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 | Pad | Pad | 88.67 | 9.00 | 79.67 | 0 | 13312.00 | 354346.67 | 46.20 | 0.00 | 0.00 | true |
2 | convolution-0-TransposeNHWCToNCHW-LayoutOptimizer | Transpose | 51.33 | 6.00 | 45.33 | 0 | 6944.00 | 564469.33 | 61.80 | 0.00 | 0.00 | true |
3 | convolution | Conv2D | 222.00 | 46.67 | 175.33 | 244858880 | 239178.67 | 2803744.00 | 23.34 | 80.47 | 5246.94 | false |
4 | add | Add | 42.33 | 6.00 | 36.33 | 802816 | 2901.33 | 1398581.33 | 83.30 | 0.57 | 133.80 | true |
5 | conv1/relu_7x7 | Relu | 32.00 | 5.00 | 27.00 | 0 | 1024.00 | 965536.00 | 75.40 | 0.00 | 0.00 | true |
6 | conv1/relu_7x7-0-0-TransposeNCHWToNHWC-LayoutOptimizer | Transpose | 36.00 | 8.00 | 28.00 | 0 | 9130.67 | 2977706.67 | 68.60 | 0.00 | 0.00 | true |
7 | PadV2 | PadV2 | 38.33 | 9.00 | 29.33 | 0 | 323306.67 | 3266218.67 | 76.90 | 0.00 | 0.00 | true |
8 | pool1/3x3_s2-0-TransposeNHWCToNCHW-LayoutOptimizer | Transpose | 28.33 | 8.00 | 20.33 | 0 | 32629.33 | 3552746.67 | 65.80 | 0.00 | 0.00 | true |
9 | pool1/3x3_s2 | MaxPool | 55.33 | 8.33 | 47.00 | 200704 | 14976.00 | 870346.67 | 57.50 | 0.23 | 24.09 | true |
10 | pool1/3x3_s2-0-0-TransposeNCHWToNHWC-LayoutOptimizer | Transpose | 29.00 | 4.00 | 25.00 | 0 | 192.00 | 409280.00 | 26.00 | 0.00 | 0.00 | true |
11 | pool1/norm1 | LRN | 67.33 | 37.00 | 30.33 | 2145024 | 6144.00 | 388010.67 | 12.20 | 5.44 | 57.97 | true |
12 | convolution_1-0-TransposeNHWCToNCHW-LayoutOptimizer | Transpose | 22.00 | 4.00 | 18.00 | 0 | 0.00 | 320.00 | 26.80 | 0.00 | 0.00 | true |
13 | convolution_1 | Conv2D | 105.67 | 15.00 | 90.67 | 25890816 | 16832.00 | 17621.33 | 16.04 | 751.47 | 1726.05 | false |
14 | add_1 | Add | 26.00 | 4.00 | 22.00 | 200704 | 512.00 | 448.00 | 60.00 | 209.07 | 50.18 | false |
15 | conv2/relu_3x3_reduce | Relu | 19.67 | 3.00 | 16.67 | 0 | 0.00 | 320.00 | 67.20 | 0.00 | 0.00 | true |
16 | Pad_1 | Pad | 35.33 | 7.00 | 28.33 | 0 | 10240.00 | 28234.67 | 58.30 | 0.00 | 0.00 | true |
17 | convolution_2 | Conv2D | 177.00 | 62.00 | 115.00 | 374022144 | 462592.00 | 3376522.67 | 22.88 | 97.42 | 6032.62 | false |
18 | add_2 | Add | 26.67 | 5.00 | 21.67 | 602112 | 1024.00 | 92426.67 | 73.30 | 6.44 | 120.42 | true |
19 | conv2/relu_3x3 | Relu | 19.00 | 4.00 | 15.00 | 0 | 2218.67 | 11296.00 | 65.00 | 0.00 | 0.00 | true |
20 | conv2/relu_3x3-0-0-TransposeNCHWToNHWC-LayoutOptimizer | Transpose | 26.00 | 6.00 | 20.00 | 0 | 576.00 | 2007338.67 | 55.00 | 0.00 | 0.00 | true |
21 | conv2/norm2 | LRN | 131.00 | 106.00 | 25.00 | 6409984 | 7562.67 | 2753674.67 | 12.20 | 2.32 | 60.47 | true |
22 | PadV2_1 | PadV2 | 30.67 | 6.00 | 24.67 | 0 | 2368.00 | 1784586.67 | 76.60 | 0.00 | 0.00 | true |
23 | pool2/3x3_s2-0-TransposeNHWCToNCHW-LayoutOptimizer | Transpose | 26.67 | 7.00 | 19.67 | 0 | 0.00 | 2657482.67 | 57.60 | 0.00 | 0.00 | true |
24 | pool2/3x3_s2 | MaxPool | 37.67 | 6.00 | 31.67 | 150528 | 2560.00 | 618709.33 | 53.00 | 0.24 | 25.09 | true |
25 | pool2/3x3_s2-0-0-TransposeNCHWToNHWC-LayoutOptimizer | Transpose | 23.33 | 3.00 | 20.33 | 0 | 0.00 | 49536.00 | 0.00 | 0.00 | 0.00 | true |
26 | convolution_5 | Conv2D | 100.00 | 19.00 | 81.00 | 9842944 | 15296.00 | 137322.67 | 11.75 | 64.49 | 518.05 | false |
27 | convolution_4 | Conv2D | 94.00 | 19.00 | 75.00 | 29566464 | 74240.00 | 119562.67 | 11.31 | 152.56 | 1556.13 | false |
28 | convolution_3 | Conv2D | 98.67 | 19.00 | 79.67 | 19710976 | 49152.00 | 97301.33 | 11.21 | 134.59 | 1037.42 | false |
29 | PadV2_2 | PadV2 | 31.00 | 4.00 | 27.00 | 0 | 0.00 | 91264.00 | 60.10 | 0.00 | 0.00 | true |
30 | add_5 | Add | 22.67 | 3.00 | 19.67 | 12544 | 64.00 | 32608.00 | 42.40 | 0.38 | 4.18 | true |
31 | add_4 | Add | 22.33 | 3.00 | 19.33 | 75264 | 384.00 | 111370.67 | 44.70 | 0.67 | 25.09 | true |
32 | add_3 | Add | 18.67 | 3.00 | 15.67 | 50176 | 256.00 | 18869.33 | 44.30 | 2.62 | 16.73 | true |
33 | inception_3a/pool-0-TransposeNHWCToNCHW-LayoutOptimizer | Transpose | 23.33 | 4.00 | 19.33 | 0 | 0.00 | 390677.33 | 23.70 | 0.00 | 0.00 | true |
34 | inception_3a/relu_5x5_reduce | Relu | 19.00 | 3.00 | 16.00 | 0 | 170.67 | 6026.67 | 43.30 | 0.00 | 0.00 | true |
35 | inception_3a/relu_3x3_reduce | Relu | 17.33 | 3.00 | 14.33 | 0 | 0.00 | 29290.67 | 43.40 | 0.00 | 0.00 | true |
36 | inception_3a/pool | MaxPool | 38.33 | 6.00 | 32.33 | 150528 | 0.00 | 86506.67 | 52.10 | 1.74 | 25.09 | true |
37 | Pad_3 | Pad | 30.00 | 5.00 | 25.00 | 0 | 1280.00 | 0.00 | 45.80 | 0.00 | 0.00 | true |
38 | Pad_2 | Pad | 25.67 | 6.00 | 19.67 | 0 | 0.00 | 7189.33 | 40.90 | 0.00 | 0.00 | true |
39 | convolution_6 | Conv2D | 98.00 | 19.00 | 79.00 | 9855488 | 24576.00 | 8832.00 | 11.16 | 295.00 | 518.71 | false |
40 | convolution_8 | Conv2D | 107.33 | 29.00 | 78.33 | 20505088 | 51200.00 | 352.00 | 8.55 | 397.76 | 707.07 | false |
41 | convolution_7 | Conv2D | 126.33 | 37.00 | 89.33 | 106684416 | 445696.00 | 295605.33 | 17.25 | 143.92 | 2883.36 | false |
42 | add_6 | Add | 24.00 | 4.00 | 20.00 | 25088 | 640.00 | 0.00 | 44.00 | 39.20 | 6.27 | false |
43 | add_8 | Add | 19.00 | 3.00 | 16.00 | 25088 | 128.00 | 85.33 | 43.50 | 117.60 | 8.36 | false |
44 | add_7 | Add | 19.00 | 3.00 | 16.00 | 100352 | 512.00 | 0.00 | 50.00 | 196.00 | 33.45 | false |
45 | inception_3a/output | ConcatV2 | 75.00 | 0.00 | 75.00 | 0 | 0.00 | 0.00 | 0.00 | 0.00 | NaN | true |
46 | inception_3a/relu_1x1 | Relu | 21.33 | 3.00 | 18.33 | 0 | 597.33 | 42.67 | 67.60 | 0.00 | 0.00 | true |
47 | inception_3a/relu_1x1-3-0-TransposeNCHWToNHWC-LayoutOptimizer | Transpose | 25.67 | 4.00 | 21.67 | 0 | 0.00 | 8682.67 | 26.50 | 0.00 | 0.00 | true |
48 | convolution_11 | Conv2D | 101.00 | 22.00 | 79.00 | 13132288 | 33024.00 | 14346.67 | 10.06 | 277.22 | 596.92 | false |
49 | convolution_10 | Conv2D | 109.67 | 30.00 | 79.67 | 52529152 | 131072.00 | 18784.00 | 8.91 | 350.53 | 1750.97 | false |
50 | convolution_9 | Conv2D | 103.00 | 30.00 | 73.00 | 52529152 | 131072.00 | 170250.67 | 8.85 | 174.33 | 1750.97 | false |
51 | PadV2_3 | PadV2 | 32.67 | 4.33 | 28.33 | 0 | 512.00 | 469941.33 | 57.70 | 0.00 | 0.00 | true |
52 | add_11 | Add | 25.33 | 3.00 | 22.33 | 25088 | 128.00 | 100821.33 | 44.00 | 0.25 | 8.36 | true |
53 | add_10 | Add | 20.00 | 3.00 | 17.00 | 100352 | 512.00 | 340074.67 | 50.00 | 0.29 | 33.45 | true |
54 | add_9 | Add | 20.00 | 3.33 | 16.67 | 100352 | 512.00 | 294346.67 | 50.20 | 0.34 | 30.11 | true |
55 | inception_3b/pool-0-TransposeNHWCToNCHW-LayoutOptimizer | Transpose | 25.00 | 4.00 | 21.00 | 0 | 256.00 | 311893.33 | 29.70 | 0.00 | 0.00 | true |
56 | inception_3b/relu_5x5_reduce | Relu | 19.00 | 3.00 | 16.00 | 0 | 0.00 | 7509.33 | 43.70 | 0.00 | 0.00 | true |
57 | inception_3b/relu_3x3_reduce | Relu | 18.33 | 3.00 | 15.33 | 0 | 0.00 | 10549.33 | 52.60 | 0.00 | 0.00 | true |
58 | inception_3b/pool | MaxPool | 38.67 | 7.00 | 31.67 | 200704 | 0.00 | 876362.67 | 53.70 | 0.23 | 28.67 | true |
59 | Pad_5 | Pad | 31.33 | 5.00 | 26.33 | 0 | 85.33 | 256.00 | 45.70 | 0.00 | 0.00 | true |
60 | Pad_4 | Pad | 29.67 | 6.00 | 23.67 | 0 | 0.00 | 7466.67 | 46.80 | 0.00 | 0.00 | true |
61 | convolution_12 | Conv2D | 101.00 | 22.00 | 79.00 | 26264576 | 65536.00 | 6378.67 | 10.03 | 365.22 | 1193.84 | false |
62 | convolution_14 | Conv2D | 127.33 | 51.33 | 76.00 | 122955264 | 307200.00 | 661.33 | 6.64 | 399.39 | 2395.25 | false |
63 | convolution_13 | Conv2D | 133.33 | 46.00 | 87.33 | 212680704 | 922016.00 | 2799050.67 | 18.17 | 57.16 | 4623.49 | false |
64 | add_12 | Add | 28.33 | 3.00 | 25.33 | 50176 | 512.00 | 2090.67 | 44.90 | 19.28 | 16.73 | true |
65 | add_14 | Add | 19.67 | 3.00 | 16.67 | 75264 | 384.00 | 109909.33 | 44.70 | 0.68 | 25.09 | true |
66 | add_13 | Add | 19.00 | 4.00 | 15.00 | 150528 | 768.00 | 381098.67 | 58.90 | 0.39 | 37.63 | true |
67 | inception_3b/output | ConcatV2 | 66.67 | 0.00 | 66.67 | 0 | 0.00 | 0.00 | 0.00 | 0.00 | NaN | true |
68 | inception_3b/relu_1x1 | Relu | 22.00 | 4.00 | 18.00 | 0 | 768.00 | 398101.33 | 61.60 | 0.00 | 0.00 | true |
69 | inception_3b/relu_1x1-0-0-TransposeNCHWToNHWC-LayoutOptimizer | Transpose | 24.67 | 5.00 | 19.67 | 0 | 0.00 | 921482.67 | 40.90 | 0.00 | 0.00 | true |
70 | PadV2_4 | PadV2 | 29.33 | 5.00 | 24.33 | 0 | 768.00 | 572213.33 | 61.80 | 0.00 | 0.00 | true |
71 | pool3/3x3_s2-0-TransposeNHWCToNCHW-LayoutOptimizer | Transpose | 24.67 | 6.00 | 18.67 | 0 | 256.00 | 696032.00 | 43.00 | 0.00 | 0.00 | true |
72 | pool3/3x3_s2 | MaxPool | 37.67 | 5.00 | 32.67 | 94080 | 1792.00 | 290720.00 | 41.00 | 0.32 | 18.82 | true |
73 | pool3/3x3_s2-2-0-TransposeNCHWToNHWC-LayoutOptimizer | Transpose | 25.00 | 3.00 | 22.00 | 0 | 0.00 | 0.00 | 0.00 | 0.00 | 0.00 | true |
74 | convolution_17 | Conv2D | 112.00 | 32.00 | 80.00 | 6884416 | 33109.33 | 9120.00 | 6.75 | 163.02 | 215.14 | false |
75 | convolution_16 | Conv2D | 111.67 | 33.00 | 78.67 | 20662656 | 184320.00 | 9514.67 | 8.06 | 106.60 | 626.14 | false |
76 | convolution_15 | Conv2D | 107.67 | 35.00 | 72.67 | 41325312 | 368640.00 | 9610.67 | 9.13 | 109.25 | 1180.72 | false |
77 | PadV2_5 | PadV2 | 30.33 | 4.00 | 26.33 | 0 | 0.00 | 2240.00 | 53.70 | 0.00 | 0.00 | true |
78 | add_17 | Add | 25.67 | 3.00 | 22.67 | 3136 | 64.00 | 0.00 | 42.50 | 49.00 | 1.05 | false |
79 | add_16 | Add | 19.33 | 3.00 | 16.33 | 18816 | 384.00 | 42.67 | 44.40 | 44.10 | 6.27 | false |
80 | add_15 | Add | 19.00 | 3.00 | 16.00 | 37632 | 768.00 | 426.67 | 44.30 | 31.50 | 12.54 | false |
81 | inception_4a/pool-0-TransposeNHWCToNCHW-LayoutOptimizer | Transpose | 24.00 | 3.33 | 20.67 | 0 | 0.00 | 2645.33 | 17.80 | 0.00 | 0.00 | true |
82 | inception_4a/relu_5x5_reduce | Relu | 18.00 | 3.00 | 15.00 | 0 | 0.00 | 298.67 | 42.90 | 0.00 | 0.00 | true |
83 | inception_4a/relu_3x3_reduce | Relu | 17.00 | 3.00 | 14.00 | 0 | 0.00 | 725.33 | 43.40 | 0.00 | 0.00 | true |
84 | inception_4a/pool | MaxPool | 36.00 | 5.00 | 31.00 | 94080 | 0.00 | 4778.67 | 39.60 | 19.69 | 18.82 | true |
85 | Pad_7 | Pad | 29.00 | 5.00 | 24.00 | 0 | 512.00 | 0.00 | 38.90 | 0.00 | 0.00 | true |
86 | Pad_6 | Pad | 27.33 | 5.00 | 22.33 | 0 | 0.00 | 0.00 | 46.70 | 0.00 | 0.00 | true |
87 | convolution_18 | Conv2D | 110.00 | 33.00 | 77.00 | 13775104 | 122880.00 | 394.67 | 7.95 | 111.74 | 417.43 | false |
88 | convolution_20 | Conv2D | 201.33 | 59.00 | 142.33 | 20395552 | 140064.00 | 2462528.00 | 29.71 | 7.84 | 345.69 | true |
89 | convolution_19 | Conv2D | 133.67 | 42.00 | 91.67 | 47609856 | 739157.33 | 1352672.00 | 18.03 | 22.76 | 1133.57 | true |
90 | add_18 | Add | 25.00 | 4.00 | 21.00 | 12544 | 3157.33 | 170.67 | 44.50 | 3.77 | 3.14 | true |
91 | add_20 | Add | 19.00 | 3.00 | 16.00 | 9408 | 448.00 | 42.67 | 42.00 | 19.17 | 3.14 | true |
92 | add_19 | Add | 17.67 | 3.00 | 14.67 | 40768 | 832.00 | 298.67 | 45.40 | 36.06 | 13.59 | false |
93 | inception_4a/output | ConcatV2 | 68.67 | 0.00 | 68.67 | 0 | 0.00 | 0.00 | 0.00 | 0.00 | NaN | true |
94 | inception_4a/relu_1x1 | Relu | 22.67 | 3.00 | 19.67 | 0 | 768.00 | 27776.00 | 53.80 | 0.00 | 0.00 | true |
96 | convolution_23 | Conv2D | 121.00 | 36.00 | 85.00 | 7344736 | 57472.00 | 169504.00 | 6.56 | 32.36 | 204.02 | false |
97 | convolution_22 | Conv2D | 109.67 | 35.00 | 74.67 | 29382080 | 229376.00 | 6026.67 | 7.82 | 124.82 | 839.49 | false |
98 | convolution_21 | Conv2D | 107.67 | 35.00 | 72.67 | 36731520 | 327680.00 | 23370.67 | 7.97 | 104.63 | 1049.47 | false |
99 | PadV2_6 | PadV2 | 35.67 | 5.00 | 30.67 | 0 | 4096.00 | 14954.67 | 58.00 | 0.00 | 0.00 | true |
100 | add_23 | Add | 22.33 | 3.00 | 19.33 | 4704 | 128.00 | 0.00 | 38.30 | 36.75 | 1.57 | false |
101 | add_22 | Add | 18.33 | 3.00 | 15.33 | 21952 | 448.00 | 213.33 | 44.10 | 33.19 | 7.32 | false |
102 | add_21 | Add | 18.67 | 3.00 | 15.67 | 31360 | 640.00 | 3285.33 | 44.70 | 7.99 | 10.45 | true |
103 | inception_4b/pool-0-TransposeNHWCToNCHW-LayoutOptimizer | Transpose | 23.00 | 4.00 | 19.00 | 0 | 0.00 | 30421.33 | 18.80 | 0.00 | 0.00 | true |
104 | inception_4b/relu_5x5_reduce | Relu | 18.33 | 3.00 | 15.33 | 0 | 0.00 | 128.00 | 41.60 | 0.00 | 0.00 | true |
105 | inception_4b/relu_3x3_reduce | Relu | 17.00 | 3.00 | 14.00 | 0 | 0.00 | 16906.67 | 43.30 | 0.00 | 0.00 | true |
106 | inception_4b/pool | MaxPool | 36.00 | 6.00 | 30.00 | 100352 | 6912.00 | 50602.67 | 45.50 | 1.74 | 16.73 | true |
107 | Pad_9 | Pad | 33.67 | 5.33 | 28.33 | 0 | 5888.00 | 0.00 | 46.60 | 0.00 | 0.00 | true |
108 | Pad_8 | Pad | 25.33 | 5.00 | 20.33 | 0 | 0.00 | 0.00 | 46.80 | 0.00 | 0.00 | true |
109 | convolution_24 | Conv2D | 113.67 | 35.00 | 78.67 | 14692608 | 131072.00 | 586.67 | 7.69 | 111.60 | 419.79 | false |
110 | convolution_26 | Conv2D | 173.00 | 59.33 | 113.67 | 39376384 | 3106165.33 | 6885077.33 | 31.81 | 3.94 | 663.65 | true |
111 | convolution_25 | Conv2D | 144.67 | 47.67 | 97.00 | 55444480 | 1039136.00 | 1715605.33 | 18.18 | 20.13 | 1163.16 | true |
112 | add_24 | Add | 26.33 | 4.00 | 22.33 | 12544 | 54016.00 | 1194.67 | 44.10 | 0.23 | 3.14 | true |
113 | add_26 | Add | 18.67 | 3.33 | 15.33 | 12544 | 256.00 | 512.00 | 42.00 | 16.33 | 3.76 | true |
114 | add_25 | Add | 18.67 | 3.00 | 15.67 | 43904 | 896.00 | 23093.33 | 45.00 | 1.83 | 14.63 | true |
115 | inception_4b/output | ConcatV2 | 69.00 | 0.00 | 69.00 | 0 | 0.00 | 0.00 | 0.00 | 0.00 | NaN | true |
116 | inception_4b/relu_1x1 | Relu | 21.00 | 3.00 | 18.00 | 0 | 1536.00 | 211978.67 | 53.50 | 0.00 | 0.00 | true |
117 | inception_4b/relu_1x1-3-0-TransposeNCHWToNHWC-LayoutOptimizer | Transpose | 23.33 | 4.67 | 18.67 | 0 | 6656.00 | 44373.33 | 16.80 | 0.00 | 0.00 | true |
118 | convolution_29 | Conv2D | 118.67 | 37.00 | 81.67 | 7344736 | 58453.33 | 169536.00 | 7.59 | 32.22 | 198.51 | false |
119 | convolution_28 | Conv2D | 109.33 | 35.00 | 74.33 | 29385216 | 262144.00 | 261621.33 | 7.88 | 56.10 | 839.58 | false |
120 | convolution_27 | Conv2D | 110.67 | 35.00 | 75.67 | 29385216 | 262741.33 | 73845.33 | 7.82 | 87.30 | 839.58 | false |
121 | PadV2_7 | PadV2 | 31.33 | 5.00 | 26.33 | 0 | 4608.00 | 81909.33 | 57.10 | 0.00 | 0.00 | true |
122 | add_29 | Add | 22.67 | 3.00 | 19.67 | 4704 | 128.00 | 9856.00 | 38.40 | 0.47 | 1.57 | true |
123 | add_28 | Add | 18.67 | 3.00 | 15.67 | 25088 | 512.00 | 5674.67 | 43.80 | 4.06 | 8.36 | true |
124 | add_27 | Add | 17.67 | 3.00 | 14.67 | 25088 | 512.00 | 23040.00 | 43.40 | 1.07 | 8.36 | true |
125 | inception_4c/pool-0-TransposeNHWCToNCHW-LayoutOptimizer | Transpose | 27.67 | 4.00 | 23.67 | 0 | 256.00 | 59477.33 | 18.80 | 0.00 | 0.00 | true |
126 | inception_4c/relu_5x5_reduce | Relu | 19.00 | 3.00 | 16.00 | 0 | 0.00 | 213.33 | 41.80 | 0.00 | 0.00 | true |
127 | inception_4c/relu_3x3_reduce | Relu | 17.00 | 3.00 | 14.00 | 0 | 0.00 | 30762.67 | 43.60 | 0.00 | 0.00 | true |
128 | inception_4c/pool | MaxPool | 39.33 | 6.00 | 33.33 | 100352 | 6912.00 | 169685.33 | 45.20 | 0.57 | 16.73 | true |
129 | Pad_11 | Pad | 33.33 | 6.00 | 27.33 | 0 | 6656.00 | 42.67 | 46.40 | 0.00 | 0.00 | true |
130 | Pad_10 | Pad | 25.67 | 5.00 | 20.67 | 0 | 0.00 | 0.00 | 46.80 | 0.00 | 0.00 | true |
131 | convolution_30 | Conv2D | 115.67 | 35.00 | 80.67 | 14692608 | 131584.00 | 373.33 | 7.92 | 111.34 | 419.79 | false |
132 | convolution_32 | Conv2D | 170.00 | 57.33 | 112.67 | 39376384 | 2684469.33 | 7003146.67 | 32.29 | 4.06 | 686.80 | true |
133 | convolution_31 | Conv2D | 145.00 | 54.00 | 91.00 | 72318976 | 1330378.67 | 2588245.33 | 19.25 | 18.46 | 1339.24 | true |
134 | add_30 | Add | 25.33 | 4.00 | 21.33 | 12544 | 53504.00 | 6656.00 | 44.20 | 0.21 | 3.14 | true |
135 | add_32 | Add | 18.00 | 3.00 | 15.00 | 12544 | 256.00 | 35157.33 | 42.00 | 0.35 | 4.18 | true |
136 | add_31 | Add | 18.67 | 3.33 | 15.33 | 50176 | 1536.00 | 85461.33 | 44.60 | 0.58 | 15.05 | true |
137 | inception_4c/output | ConcatV2 | 66.67 | 0.00 | 66.67 | 0 | 0.00 | 0.00 | 0.00 | 0.00 | NaN | true |
138 | inception_4c/relu_1x1 | Relu | 21.33 | 3.00 | 18.33 | 0 | 1536.00 | 295402.67 | 53.70 | 0.00 | 0.00 | true |
139 | inception_4c/relu_1x1-3-0-TransposeNCHWToNHWC-LayoutOptimizer | Transpose | 24.00 | 4.33 | 19.67 | 0 | 6144.00 | 4480.00 | 0.00 | 0.00 | 0.00 | true |
140 | convolution_35 | Conv2D | 119.00 | 37.00 | 82.00 | 7346304 | 74602.67 | 235317.33 | 7.64 | 23.70 | 198.55 | true |
141 | convolution_34 | Conv2D | 112.67 | 35.00 | 77.67 | 36728384 | 294912.00 | 326965.33 | 7.80 | 59.06 | 1049.38 | false |
142 | convolution_33 | Conv2D | 108.67 | 35.00 | 73.67 | 29382080 | 229376.00 | 22538.67 | 7.90 | 116.64 | 839.49 | false |
143 | PadV2_8 | PadV2 | 31.33 | 5.00 | 26.33 | 0 | 4352.00 | 51328.00 | 58.40 | 0.00 | 0.00 | true |
144 | add_35 | Add | 23.67 | 3.00 | 20.67 | 6272 | 469.33 | 640.00 | 42.40 | 5.65 | 2.09 | true |
145 | add_34 | Add | 19.33 | 3.00 | 16.33 | 28224 | 576.00 | 14730.67 | 45.30 | 1.84 | 9.41 | true |
146 | add_33 | Add | 19.00 | 3.00 | 16.00 | 21952 | 448.00 | 51370.67 | 43.80 | 0.42 | 7.32 | true |
147 | inception_4d/pool-0-TransposeNHWCToNCHW-LayoutOptimizer | Transpose | 23.00 | 4.00 | 19.00 | 0 | 0.00 | 33024.00 | 18.80 | 0.00 | 0.00 | true |
148 | inception_4d/relu_5x5_reduce | Relu | 18.67 | 3.00 | 15.67 | 0 | 0.00 | 170.67 | 42.80 | 0.00 | 0.00 | true |
149 | inception_4d/relu_3x3_reduce | Relu | 16.67 | 3.00 | 13.67 | 0 | 0.00 | 50250.67 | 43.80 | 0.00 | 0.00 | true |
150 | inception_4d/pool | MaxPool | 37.33 | 5.67 | 31.67 | 100352 | 6400.00 | 41728.00 | 44.60 | 2.09 | 17.71 | true |
151 | Pad_13 | Pad | 30.67 | 5.00 | 25.67 | 0 | 6656.00 | 0.00 | 43.70 | 0.00 | 0.00 | true |
152 | Pad_12 | Pad | 26.00 | 5.00 | 21.00 | 0 | 0.00 | 42.67 | 46.80 | 0.00 | 0.00 | true |
153 | convolution_36 | Conv2D | 111.67 | 35.00 | 76.67 | 14692608 | 131072.00 | 330.67 | 7.74 | 111.81 | 419.79 | false |
154 | convolution_38 | Conv2D | 178.00 | 62.33 | 115.67 | 51679104 | 6614922.67 | 10286144.00 | 36.30 | 3.06 | 829.08 | true |
155 | convolution_37 | Conv2D | 146.00 | 59.33 | 86.67 | 91431936 | 1660298.67 | 2933834.67 | 20.27 | 19.90 | 1541.00 | true |
156 | add_36 | Add | 25.33 | 4.00 | 21.33 | 12544 | 53760.00 | 6741.33 | 44.50 | 0.21 | 3.14 | true |
157 | add_38 | Add | 18.67 | 3.00 | 15.67 | 12544 | 7125.33 | 6474.67 | 41.90 | 0.92 | 4.18 | true |
158 | add_37 | Add | 18.00 | 3.33 | 14.67 | 56448 | 1664.00 | 109909.33 | 45.10 | 0.51 | 16.94 | true |
159 | inception_4d/output | ConcatV2 | 67.67 | 0.00 | 67.67 | 0 | 0.00 | 0.00 | 0.00 | 0.00 | NaN | true |
160 | inception_4d/relu_1x1 | Relu | 21.33 | 3.00 | 18.33 | 0 | 1536.00 | 265568.00 | 55.50 | 0.00 | 0.00 | true |
161 | inception_4d/relu_1x1-0-0-TransposeNCHWToNHWC-LayoutOptimizer | Transpose | 24.00 | 4.00 | 20.00 | 0 | 0.00 | 0.00 | 0.00 | 0.00 | 0.00 | true |
162 | convolution_41 | Conv2D | 119.67 | 38.00 | 81.67 | 7575680 | 76885.33 | 63125.33 | 7.33 | 54.11 | 199.36 | false |
163 | convolution_40 | Conv2D | 110.67 | 37.00 | 73.67 | 37878400 | 338176.00 | 375818.67 | 8.88 | 53.05 | 1023.74 | false |
164 | convolution_39 | Conv2D | 111.00 | 37.67 | 73.33 | 60605440 | 540672.00 | 411232.00 | 9.48 | 63.67 | 1608.98 | false |
165 | PadV2_9 | PadV2 | 32.00 | 5.00 | 27.00 | 0 | 5376.00 | 85461.33 | 59.30 | 0.00 | 0.00 | true |
166 | add_41 | Add | 22.33 | 3.00 | 19.33 | 6272 | 128.00 | 11690.67 | 42.40 | 0.53 | 2.09 | true |
167 | add_40 | Add | 18.33 | 3.00 | 15.33 | 31360 | 640.00 | 18560.00 | 44.90 | 1.63 | 10.45 | true |
168 | add_39 | Add | 19.33 | 3.00 | 16.33 | 50176 | 1024.00 | 2816.00 | 44.40 | 13.07 | 16.73 | true |
169 | inception_4e/pool-0-TransposeNHWCToNCHW-LayoutOptimizer | Transpose | 22.67 | 4.00 | 18.67 | 0 | 0.00 | 7594.67 | 19.70 | 0.00 | 0.00 | true |
170 | inception_4e/relu_5x5_reduce | Relu | 19.00 | 3.00 | 16.00 | 0 | 0.00 | 0.00 | 42.80 | 0.00 | 0.00 | true |
171 | inception_4e/relu_3x3_reduce | Relu | 17.00 | 3.00 | 14.00 | 0 | 0.00 | 42.67 | 43.50 | 0.00 | 0.00 | true |
172 | inception_4e/pool | MaxPool | 37.00 | 5.00 | 32.00 | 103488 | 6656.00 | 128.00 | 45.90 | 15.25 | 20.70 | true |
173 | Pad_15 | Pad | 31.00 | 5.00 | 26.00 | 0 | 6656.00 | 42.67 | 43.70 | 0.00 | 0.00 | true |
174 | Pad_14 | Pad | 25.33 | 5.00 | 20.33 | 0 | 0.00 | 0.00 | 46.80 | 0.00 | 0.00 | true |
175 | convolution_42 | Conv2D | 114.67 | 36.00 | 78.67 | 30302720 | 270336.00 | 34698.67 | 7.69 | 99.34 | 841.74 | false |
176 | convolution_44 | Conv2D | 205.00 | 93.67 | 111.33 | 102635776 | 17463093.33 | 20911808.00 | 42.31 | 2.67 | 1095.75 | true |
177 | convolution_43 | Conv2D | 151.67 | 66.00 | 85.67 | 112783360 | 2028149.33 | 2913141.33 | 20.68 | 22.82 | 1708.84 | true |
178 | add_42 | Add | 25.67 | 4.00 | 21.67 | 25088 | 104448.00 | 6869.33 | 46.20 | 0.23 | 6.27 | true |
179 | add_44 | Add | 19.67 | 4.00 | 15.67 | 25088 | 55722.67 | 74325.33 | 43.70 | 0.19 | 6.27 | true |
180 | add_43 | Add | 18.00 | 3.00 | 15.00 | 62720 | 1280.00 | 93600.00 | 44.60 | 0.66 | 20.91 | true |
181 | inception_4e/output | ConcatV2 | 67.00 | 0.00 | 67.00 | 0 | 0.00 | 0.00 | 0.00 | 0.00 | NaN | true |
182 | inception_4e/relu_1x1 | Relu | 21.00 | 3.00 | 18.00 | 0 | 1536.00 | 269909.33 | 75.90 | 0.00 | 0.00 | true |
183 | inception_4e/relu_1x1-0-0-TransposeNCHWToNHWC-LayoutOptimizer | Transpose | 24.00 | 5.00 | 19.00 | 0 | 6826.67 | 77152.00 | 25.40 | 0.00 | 0.00 | true |
184 | PadV2_10 | PadV2 | 30.00 | 5.33 | 24.67 | 0 | 5120.00 | 259850.67 | 62.40 | 0.00 | 0.00 | true |
185 | pool4/3x3_s2-0-TransposeNHWCToNCHW-LayoutOptimizer | Transpose | 23.33 | 4.00 | 19.33 | 0 | 170.67 | 746890.67 | 27.50 | 0.00 | 0.00 | true |
186 | pool4/3x3_s2 | MaxPool | 38.00 | 5.33 | 32.67 | 40768 | 6912.00 | 28586.67 | 19.40 | 1.15 | 7.64 | true |
187 | pool4/3x3_s2-0-0-TransposeNCHWToNHWC-LayoutOptimizer | Transpose | 23.33 | 3.00 | 20.33 | 0 | 0.00 | 42.67 | 12.40 | 0.00 | 0.00 | true |
188 | convolution_47 | Conv2D | 138.00 | 56.00 | 82.00 | 3409440 | 117333.33 | 586.67 | 5.95 | 28.91 | 60.88 | false |
189 | convolution_46 | Conv2D | 172.67 | 94.67 | 78.00 | 17047200 | 551424.00 | 362.67 | 5.77 | 30.89 | 180.08 | false |
190 | convolution_45 | Conv2D | 130.33 | 57.00 | 73.33 | 27275520 | 852309.33 | 330.67 | 8.97 | 31.99 | 478.52 | false |
191 | PadV2_11 | PadV2 | 30.33 | 4.00 | 26.33 | 0 | 0.00 | 0.00 | 43.90 | 0.00 | 0.00 | true |
192 | add_47 | Add | 23.33 | 3.00 | 20.33 | 1568 | 384.00 | 42.67 | 32.50 | 3.67 | 0.52 | true |
193 | add_46 | Add | 19.33 | 3.33 | 16.00 | 7840 | 640.00 | 0.00 | 45.20 | 12.25 | 2.35 | true |
194 | add_45 | Add | 20.33 | 4.00 | 16.33 | 12544 | 1024.00 | 42.67 | 42.50 | 11.76 | 3.14 | true |
195 | inception_5a/pool-0-TransposeNHWCToNCHW-LayoutOptimizer | Transpose | 22.00 | 3.00 | 19.00 | 0 | 0.00 | 0.00 | 12.30 | 0.00 | 0.00 | true |
196 | inception_5a/relu_5x5_reduce | Relu | 18.33 | 2.00 | 16.33 | 0 | 0.00 | 42.67 | 38.20 | 0.00 | 0.00 | true |
197 | inception_5a/relu_3x3_reduce | Relu | 17.33 | 3.00 | 14.33 | 0 | 0.00 | 0.00 | 43.90 | 0.00 | 0.00 | true |
198 | inception_5a/pool | MaxPool | 37.33 | 4.00 | 33.33 | 40768 | 85.33 | 170.67 | 18.90 | 159.25 | 10.19 | false |
199 | Pad_17 | Pad | 35.00 | 8.00 | 27.00 | 0 | 10240.00 | 0.00 | 47.60 | 0.00 | 0.00 | true |
200 | Pad_16 | Pad | 30.67 | 6.00 | 24.67 | 0 | 0.00 | 32.00 | 42.90 | 0.00 | 0.00 | true |
201 | convolution_48 | Conv2D | 132.67 | 53.00 | 79.67 | 13637760 | 425984.00 | 416.00 | 7.83 | 31.98 | 257.32 | false |
202 | convolution_50 | Conv2D | 184.67 | 54.33 | 130.33 | 25072512 | 455552.00 | 4456128.00 | 24.98 | 5.10 | 461.46 | true |
203 | convolution_49 | Conv2D | 153.00 | 65.00 | 88.00 | 57876480 | 1918282.67 | 5052960.00 | 21.27 | 8.30 | 890.41 | true |
204 | add_48 | Add | 26.00 | 4.00 | 22.00 | 6272 | 29013.33 | 512.00 | 44.40 | 0.21 | 1.57 | true |
205 | add_50 | Add | 18.33 | 4.00 | 14.33 | 6272 | 3456.00 | 213.33 | 41.60 | 1.71 | 1.57 | true |
206 | add_49 | Add | 18.00 | 4.00 | 14.00 | 15680 | 1280.00 | 128.00 | 45.20 | 11.14 | 3.92 | true |
207 | inception_5a/output | ConcatV2 | 67.33 | 0.00 | 67.33 | 0 | 0.00 | 0.00 | 0.00 | 0.00 | NaN | true |
208 | inception_5a/relu_1x1 | Relu | 20.33 | 3.00 | 17.33 | 0 | 1450.67 | 161226.67 | 45.70 | 0.00 | 0.00 | true |
209 | inception_5a/relu_1x1-0-0-TransposeNCHWToNHWC-LayoutOptimizer | Transpose | 22.33 | 3.67 | 18.67 | 0 | 4864.00 | 32480.00 | 12.40 | 0.00 | 0.00 | true |
210 | convolution_53 | Conv2D | 131.67 | 52.00 | 79.67 | 6818096 | 163242.67 | 187829.33 | 6.28 | 19.42 | 131.12 | true |
211 | convolution_52 | Conv2D | 127.00 | 53.00 | 74.00 | 20456640 | 640597.33 | 528725.33 | 7.79 | 17.49 | 385.97 | true |
212 | convolution_51 | Conv2D | 133.00 | 58.00 | 75.00 | 40913280 | 1277952.00 | 712650.67 | 10.32 | 20.55 | 705.40 | true |
213 | PadV2_12 | PadV2 | 33.00 | 5.00 | 28.00 | 0 | 3072.00 | 42.67 | 44.30 | 0.00 | 0.00 | true |
214 | add_53 | Add | 22.33 | 3.00 | 19.33 | 2352 | 192.00 | 0.00 | 38.30 | 12.25 | 0.78 | true |
215 | add_52 | Add | 19.33 | 3.67 | 15.67 | 9408 | 853.33 | 42.67 | 41.90 | 10.50 | 2.57 | true |
216 | add_51 | Add | 18.00 | 4.00 | 14.00 | 18816 | 2304.00 | 0.00 | 44.20 | 8.17 | 4.70 | true |
217 | inception_5b/pool-0-TransposeNHWCToNCHW-LayoutOptimizer | Transpose | 22.67 | 4.00 | 18.67 | 0 | 426.67 | 170.67 | 12.30 | 0.00 | 0.00 | true |
218 | inception_5b/relu_5x5_reduce | Relu | 19.00 | 3.00 | 16.00 | 0 | 85.33 | 0.00 | 40.80 | 0.00 | 0.00 | true |
219 | inception_5b/relu_3x3_reduce | Relu | 16.67 | 3.00 | 13.67 | 0 | 0.00 | 42.67 | 42.80 | 0.00 | 0.00 | true |
220 | inception_5b/pool | MaxPool | 35.67 | 5.00 | 30.67 | 40768 | 4864.00 | 170.67 | 19.20 | 8.10 | 8.15 | true |
221 | Pad_19 | Pad | 31.67 | 6.33 | 25.33 | 0 | 5376.00 | 42.67 | 39.80 | 0.00 | 0.00 | true |
222 | Pad_18 | Pad | 26.67 | 7.00 | 19.67 | 0 | 0.00 | 42.67 | 45.50 | 0.00 | 0.00 | true |
223 | convolution_54 | Conv2D | 129.33 | 52.67 | 76.67 | 13637760 | 426069.33 | 25450.67 | 7.57 | 30.20 | 258.94 | false |
224 | convolution_56 | Conv2D | 179.00 | 60.00 | 119.00 | 36751104 | 3332192.00 | 8459488.00 | 27.02 | 3.12 | 612.52 | true |
225 | convolution_55 | Conv2D | 162.67 | 76.00 | 86.67 | 83238912 | 3038624.00 | 6647392.00 | 22.44 | 8.59 | 1095.25 | true |
226 | add_54 | Add | 25.67 | 4.00 | 21.67 | 6272 | 26773.33 | 1450.67 | 44.20 | 0.22 | 1.57 | true |
227 | add_56 | Add | 21.00 | 4.00 | 17.00 | 6272 | 13909.33 | 1984.00 | 41.60 | 0.39 | 1.57 | true |
228 | add_55 | Add | 18.67 | 4.00 | 14.67 | 18816 | 2560.00 | 597.33 | 44.50 | 5.96 | 4.70 | true |
229 | inception_5b/output | ConcatV2 | 67.67 | 0.00 | 67.67 | 0 | 0.00 | 0.00 | 0.00 | 0.00 | NaN | true |
230 | inception_5b/relu_1x1 | Relu | 21.00 | 3.00 | 18.00 | 0 | 1280.00 | 149728.00 | 45.60 | 0.00 | 0.00 | true |
231 | pool5/7x7_s1 | AvgPool | 47.33 | 10.00 | 37.33 | 66479 | 12800.00 | 4693.33 | 10.70 | 3.80 | 6.65 | true |
232 | Flatten/flatten/Shape | Shape | 8.33 | 0.00 | 8.33 | 0 | 0.00 | 0.00 | 0.00 | 0.00 | NaN | true |
233 | pool5/7x7_s1-1-0-TransposeNCHWToNHWC-LayoutOptimizer | Transpose | 4.67 | 0.00 | 4.67 | 0 | 0.00 | 0.00 | 0.00 | 0.00 | NaN | true |
234 | Flatten/flatten/Shape-0-0-VecPermuteNCHWToNHWC-LayoutOptimizer | DataFormatVecPermute | 10.00 | 0.00 | 10.00 | 0 | 0.00 | 0.00 | 0.00 | 0.00 | NaN | true |
235 | Flatten/flatten/strided_slice | StridedSlice | 15.67 | 0.00 | 15.67 | 0 | 0.00 | 0.00 | 0.00 | 0.00 | NaN | true |
236 | Flatten/flatten/Reshape/shape | Pack | 9.33 | 0.00 | 9.33 | 0 | 0.00 | 0.00 | 0.00 | 0.00 | NaN | true |
237 | Flatten/flatten/Reshape | Reshape | 5.00 | 0.00 | 5.00 | 0 | 0.00 | 0.00 | 0.00 | 0.00 | NaN | true |
238 | dense/MatMul | MatMul | 69.00 | 15.00 | 54.00 | 2135512 | 4103168.00 | 848597.33 | 6.20 | 0.43 | 142.37 | true |
239 | dense/BiasAdd | BiasAdd | 24.33 | 3.00 | 21.33 | 1000 | 5568.00 | 341.33 | 47.20 | 0.17 | 0.33 | true |
240 | prob | Softmax | 61.33 | 15.00 | 46.33 | 34421 | 12693.33 | 2645.33 | 3.59 | 2.24 | 2.29 | true |
Showing 1 to 239 of 239 entries