Navigation :

GPU Kernel Information Aggregated by Layer

Search:

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_memory_bound
1	resnetv20_conv0_fwd	Convolution	18797.67	133.67	18664.00	true
2	resnetv20_batchnorm1_fwd	BatchNorm	166.67	49.67	117.00	true
3	resnetv20_relu0_fwd	Activation	56.00	45.33	10.67	true
4	resnetv20_pool0_fwd	Pooling	3994.00	52.00	3942.00	true
5	resnetv20_stage1_batchnorm0_fwd	BatchNorm	142.00	10.33	131.67	true
6	resnetv20_stage1_activation0	Activation	21.00	6.00	15.00	true
7	resnetv20_stage1_conv0_fwd	Convolution	18032.00	74.67	17957.33	true
8	resnetv20_stage1_batchnorm1_fwd	BatchNorm	126.00	8.00	118.00	true
9	resnetv20_stage1_activation1	Activation	20.67	6.00	14.67	true
10	resnetv20_stage1_conv1_fwd	Convolution	17993.00	58.67	17934.33	true
11	resnetv20_stage1__plus0	elemwise_add	119.33	10.00	109.33	true
12	resnetv20_stage1_batchnorm2_fwd	BatchNorm	33.67	10.00	23.67	true
13	resnetv20_stage1_activation2	Activation	23.00	6.00	17.00	true
14	resnetv20_stage1_conv2_fwd	Convolution	17997.33	59.33	17938.00	true
15	resnetv20_stage1_batchnorm3_fwd	BatchNorm	123.33	8.00	115.33	true
16	resnetv20_stage1_activation3	Activation	22.33	6.00	16.33	true
17	resnetv20_stage1_conv3_fwd	Convolution	17999.67	58.67	17941.00	true
18	resnetv20_stage1__plus1	elemwise_add	120.00	10.00	110.00	true
19	resnetv20_stage2_batchnorm0_fwd	BatchNorm	25.67	8.00	17.67	true
20	resnetv20_stage2_activation0	Activation	22.33	6.00	16.33	true
21	resnetv20_stage2_conv0_fwd	Convolution	8637.33	144.00	8493.33	true
22	resnetv20_stage2_batchnorm1_fwd	BatchNorm	117.67	6.00	111.67	true
23	resnetv20_stage2_activation1	Activation	16.00	4.00	12.00	true
24	resnetv20_stage2_conv1_fwd	Convolution	16159.00	80.67	16078.33	true
25	resnetv20_stage2_conv2_fwd	Convolution	974.00	44.33	929.67	true
26	resnetv20_stage2__plus0	elemwise_add	114.67	5.00	109.67	true
27	resnetv20_stage2_batchnorm2_fwd	BatchNorm	24.33	6.00	18.33	true
28	resnetv20_stage2_activation2	Activation	16.33	5.00	11.33	true
29	resnetv20_stage2_conv3_fwd	Convolution	16147.33	82.00	16065.33	true
30	resnetv20_stage2_batchnorm3_fwd	BatchNorm	120.00	6.00	114.00	true
31	resnetv20_stage2_activation3	Activation	16.67	4.67	12.00	true
32	resnetv20_stage2_conv4_fwd	Convolution	16143.67	70.00	16073.67	true
33	resnetv20_stage2__plus1	elemwise_add	114.67	6.00	108.67	true
34	resnetv20_stage3_batchnorm0_fwd	BatchNorm	25.00	6.00	19.00	true
35	resnetv20_stage3_activation0	Activation	15.00	4.33	10.67	true
36	resnetv20_stage3_conv0_fwd	Convolution	8252.00	223.33	8028.67	true
37	resnetv20_stage3_batchnorm1_fwd	BatchNorm	119.33	6.00	113.33	true
38	resnetv20_stage3_activation1	Activation	12.67	4.00	8.67	true
39	resnetv20_stage3_conv1_fwd	Convolution	15538.67	133.67	15405.00	true
40	resnetv20_stage3_conv2_fwd	Convolution	845.00	56.00	789.00	true
41	resnetv20_stage3__plus0	elemwise_add	114.00	4.00	110.00	true
42	resnetv20_stage3_batchnorm2_fwd	BatchNorm	21.67	6.00	15.67	true
43	resnetv20_stage3_activation2	Activation	13.67	5.00	8.67	true
44	resnetv20_stage3_conv3_fwd	Convolution	15450.33	136.00	15314.33	true
45	resnetv20_stage3_batchnorm3_fwd	BatchNorm	119.33	5.67	113.67	true
46	resnetv20_stage3_activation3	Activation	14.00	4.00	10.00	true
47	resnetv20_stage3_conv4_fwd	Convolution	15489.00	119.67	15369.33	true
48	resnetv20_stage3__plus1	elemwise_add	113.33	4.00	109.33	true
49	resnetv20_stage4_batchnorm0_fwd	BatchNorm	22.00	6.00	16.00	true
50	resnetv20_stage4_activation0	Activation	13.00	4.00	9.00	true
51	resnetv20_stage4_conv0_fwd	Convolution	9796.67	355.67	9441.00	true
52	resnetv20_stage4_batchnorm1_fwd	BatchNorm	121.00	5.67	115.33	true
53	resnetv20_stage4_activation1	Activation	11.33	4.00	7.33	true
54	resnetv20_stage4_conv1_fwd	Convolution	18875.33	359.00	18516.33	true
55	resnetv20_stage4_conv2_fwd	Convolution	1013.33	82.00	931.33	true
56	resnetv20_stage4__plus0	elemwise_add	115.67	3.00	112.67	true
57	resnetv20_stage4_batchnorm2_fwd	BatchNorm	21.00	5.00	16.00	true
58	resnetv20_stage4_activation2	Activation	12.00	4.00	8.00	true
59	resnetv20_stage4_conv3_fwd	Convolution	18874.33	358.00	18516.33	true
60	resnetv20_stage4_batchnorm3_fwd	BatchNorm	120.67	5.00	115.67	true
61	resnetv20_stage4_activation3	Activation	12.33	4.00	8.33	true
62	resnetv20_stage4_conv4_fwd	Convolution	18894.00	341.00	18553.00	true
63	resnetv20_stage4__plus1	elemwise_add	119.00	3.67	115.33	true
64	resnetv20_batchnorm2_fwd	BatchNorm	21.67	5.00	16.67	true
65	resnetv20_relu1_fwd	Activation	12.00	4.00	8.00	true
66	resnetv20_pool1_fwd	Pooling	77.67	18.00	59.67	true
67	resnetv20_flatten0_flatten0	Flatten	2.00	0.00	2.00	true
68	resnetv20_dense0_fwd	FullyConnected	828.67	25.00	803.67	true

Showing 1 to 68 of 68 entries

Download as CSV