GPU Kernel Information Aggregated by Layer
layer_index | layer_name | layer_type | layer_duration (us) | layer_gpu_duration (us) | layer_cpu_duration (us) | layer_flops | layer_dram_read_bytes | layer_dram_write_bytes | layer_achieved_occupancy (%) | layer_arithmetic_intensity (flops/byte) | layer_arithmetic_throughput (GFlops) | layer_memory_bound |
---|
layer_index | layer_name | layer_type | layer_duration (us) | layer_gpu_duration (us) | layer_cpu_duration (us) | layer_flops | layer_dram_read_bytes | layer_dram_write_bytes | layer_achieved_occupancy (%) | layer_arithmetic_intensity (flops/byte) | layer_arithmetic_throughput (GFlops) | layer_memory_bound |
---|---|---|---|---|---|---|---|---|---|---|---|---|
1 | mobilenet2_batchnorm0_fwd | BatchNorm | 404.67 | 9.83 | 394.83 | 1212416 | 416.00 | 109776.00 | 23.80 | 11.00 | 123.30 | true |
2 | mobilenet2_relu0_fwd | Activation | 48.50 | 5.00 | 43.50 | 401408 | 224.00 | 25408.00 | 56.60 | 15.66 | 80.28 | true |
3 | mobilenet2_conv1_fwd | Convolution | 2975.50 | 256.00 | 2719.50 | 494772224 | 650.67 | 1045.34 | 13.18 | 291726.15 | 1932.70 | false |
4 | mobilenet2_batchnorm1_fwd | BatchNorm | 398.00 | 9.17 | 388.83 | 1212416 | 341.33 | 341.33 | 23.80 | 1776.00 | 132.26 | false |
5 | mobilenet2_relu1_fwd | Activation | 46.67 | 4.67 | 42.00 | 401408 | 42.67 | 0.00 | 52.50 | 9407.93 | 86.01 | false |
6 | mobilenet2_conv2_fwd | Convolution | 3261.17 | 16.83 | 3244.33 | 27295744 | 11648.00 | 142245.33 | 7.66 | 177.37 | 1621.56 | false |
7 | mobilenet2_batchnorm2_fwd | BatchNorm | 132.83 | 9.33 | 123.50 | 2424832 | 682.67 | 999152.00 | 23.80 | 2.43 | 259.81 | true |
8 | mobilenet2_relu2_fwd | Activation | 85.67 | 5.17 | 80.50 | 802816 | 384.00 | 328224.00 | 64.80 | 2.44 | 155.37 | true |
9 | mobilenet2_conv3_fwd | Convolution | 1575.50 | 7.00 | 1568.50 | 1885248 | 1802.67 | 105957.33 | 57.70 | 17.49 | 269.32 | false |
10 | mobilenet2_batchnorm3_fwd | BatchNorm | 37.67 | 5.33 | 32.33 | 618496 | 512.00 | 93594.67 | 22.70 | 6.57 | 115.98 | true |
11 | mobilenet2_relu3_fwd | Activation | 26.33 | 4.17 | 22.17 | 200704 | 0.00 | 1024.00 | 47.20 | 196.00 | 48.17 | false |
12 | mobilenet2_conv4_fwd | Convolution | 1579.33 | 10.00 | 1569.33 | 13045760 | 10208.00 | 199093.33 | 7.30 | 62.33 | 1304.58 | false |
13 | mobilenet2_batchnorm4_fwd | BatchNorm | 47.67 | 5.50 | 42.17 | 1236992 | 1024.00 | 256.00 | 22.80 | 966.40 | 224.91 | false |
14 | mobilenet2_relu4_fwd | Activation | 46.50 | 4.50 | 42.00 | 401408 | 0.00 | 320.00 | 52.10 | 1254.40 | 89.20 | false |
15 | mobilenet2_conv5_fwd | Convolution | 3019.50 | 10.00 | 3009.50 | 4214784 | 4352.00 | 384.00 | 12.20 | 889.95 | 421.48 | false |
16 | mobilenet2_batchnorm5_fwd | BatchNorm | 93.33 | 5.33 | 88.00 | 1236992 | 1024.00 | 64.00 | 22.70 | 1136.94 | 231.95 | false |
17 | mobilenet2_relu5_fwd | Activation | 47.00 | 5.00 | 42.00 | 401408 | 0.00 | 0.00 | 50.50 | 0.00 | 80.28 | true |
18 | mobilenet2_conv6_fwd | Convolution | 1858.50 | 13.00 | 1845.50 | 25890816 | 16384.00 | 373.33 | 7.00 | 1545.04 | 1991.60 | false |
19 | mobilenet2_batchnorm6_fwd | BatchNorm | 86.83 | 5.50 | 81.33 | 1236992 | 1024.00 | 128.00 | 22.80 | 1073.78 | 224.91 | false |
20 | mobilenet2_relu6_fwd | Activation | 51.50 | 4.67 | 46.83 | 401408 | 0.00 | 0.00 | 52.10 | 0.00 | 86.01 | true |
21 | mobilenet2_conv7_fwd | Convolution | 824.33 | 5.33 | 819.00 | 931968 | 2304.00 | 0.00 | 46.90 | 404.50 | 174.75 | false |
22 | mobilenet2_batchnorm7_fwd | BatchNorm | 19.00 | 4.67 | 14.33 | 317440 | 1024.00 | 0.00 | 11.40 | 310.00 | 68.02 | false |
23 | mobilenet2_relu7_fwd | Activation | 16.17 | 4.00 | 12.17 | 100352 | 0.00 | 320.00 | 34.70 | 313.60 | 25.09 | false |
24 | mobilenet2_conv8_fwd | Convolution | 892.17 | 14.00 | 878.17 | 13207552 | 41002.67 | 288.00 | 4.40 | 319.87 | 943.40 | false |
25 | mobilenet2_batchnorm8_fwd | BatchNorm | 28.83 | 4.50 | 24.33 | 634880 | 2048.00 | 0.00 | 17.80 | 310.00 | 141.08 | false |
26 | mobilenet2_relu8_fwd | Activation | 25.83 | 4.17 | 21.67 | 200704 | 0.00 | 0.00 | 47.90 | 0.00 | 48.17 | true |
27 | mobilenet2_conv9_fwd | Convolution | 1568.83 | 6.50 | 1562.33 | 2107392 | 4608.00 | 128.00 | 18.60 | 444.97 | 324.21 | false |
28 | mobilenet2_batchnorm9_fwd | BatchNorm | 39.50 | 4.50 | 35.00 | 634880 | 2048.00 | 0.00 | 17.70 | 310.00 | 141.08 | false |
29 | mobilenet2_relu9_fwd | Activation | 27.17 | 4.00 | 23.17 | 200704 | 0.00 | 0.00 | 47.30 | 0.00 | 50.18 | true |
30 | mobilenet2_conv10_fwd | Convolution | 1748.33 | 17.00 | 1731.33 | 26314752 | 65536.00 | 288.00 | 3.90 | 399.77 | 1547.93 | false |
31 | mobilenet2_batchnorm10_fwd | BatchNorm | 69.17 | 5.00 | 64.17 | 634880 | 2048.00 | 128.00 | 17.80 | 291.76 | 126.98 | false |
32 | mobilenet2_relu10_fwd | Activation | 26.00 | 4.17 | 21.83 | 200704 | 0.00 | 0.00 | 47.90 | 0.00 | 48.17 | true |
33 | mobilenet2_conv11_fwd | Convolution | 452.50 | 5.50 | 447.00 | 455424 | 4608.00 | 0.00 | 46.20 | 98.83 | 82.80 | false |
34 | mobilenet2_batchnorm11_fwd | BatchNorm | 14.50 | 4.67 | 9.83 | 158720 | 2048.00 | 0.00 | 4.70 | 77.50 | 34.01 | false |
35 | mobilenet2_relu11_fwd | Activation | 11.00 | 4.00 | 7.00 | 50176 | 0.00 | 128.00 | 27.50 | 392.00 | 12.54 | false |
36 | mobilenet2_conv12_fwd | Convolution | 780.17 | 13.17 | 767.00 | 14730240 | 131072.00 | 288.00 | 3.10 | 112.14 | 1118.72 | false |
37 | mobilenet2_batchnorm12_fwd | BatchNorm | 20.00 | 5.00 | 15.00 | 317440 | 4096.00 | 0.00 | 8.80 | 77.50 | 63.49 | false |
38 | mobilenet2_relu12_fwd | Activation | 16.00 | 4.00 | 12.00 | 100352 | 0.00 | 0.00 | 35.20 | 0.00 | 25.09 | true |
39 | mobilenet2_conv13_fwd | Convolution | 868.33 | 5.00 | 863.33 | 869376 | 9216.00 | 128.00 | 47.20 | 93.04 | 173.88 | false |
40 | mobilenet2_batchnorm13_fwd | BatchNorm | 18.83 | 5.00 | 13.83 | 317440 | 4096.00 | 0.00 | 8.70 | 77.50 | 63.49 | false |
41 | mobilenet2_relu13_fwd | Activation | 16.33 | 4.00 | 12.33 | 100352 | 0.00 | 0.00 | 34.50 | 0.00 | 25.09 | true |
42 | mobilenet2_conv14_fwd | Convolution | 1510.00 | 21.00 | 1489.00 | 29410304 | 262144.00 | 288.00 | 3.10 | 112.07 | 1400.49 | false |
43 | mobilenet2_batchnorm14_fwd | BatchNorm | 23.17 | 5.00 | 18.17 | 317440 | 4096.00 | 128.00 | 8.80 | 75.15 | 63.49 | false |
44 | mobilenet2_relu14_fwd | Activation | 16.33 | 4.00 | 12.33 | 100352 | 0.00 | 0.00 | 35.20 | 0.00 | 25.09 | true |
45 | mobilenet2_conv15_fwd | Convolution | 861.00 | 5.00 | 856.00 | 869376 | 9216.00 | 0.00 | 47.20 | 94.33 | 173.88 | false |
46 | mobilenet2_batchnorm15_fwd | BatchNorm | 19.83 | 5.00 | 14.83 | 317440 | 4096.00 | 0.00 | 8.70 | 77.50 | 63.49 | false |
47 | mobilenet2_relu15_fwd | Activation | 16.00 | 4.00 | 12.00 | 100352 | 0.00 | 128.00 | 34.70 | 784.00 | 25.09 | false |
48 | mobilenet2_conv16_fwd | Convolution | 1504.33 | 21.00 | 1483.33 | 29410304 | 262144.00 | 288.00 | 3.10 | 112.07 | 1400.49 | false |
49 | mobilenet2_batchnorm16_fwd | BatchNorm | 22.33 | 4.50 | 17.83 | 317440 | 4096.00 | 0.00 | 8.80 | 77.50 | 70.54 | false |
50 | mobilenet2_relu16_fwd | Activation | 16.00 | 4.00 | 12.00 | 100352 | 0.00 | 0.00 | 35.20 | 0.00 | 25.09 | true |
51 | mobilenet2_conv17_fwd | Convolution | 862.67 | 5.17 | 857.50 | 869376 | 9216.00 | 128.00 | 47.20 | 93.04 | 168.26 | false |
52 | mobilenet2_batchnorm17_fwd | BatchNorm | 18.83 | 4.67 | 14.17 | 317440 | 4778.67 | 0.00 | 8.70 | 66.43 | 68.02 | false |
53 | mobilenet2_relu17_fwd | Activation | 15.83 | 4.00 | 11.83 | 100352 | 0.00 | 0.00 | 34.60 | 0.00 | 25.09 | true |
54 | mobilenet2_conv18_fwd | Convolution | 1508.50 | 20.83 | 1487.67 | 29410304 | 262144.00 | 288.00 | 3.10 | 112.07 | 1411.72 | false |
55 | mobilenet2_batchnorm18_fwd | BatchNorm | 27.00 | 5.00 | 22.00 | 317440 | 4096.00 | 128.00 | 8.80 | 75.15 | 63.49 | false |
56 | mobilenet2_relu18_fwd | Activation | 15.83 | 4.00 | 11.83 | 100352 | 0.00 | 0.00 | 35.20 | 0.00 | 25.09 | true |
57 | mobilenet2_conv19_fwd | Convolution | 858.50 | 5.17 | 853.33 | 869376 | 9216.00 | 0.00 | 47.20 | 94.33 | 168.26 | false |
58 | mobilenet2_batchnorm19_fwd | BatchNorm | 18.17 | 5.00 | 13.17 | 317440 | 4096.00 | 0.00 | 8.70 | 77.50 | 63.49 | false |
59 | mobilenet2_relu19_fwd | Activation | 16.00 | 4.00 | 12.00 | 100352 | 0.00 | 128.00 | 34.60 | 784.00 | 25.09 | false |
60 | mobilenet2_conv20_fwd | Convolution | 1513.17 | 21.00 | 1492.17 | 29410304 | 262144.00 | 314.67 | 3.10 | 112.06 | 1400.49 | false |
61 | mobilenet2_batchnorm20_fwd | BatchNorm | 26.17 | 5.00 | 21.17 | 317440 | 4096.00 | 0.00 | 8.80 | 77.50 | 63.49 | false |
62 | mobilenet2_relu20_fwd | Activation | 16.00 | 4.00 | 12.00 | 100352 | 0.00 | 0.00 | 35.20 | 0.00 | 25.09 | true |
63 | mobilenet2_conv21_fwd | Convolution | 871.17 | 5.00 | 866.17 | 869376 | 9216.00 | 128.00 | 47.20 | 93.04 | 173.88 | false |
64 | mobilenet2_batchnorm21_fwd | BatchNorm | 19.50 | 5.00 | 14.50 | 317440 | 4096.00 | 0.00 | 8.60 | 77.50 | 63.49 | false |
65 | mobilenet2_relu21_fwd | Activation | 15.83 | 4.00 | 11.83 | 100352 | 0.00 | 0.00 | 34.60 | 0.00 | 25.09 | true |
66 | mobilenet2_conv22_fwd | Convolution | 1498.83 | 21.00 | 1477.83 | 29410304 | 262261.33 | 1306.67 | 3.10 | 111.59 | 1400.49 | false |
67 | mobilenet2_batchnorm22_fwd | BatchNorm | 24.33 | 4.83 | 19.50 | 317440 | 4128.00 | 186.67 | 8.70 | 73.57 | 65.68 | false |
68 | mobilenet2_relu22_fwd | Activation | 16.00 | 4.00 | 12.00 | 100352 | 0.00 | 0.00 | 35.10 | 0.00 | 25.09 | true |
69 | mobilenet2_conv23_fwd | Convolution | 292.00 | 5.50 | 286.50 | 217344 | 9216.00 | 277.33 | 45.50 | 22.89 | 39.52 | false |
70 | mobilenet2_batchnorm23_fwd | BatchNorm | 14.17 | 4.17 | 10.00 | 87808 | 4096.00 | 91936.00 | 8.50 | 0.91 | 21.07 | true |
71 | mobilenet2_relu23_fwd | Activation | 8.00 | 4.00 | 4.00 | 25088 | 0.00 | 7136.00 | 25.00 | 3.52 | 6.27 | true |
72 | mobilenet2_conv24_fwd | Convolution | 924.17 | 21.17 | 903.00 | 16802304 | 524421.33 | 1109.33 | 3.10 | 31.97 | 793.80 | false |
73 | mobilenet2_batchnorm24_fwd | BatchNorm | 17.33 | 4.83 | 12.50 | 175616 | 8192.00 | 0.00 | 14.40 | 21.44 | 36.34 | false |
74 | mobilenet2_relu24_fwd | Activation | 11.00 | 4.00 | 7.00 | 50176 | 0.00 | 0.00 | 28.60 | 0.00 | 12.54 | true |
75 | mobilenet2_conv25_fwd | Convolution | 558.00 | 5.00 | 553.00 | 394752 | 18432.00 | 128.00 | 46.70 | 21.27 | 78.95 | false |
76 | mobilenet2_batchnorm25_fwd | BatchNorm | 16.33 | 4.33 | 12.00 | 175616 | 8192.00 | 0.00 | 14.00 | 21.44 | 40.53 | false |
77 | mobilenet2_relu25_fwd | Activation | 10.33 | 4.00 | 6.33 | 50176 | 0.00 | 0.00 | 27.50 | 0.00 | 12.54 | true |
78 | mobilenet2_conv26_fwd | Convolution | 1813.33 | 36.00 | 1777.33 | 33579520 | 1048618.67 | 254970.67 | 3.10 | 25.76 | 932.76 | false |
79 | mobilenet2_batchnorm26_fwd | BatchNorm | 71.00 | 5.00 | 66.00 | 175616 | 8192.00 | 6186.67 | 14.40 | 12.21 | 35.12 | true |
80 | mobilenet2_relu26_fwd | Activation | 11.00 | 4.00 | 7.00 | 50176 | 0.00 | 0.00 | 28.70 | 0.00 | 12.54 | true |
81 | mobilenet2_pool0_fwd | Pooling | 42.67 | 8.00 | 34.67 | 32287 | 4608.00 | 4565.33 | 10.70 | 3.52 | 4.04 | true |
82 | mobilenet2_flatten0_flatten0 | Flatten | 3.50 | 0.00 | 3.50 | 0 | 0.00 | 0.00 | 0.00 | 0.00 | 0.00 | true |
83 | mobilenet2_dense0_fwd | FullyConnected | 507.00 | 12.67 | 494.33 | 1120000 | 2057109.33 | 927296.00 | 10.49 | 0.38 | 88.42 | true |
Showing 1 to 83 of 83 entries