiB73640.017522 ms215.037402 mW3767885.360000 pJ
64gemm/largePASS0.067 s0.02 MiB0.03 MiB17160.011229 ms140.152181 mW1573768.840000 pJ
65gemm/large_k_small_nPASS0.097 s0.01 MiB0.01 MiB980.004748 ms133.481449 mW633769.920000 pJ
66gemm/non_squarePASS0.060 s0.00 MiB0.01 MiB540.003527 ms118.958310 mW419565.960000 pJ
67gemm/scalar_biasPASS0.054 s0.00 MiB0.01 MiB540.007072 ms105.979208 mW749484.960000 pJ
68gemm/simplePASS0.071 s0.03 MiB0.08 MiB42400.021640 ms151.774196 mW3284393.600000 pJ
69gemm/smallPASS0.052 s0.00 MiB0.00 MiB220.004420 ms90.144000 mW398436.480000 pJ
70gemm/small_k_large_nPASS0.089 s0.01 MiB0.02 MiB1780.007962 ms131.005014 mW1043061.920000 pJ
71gemm/transAPASS0.057 s0.00 MiB0.01 MiB540.005762 ms109.140743 mW628868.960000 pJ
72gemm/transA_transBPASS0.053 s0.00 MiB0.01 MiB540.005762 ms109.140743 mW628868.960000 pJ
73gemm/transBPASS0.062 s0.00 MiB0.01 MiB540.003527 ms118.958310 mW419565.960000 pJ
74gemm/transB_with_biasPASS0.057 s0.01 MiB0.01 MiB540.005046 ms110.546762 mW557818.960000 pJ
75gemm/with_biasPASS0.056 s0.01 MiB0.01 MiB540.005562 ms108.767882 mW604966.960000 pJ
76gemv/constantPASS0.051 s0.00 MiB0.00 MiB000.000000 ms2.000000 mW0.000000 pJ
77gemv/simplePASS0.066 s0.00 MiB0.01 MiB640.005160 ms111.150380 mW573535.960000 pJ
78gemv/with_heterogeneous_constantPASS0.066 s0.00 MiB0.01 MiB640.005549 ms109.816536 mW609371.960000 pJ
79gemv/with_homogeneous_constantPASS0.063 s0.00 MiB0.01 MiB640.005549 ms109.816536 mW609371.960000 pJ
80gemv/with_scalar_constantPASS0.064 s0.00 MiB0.01 MiB640.005549 ms109.816536 mW609371.960000 pJ
81matmul/basicPASS0.064 s0.00 MiB0.00 MiB220.004420 ms90.144000 mW398436.480000 pJ
82matmul/batched_3dPASS0.058 s0.00 MiB0.01 MiB540.005958 ms108.588949 mW646972.960000 pJ
83matmul/batched_3d_dynamicPASS0.053 s0.00 MiB0.00 MiB400.001822 ms92.192645 mW167975.000000 pJ
84matmul/batched_left_constantPASS0.058 s0.00 MiB0.02 MiB980.008822 ms114.385164 mW1009105.920000 pJ
85matmul/batched_lhs_broadcastPASS0.056 s0.00 MiB0.01 MiB540.005681 ms109.389361 mW621440.960000 pJ
86matmul/batched_rhs_broadcastPASS0.059 s0.00 MiB0.01 MiB540.005958 ms108.588949 mW646972.960000 pJ
87matmul/dynamicPASS0.054 s0.00 MiB0.00 MiB500.001621 ms91.421962 mW148195.000000 pJ
88matmul/huge_1024PASS0.142 s0.01 MiB0.10 MiB73640.017522 ms215.037402 mW3767885.360000 pJ
89matmul/left_constantPASS0.064 s0.00 MiB0.01 MiB540.005853 ms108.861944 mW637168.960000 pJ
90matmul/matrix_vectorPASS0.097 s0.52 MiB0.78 MiB1681730.384660 ms202.131271 mW77751814.880000 pJ
91matmul/vector_matrixPASS0.086 s0.01 MiB0.01 MiB980.007409 ms118.680243 mW879301.920000 pJ
92matmul/yolo_attentionPASS0.417 s1.02 MiB43.44 MiB16808.151445 ms170.003707 mW1385775865.000000 pJ
93mul/after_convPASS0.060 s0.00 MiB0.00 MiB430.005453 ms107.639046 mW586955.720000 pJ
94mul/basicPASS0.051 s0.00 MiB0.00 MiB100.000323 ms78.222910 mW25266.000000 pJ
95mul/channel_broadcast_1024PASS0.060 s0.02 MiB0.01 MiB100.006913 ms78.118038 mW540030.000000 pJ
96mul/leading_dimension_broadcastPASS0.058 s0.00 MiB0.00 MiB100.000323 ms78.222910 mW25266.000000 pJ
97mul/scalar_constantPASS0.053 s0.00 MiB0.00 MiB100.000323 ms78.222910 mW25266.000000 pJ
98pool/avg_basicPASS0.057 s0.00 MiB0.00 MiB100.011939 ms78.022112 mW931506.000000 pJ
99pool/avg_ceil_modePASS0.054 s0.00 MiB0.00 MiB100.004359 ms78.033035 mW340146.000000 pJ
100pool/avg_explicit_paddingPASS0.062 s0.00 MiB0.00 MiB100.008822 ms78.027205 mW688356.000000 pJ
101pool/avg_include_padPASS0.055 s0.00 MiB0.00 MiB100.008506 ms78.016929 mW663612.000000 pJ
102pool/avg_large_channelsPASS0.067 s0.04 MiB0.02 MiB100.178249 ms78.280327 mW13953390.000000 pJ
103pool/avg_non_uniform_stridePASS0.060 s0.00 MiB0.00 MiB100.014513 ms78.016537 mW1132254.000000 pJ
104pool/avg_real_asymmetric_paddingPASS0.071 s0.00 MiB0.00 MiB100.025206 ms78.024756 mW1966692.000000 pJ
105pool/max_after_convPASS0.066 s0.00 MiB0.00 MiB640.006452 ms96.374606 mW621808.960000 pJ
106pool/max_basicPASS0.063 s0.00 MiB0.00 MiB300.001634 ms92.132191 mW150544.000000 pJ
107pool/max_ceil_modePASS0.078 s0.00 MiB0.00 MiB200.001297 ms79.111025 mW102607.000000 pJ
108pool/max_global_style_kernel_equals_inputPASS0.067 s0.00 MiB0.00 MiB100.004366 ms78.010994 mW340596.000000 pJ
109pool/max_non_square_kernelPASS0.073 s0.00 MiB0.00 MiB400.003409 ms93.253447 mW317901.000000 pJ
110pool/max_real_asymmetric_paddingPASS0.064 s0.00 MiB0.00 MiB400.003078 ms93.124756 mW286638.000000 pJ
111pool/max_same_upperPASS0.063 s0.00 MiB0.00 MiB300.003024 ms92.095238 mW278496.000000 pJ
112pool/max_stride2_multichannelPASS0.066 s0.00 MiB0.00 MiB300.004012 ms92.269192 mW370184.000000 pJ
113reduce_mean/4d_spatialPASS0.078 s0.00 MiB0.00 MiB300.000321 ms92.448598 mW29676.000000 pJ
114reduce_mean/4d_spatial_keepdims_0PASS0.066 s0.00 MiB0.00 MiB400.000655 ms94.352672 mW61801.000000 pJ
115reduce_mean/after_convPASS0.068 s0.00 MiB0.00 MiB530.005342 ms106.951089 mW571332.720000 pJ
116reduce_mean/all_axes_keepdims_0PASS0.067 s0.00 MiB0.00 MiB200.000391 ms79.237852 mW30982.000000 pJ
117reduce_mean/all_axes_keepdims_1PASS0.056 s0.00 MiB0.00 MiB100.000221 ms78.217195 mW17286.000000 pJ
118reduce_mean/basicPASS0.052 s0.00 MiB0.00 MiB400.000373 ms93.514745 mW34881.000000 pJ
119reduce_mean/channel_axis_nchwPASS0.054 s0.03 MiB0.02 MiB400.164926 ms93.596631 mW15436518.000000 pJ
120reduce_mean/keepdims_0PASS0.054 s0.00 MiB0.00 MiB500.000748 ms91.401070 mW68368.000000 pJ
121reduce_mean/large_dimension_1024PASS0.056 s0.01 MiB0.00 MiB100.002785 ms78.017235 mW217278.000000 pJ
122reduce_mean/legacy_axes_1_2_keepdims_1PASS0.064 s0.00 MiB0.00 MiB200.000271 ms79.354244 mW21505.000000 pJ
123reduce_mean/legacy_axis1_keepdims_0PASS0.071 s0.00 MiB0.00 MiB900.001986 ms92.501511 mW183708.000000 pJ
124reduce_mean/legacy_axis1_keepdims_1PASS0.065 s0.00 MiB0.00 MiB800.001373 ms94.559359 mW129830.000000 pJ
125reduce_mean/legacy_empty_axes_noopPASS0.053 s0.00 MiB0.00 MiB100.000221 ms78.217195 mW17286.000000 pJ
126reduce_mean/legacy_nchw_spatialPASS0.050 s0.00 MiB0.00 MiB300.000321 ms92.448598 mW29676.000000 pJ
127reduce_mean/legacy_negative_axisPASS0.064 s0.00 MiB0.00 MiB600.000553 ms93.520796 mW51717.000000 pJ
128reduce_mean/legacy_reduce_all_keepdims_1PASS0.054 s0.00 MiB0.00 MiB100.000221 ms78.217195 mW17286.000000 pJ
129reduce_mean/negative_axisPASS0.059 s0.00 MiB0.00 MiB600.000553 ms93.520796 mW51717.000000 pJ
130relu/4dPASS0.068 s0.00 MiB0.00 MiB100.000521 ms78.184261 mW40734.000000 pJ
131relu/after_convPASS0.057 s0.00 MiB0.00 MiB330.004956 ms106.998935 mW530286.720000 pJ
132relu/after_gemmPASS0.064 s0.01 MiB0.01 MiB540.007513 ms105.158653 mW790056.960000 pJ
133relu/basicPASS0.056 s0.00 MiB0.00 MiB100.000221 ms78.217195 mW17286.000000 pJ
134reshape/4d_to_2d_flattenPASS0.068 s0.00 MiB0.00 MiB100.000258 ms78.279070 mW20196.000000 pJ
135reshape/infer_dim_minus_onePASS0.059 s0.00 MiB0.00 MiB100.000162 ms78.296296 mW12684.000000 pJ
136reshape/same_rankPASS0.065 s0.00 MiB0.00 MiB100.000162 ms78.296296 mW12684.000000 pJ
137reshape/zero_copies_input_dimPASS0.057 s0.00 MiB0.00 MiB100.000162 ms78.296296 mW12684.000000 pJ
138resize/height_onlyPASS0.063 s0.00 MiB0.00 MiB400.000693 ms93.554113 mW64833.000000 pJ
139resize/nearest_2xPASS0.064 s0.00 MiB0.00 MiB400.001173 ms93.572890 mW109761.000000 pJ
140resize/nearest_downsamplePASS0.062 s0.00 MiB0.00 MiB200.000427 ms79.449649 mW33925.000000 pJ
141resize/non_uniformPASS0.068 s0.00 MiB0.00 MiB600.001753 ms93.575014 mW164037.000000 pJ
142resize/width_onlyPASS0.070 s0.00 MiB0.00 MiB200.000667 ms79.503748 mW53029.000000 pJ
143resize/with_sizesPASS0.052 s0.00 MiB0.00 MiB300.000797 ms92.542033 mW73756.000000 pJ
144sigmoid/4dPASS0.067 s0.00 MiB0.00 MiB100.000521 ms78.184261 mW40734.000000 pJ
145sigmoid/after_gemmPASS0.070 s0.01 MiB0.01 MiB540.007513 ms105.158653 mW790056.960000 pJ
146sigmoid/basicPASS0.062 s0.00 MiB0.00 MiB100.000221 ms78.217195 mW17286.000000 pJ
147slice/2d_basicPASS0.049 s0.00 MiB0.00 MiB100.000242 ms78.297521 mW18948.000000 pJ
148slice/after_convPASS0.062 s0.00 MiB0.01 MiB760.011296 ms118.190765 mW1335082.880000 pJ
149slice/default_axesPASS0.056 s0.00 MiB0.00 MiB100.000242 ms78.297521 mW18948.000000 pJ
150slice/large_channel_1024PASS0.049 s0.01 MiB0.00 MiB100.002832 ms78.144068 mW221304.000000 pJ
151slice/nchw_spatial_cropPASS0.063 s0.00 MiB0.00 MiB100.001302 ms78.239631 mW101868.000000 pJ
152slice/negative_axisPASS0.058 s0.00 MiB0.00 MiB100.000562 ms78.298932 mW44004.000000 pJ
153slice/negative_indicesPASS0.059 s0.00 MiB0.00 MiB100.000322 ms78.298137 mW25212.000000 pJ
154slice/step2PASS0.054 s0.00 MiB0.00 MiB100.002042 ms78.293830 mW159876.000000 pJ
155softmax/3d_last_axisPASS0.059 s0.00 MiB0.00 MiB10UNSUPPORTEDUNSUPPORTEDUNSUPPORTED
156softmax/basicPASS0.069 s0.00 MiB0.00 MiB10UNSUPPORTEDUNSUPPORTEDUNSUPPORTED
157softmax/channel_axisPASS0.061 s0.00 MiB0.00 MiB10UNSUPPORTEDUNSUPPORTEDUNSUPPORTED
158softmax/large_dimension_1024PASS0.045 s0.01 MiB0.01 MiB10UNSUPPORTEDUNSUPPORTEDUNSUPPORTED
159softmax/negative_axisPASS0.064 s0.00 MiB0.00 MiB10UNSUPPORTEDUNSUPPORTEDUNSUPPORTED
160split/basicPASS0.066 s0.00 MiB0.00 MiB100.000403 ms78.297767 mW31554.000000 pJ
161split/equal_three_wayPASS0.065 s0.00 MiB0.00 MiB100.000564 ms78.297872 mW44160.000000 pJ
162split/negative_axisPASS0.058 s0.00 MiB0.00 MiB100.001083 ms78.288089 mW84786.000000 pJ
163split/uneven_channel_axis_4dPASS0.062 s0.00 MiB0.00 MiB100.000242 ms78.297521 mW18948.000000 pJ
164sub/after_gemmPASS0.068 s0.01 MiB0.01 MiB540.007784 ms104.703618 mW815012.960000 pJ
165sub/basicPASS0.060 s0.00 MiB0.00 MiB100.000323 ms78.222910 mW25266.000000 pJ
166sub/broadcast_rowPASS0.062 s0.00 MiB0.00 MiB100.000323 ms78.222910 mW25266.000000 pJ
167sub/channel_broadcast_1024PASS0.057 s0.02 MiB0.01 MiB100.006913 ms78.118038 mW540030.000000 pJ
168sub/constant_lhs_broadcastPASS0.057 s0.00 MiB0.00 MiB100.000322 ms78.223602 mW25188.000000 pJ
169sub/leading_dimension_broadcastPASS0.058 s0.00 MiB0.00 MiB100.000323 ms78.222910 mW25266.000000 pJ