Created
February 4, 2020 02:41
-
-
Save ChaiBapchya/d5ecc3e971c5a3c558d672477b4b6b9c to your computer and use it in GitHub Desktop.
MXNet 1.6.0rc2 OpPerf CPU benchmark results
This file contains bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
{ | |
"BatchNorm": [ | |
{ | |
"avg_time_backward_BatchNorm": 1.5163, | |
"avg_time_forward_BatchNorm": 1.4379, | |
"inputs": { | |
"beta": [ | |
3 | |
], | |
"data": [ | |
32, | |
3, | |
256, | |
256 | |
], | |
"gamma": [ | |
3 | |
], | |
"moving_mean": [ | |
3 | |
], | |
"moving_var": [ | |
3 | |
] | |
}, | |
"max_storage_mem_alloc_cpu/0": 25165.8359 | |
}, | |
{ | |
"avg_time_backward_BatchNorm": 2.5419, | |
"avg_time_forward_BatchNorm": 5.6138, | |
"inputs": { | |
"beta": [ | |
3 | |
], | |
"data": [ | |
32, | |
3, | |
10000, | |
10 | |
], | |
"gamma": [ | |
3 | |
], | |
"moving_mean": [ | |
3 | |
], | |
"moving_var": [ | |
3 | |
] | |
}, | |
"max_storage_mem_alloc_cpu/0": 38400.0117 | |
} | |
], | |
"BlockGrad": [ | |
{ | |
"avg_time_forward_BlockGrad": 0.5422, | |
"inputs": { | |
"data": [ | |
1024, | |
1024 | |
] | |
}, | |
"max_storage_mem_alloc_cpu/0": 2097.1521 | |
}, | |
{ | |
"avg_time_forward_BlockGrad": 0.0103, | |
"inputs": { | |
"data": [ | |
10000, | |
1 | |
] | |
}, | |
"max_storage_mem_alloc_cpu/0": 20.0 | |
}, | |
{ | |
"avg_time_forward_BlockGrad": 0.5887, | |
"inputs": { | |
"data": [ | |
10000, | |
100 | |
] | |
}, | |
"max_storage_mem_alloc_cpu/0": 2000.0 | |
} | |
], | |
"Convolution": [ | |
{ | |
"avg_time_backward_Convolution": 0.74, | |
"avg_time_forward_Convolution": 0.4593, | |
"inputs": { | |
"bias": [ | |
64 | |
], | |
"data": [ | |
32, | |
3, | |
256 | |
], | |
"dilate": [ | |
1 | |
], | |
"kernel": [ | |
3 | |
], | |
"layout": "NCW", | |
"num_filter": 64, | |
"pad": [ | |
0 | |
], | |
"stride": [ | |
1 | |
], | |
"weight": [ | |
64, | |
3, | |
3 | |
] | |
}, | |
"max_storage_mem_alloc_cpu/0": 2080.7681 | |
} | |
], | |
"Deconvolution": [ | |
{ | |
"avg_time_backward_Deconvolution": 7.7016, | |
"avg_time_forward_Deconvolution": 2.3705, | |
"inputs": { | |
"adj": [ | |
0 | |
], | |
"bias": [ | |
64 | |
], | |
"data": [ | |
32, | |
3, | |
256 | |
], | |
"dilate": [ | |
1 | |
], | |
"kernel": [ | |
3 | |
], | |
"layout": "NCW", | |
"no_bias": false, | |
"num_filter": 64, | |
"pad": [ | |
0 | |
], | |
"stride": [ | |
1 | |
], | |
"weight": [ | |
3, | |
64, | |
3 | |
] | |
}, | |
"max_storage_mem_alloc_cpu/0": 2683671.75 | |
} | |
], | |
"Dropout": [ | |
{ | |
"avg_time_backward_Dropout": 0.7059, | |
"avg_time_forward_Dropout": 3.4594, | |
"inputs": { | |
"data": [ | |
32, | |
3, | |
256, | |
256 | |
], | |
"mode": "always", | |
"p": 0.5 | |
}, | |
"max_storage_mem_alloc_cpu/0": 25165.8242 | |
}, | |
{ | |
"avg_time_backward_Dropout": 0.1029, | |
"avg_time_forward_Dropout": 0.4645, | |
"inputs": { | |
"data": [ | |
10000, | |
10 | |
], | |
"mode": "always", | |
"p": 0.5 | |
}, | |
"max_storage_mem_alloc_cpu/0": 400.0 | |
} | |
], | |
"FullyConnected": [ | |
{ | |
"avg_time_backward_FullyConnected": 84.3464, | |
"avg_time_forward_FullyConnected": 1.1579, | |
"inputs": { | |
"bias": [ | |
64 | |
], | |
"data": [ | |
32, | |
3, | |
256, | |
256 | |
], | |
"flatten": true, | |
"num_hidden": 64, | |
"weight": [ | |
64, | |
196608 | |
] | |
}, | |
"max_storage_mem_alloc_cpu/0": 8.192 | |
}, | |
{ | |
"avg_time_backward_FullyConnected": 30.3979, | |
"avg_time_forward_FullyConnected": 1.1499, | |
"inputs": { | |
"bias": [ | |
64 | |
], | |
"data": [ | |
32, | |
3, | |
256, | |
256 | |
], | |
"flatten": false, | |
"num_hidden": 64, | |
"weight": [ | |
64, | |
256 | |
] | |
}, | |
"max_storage_mem_alloc_cpu/0": 6291.4561 | |
} | |
], | |
"LeakyReLU": [ | |
{ | |
"avg_time_backward_LeakyReLU": 0.0755, | |
"avg_time_forward_LeakyReLU": 0.0679, | |
"inputs": { | |
"act_type": "leaky", | |
"data": [ | |
1024, | |
1024 | |
], | |
"slope": 0.1 | |
}, | |
"max_storage_mem_alloc_cpu/0": 2097.1521 | |
}, | |
{ | |
"avg_time_backward_LeakyReLU": 0.0398, | |
"avg_time_forward_LeakyReLU": 0.0422, | |
"inputs": { | |
"act_type": "leaky", | |
"data": [ | |
10000, | |
1 | |
], | |
"slope": 0.1 | |
}, | |
"max_storage_mem_alloc_cpu/0": 20.0 | |
}, | |
{ | |
"avg_time_backward_LeakyReLU": 0.0872, | |
"avg_time_forward_LeakyReLU": 0.3997, | |
"inputs": { | |
"act_type": "leaky", | |
"data": [ | |
10000, | |
100 | |
], | |
"slope": 0.1 | |
}, | |
"max_storage_mem_alloc_cpu/0": 2000.0 | |
}, | |
{ | |
"avg_time_backward_LeakyReLU": 0.8136, | |
"avg_time_forward_LeakyReLU": 0.147, | |
"inputs": { | |
"act_type": "elu", | |
"data": [ | |
1024, | |
1024 | |
], | |
"slope": 0.1 | |
}, | |
"max_storage_mem_alloc_cpu/0": 2097.1521 | |
}, | |
{ | |
"avg_time_backward_LeakyReLU": 0.0528, | |
"avg_time_forward_LeakyReLU": 0.053, | |
"inputs": { | |
"act_type": "elu", | |
"data": [ | |
10000, | |
1 | |
], | |
"slope": 0.1 | |
}, | |
"max_storage_mem_alloc_cpu/0": 40.0 | |
}, | |
{ | |
"avg_time_backward_LeakyReLU": 0.7188, | |
"avg_time_forward_LeakyReLU": 0.5673, | |
"inputs": { | |
"act_type": "elu", | |
"data": [ | |
10000, | |
100 | |
], | |
"slope": 0.1 | |
}, | |
"max_storage_mem_alloc_cpu/0": 4000.0 | |
}, | |
{ | |
"avg_time_backward_LeakyReLU": 0.3818, | |
"avg_time_forward_LeakyReLU": 0.9016, | |
"inputs": { | |
"act_type": "selu", | |
"data": [ | |
1024, | |
1024 | |
] | |
}, | |
"max_storage_mem_alloc_cpu/0": 2097.1521 | |
}, | |
{ | |
"avg_time_backward_LeakyReLU": 0.0293, | |
"avg_time_forward_LeakyReLU": 0.0355, | |
"inputs": { | |
"act_type": "selu", | |
"data": [ | |
10000, | |
1 | |
] | |
}, | |
"max_storage_mem_alloc_cpu/0": 20.0 | |
}, | |
{ | |
"avg_time_backward_LeakyReLU": 0.3341, | |
"avg_time_forward_LeakyReLU": 0.8673, | |
"inputs": { | |
"act_type": "selu", | |
"data": [ | |
10000, | |
100 | |
] | |
}, | |
"max_storage_mem_alloc_cpu/0": 2000.0 | |
}, | |
{ | |
"avg_time_backward_LeakyReLU": 2.817, | |
"avg_time_forward_LeakyReLU": 0.4101, | |
"inputs": { | |
"act_type": "prelu", | |
"data": [ | |
1024, | |
1024 | |
], | |
"gamma": [ | |
1, | |
1024 | |
] | |
}, | |
"max_storage_mem_alloc_cpu/0": 2097.1521 | |
}, | |
{ | |
"avg_time_backward_LeakyReLU": 0.2074, | |
"avg_time_forward_LeakyReLU": 0.0334, | |
"inputs": { | |
"act_type": "prelu", | |
"data": [ | |
10000, | |
1 | |
], | |
"gamma": [ | |
1, | |
1 | |
] | |
}, | |
"max_storage_mem_alloc_cpu/0": 20.0 | |
}, | |
{ | |
"avg_time_backward_LeakyReLU": 2.6711, | |
"avg_time_forward_LeakyReLU": 0.5885, | |
"inputs": { | |
"act_type": "prelu", | |
"data": [ | |
10000, | |
100 | |
], | |
"gamma": [ | |
1, | |
100 | |
] | |
}, | |
"max_storage_mem_alloc_cpu/0": 4000.0 | |
} | |
], | |
"Pooling": [ | |
{ | |
"avg_time_backward_Pooling": 0.2234, | |
"avg_time_forward_Pooling": 0.1525, | |
"inputs": { | |
"data": [ | |
32, | |
3, | |
256 | |
], | |
"global_pool": 0, | |
"kernel": 3, | |
"pad": 1, | |
"pool_type": "avg", | |
"stride": 1 | |
}, | |
"max_storage_mem_alloc_cpu/0": 49.152 | |
} | |
], | |
"abs": [ | |
{ | |
"avg_time_backward_abs": 0.4008, | |
"avg_time_forward_abs": 0.0677, | |
"inputs": { | |
"data": [ | |
1024, | |
1024 | |
] | |
}, | |
"max_storage_mem_alloc_cpu/0": 2097.1521 | |
}, | |
{ | |
"avg_time_backward_abs": 0.1292, | |
"avg_time_forward_abs": 0.0255, | |
"inputs": { | |
"data": [ | |
10000, | |
1 | |
] | |
}, | |
"max_storage_mem_alloc_cpu/0": 20.0 | |
}, | |
{ | |
"avg_time_backward_abs": 0.4151, | |
"avg_time_forward_abs": 0.149, | |
"inputs": { | |
"data": [ | |
10000, | |
100 | |
] | |
}, | |
"max_storage_mem_alloc_cpu/0": 2000.0 | |
} | |
], | |
"adam_update": [ | |
{ | |
"avg_time_forward_adam_update": 0.5096, | |
"inputs": { | |
"beta1": 0.1, | |
"beta2": 0.1, | |
"epsilon": 1e-08, | |
"grad": [ | |
1024, | |
1024 | |
], | |
"lazy_update": 0, | |
"lr": 0.1, | |
"mean": [ | |
1024, | |
1024 | |
], | |
"rescale_grad": 0.4, | |
"var": [ | |
1024, | |
1024 | |
], | |
"weight": [ | |
1024, | |
1024 | |
] | |
}, | |
"max_storage_mem_alloc_cpu/0": 2616338.5 | |
}, | |
{ | |
"avg_time_forward_adam_update": 0.0419, | |
"inputs": { | |
"beta1": 0.5, | |
"beta2": 0.5, | |
"epsilon": 1e-08, | |
"grad": [ | |
10000, | |
1 | |
], | |
"lazy_update": 0, | |
"lr": 0.5, | |
"mean": [ | |
10000, | |
1 | |
], | |
"rescale_grad": 0.4, | |
"var": [ | |
10000, | |
1 | |
], | |
"weight": [ | |
10000, | |
1 | |
] | |
}, | |
"max_storage_mem_alloc_cpu/0": 2616378.5 | |
}, | |
{ | |
"avg_time_forward_adam_update": 1.7834, | |
"inputs": { | |
"beta1": 0.9, | |
"beta2": 0.9, | |
"epsilon": 1e-08, | |
"grad": [ | |
10000, | |
100 | |
], | |
"lazy_update": 0, | |
"lr": 0.9, | |
"mean": [ | |
10000, | |
100 | |
], | |
"rescale_grad": 0.4, | |
"var": [ | |
10000, | |
100 | |
], | |
"weight": [ | |
10000, | |
100 | |
] | |
}, | |
"max_storage_mem_alloc_cpu/0": 2620358.5 | |
} | |
], | |
"arccos": [ | |
{ | |
"avg_time_backward_arccos": 0.706, | |
"avg_time_forward_arccos": 0.8168, | |
"inputs": { | |
"data": [ | |
1024, | |
1024 | |
] | |
}, | |
"max_storage_mem_alloc_cpu/0": 2097.1521 | |
}, | |
{ | |
"avg_time_backward_arccos": 0.0262, | |
"avg_time_forward_arccos": 0.0284, | |
"inputs": { | |
"data": [ | |
10000, | |
1 | |
] | |
}, | |
"max_storage_mem_alloc_cpu/0": 20.0 | |
}, | |
{ | |
"avg_time_backward_arccos": 0.863, | |
"avg_time_forward_arccos": 0.8796, | |
"inputs": { | |
"data": [ | |
10000, | |
100 | |
] | |
}, | |
"max_storage_mem_alloc_cpu/0": 2000.0 | |
} | |
], | |
"arccosh": [ | |
{ | |
"avg_time_backward_arccosh": 0.9237, | |
"avg_time_forward_arccosh": 0.7149, | |
"inputs": { | |
"data": [ | |
1024, | |
1024 | |
] | |
}, | |
"max_storage_mem_alloc_cpu/0": 2097.1521 | |
}, | |
{ | |
"avg_time_backward_arccosh": 0.025, | |
"avg_time_forward_arccosh": 0.028, | |
"inputs": { | |
"data": [ | |
10000, | |
1 | |
] | |
}, | |
"max_storage_mem_alloc_cpu/0": 20.0 | |
}, | |
{ | |
"avg_time_backward_arccosh": 0.7771, | |
"avg_time_forward_arccosh": 0.7297, | |
"inputs": { | |
"data": [ | |
10000, | |
100 | |
] | |
}, | |
"max_storage_mem_alloc_cpu/0": 2000.0 | |
} | |
], | |
"arcsin": [ | |
{ | |
"avg_time_backward_arcsin": 0.7384, | |
"avg_time_forward_arcsin": 0.7926, | |
"inputs": { | |
"data": [ | |
1024, | |
1024 | |
] | |
}, | |
"max_storage_mem_alloc_cpu/0": 2097.1521 | |
}, | |
{ | |
"avg_time_backward_arcsin": 0.0315, | |
"avg_time_forward_arcsin": 0.0336, | |
"inputs": { | |
"data": [ | |
10000, | |
1 | |
] | |
}, | |
"max_storage_mem_alloc_cpu/0": 20.0 | |
}, | |
{ | |
"avg_time_backward_arcsin": 0.8458, | |
"avg_time_forward_arcsin": 0.8354, | |
"inputs": { | |
"data": [ | |
10000, | |
100 | |
] | |
}, | |
"max_storage_mem_alloc_cpu/0": 2000.0 | |
} | |
], | |
"arcsinh": [ | |
{ | |
"avg_time_backward_arcsinh": 0.4782, | |
"avg_time_forward_arcsinh": 1.6894, | |
"inputs": { | |
"data": [ | |
1024, | |
1024 | |
] | |
}, | |
"max_storage_mem_alloc_cpu/0": 2097.1521 | |
}, | |
{ | |
"avg_time_backward_arcsinh": 0.0289, | |
"avg_time_forward_arcsinh": 0.0391, | |
"inputs": { | |
"data": [ | |
10000, | |
1 | |
] | |
}, | |
"max_storage_mem_alloc_cpu/0": 20.0 | |
}, | |
{ | |
"avg_time_backward_arcsinh": 0.4585, | |
"avg_time_forward_arcsinh": 1.6076, | |
"inputs": { | |
"data": [ | |
10000, | |
100 | |
] | |
}, | |
"max_storage_mem_alloc_cpu/0": 4000.0 | |
} | |
], | |
"arctan": [ | |
{ | |
"avg_time_backward_arctan": 0.0777, | |
"avg_time_forward_arctan": 0.8957, | |
"inputs": { | |
"data": [ | |
1024, | |
1024 | |
] | |
}, | |
"max_storage_mem_alloc_cpu/0": 2097.1521 | |
}, | |
{ | |
"avg_time_backward_arctan": 0.0347, | |
"avg_time_forward_arctan": 0.037, | |
"inputs": { | |
"data": [ | |
10000, | |
1 | |
] | |
}, | |
"max_storage_mem_alloc_cpu/0": 20.0 | |
}, | |
{ | |
"avg_time_backward_arctan": 0.0824, | |
"avg_time_forward_arctan": 0.8552, | |
"inputs": { | |
"data": [ | |
10000, | |
100 | |
] | |
}, | |
"max_storage_mem_alloc_cpu/0": 2000.0 | |
} | |
], | |
"arctanh": [ | |
{ | |
"avg_time_backward_arctanh": 0.0854, | |
"avg_time_forward_arctanh": 1.3269, | |
"inputs": { | |
"data": [ | |
1024, | |
1024 | |
] | |
}, | |
"max_storage_mem_alloc_cpu/0": 2097.1521 | |
}, | |
{ | |
"avg_time_backward_arctanh": 0.0364, | |
"avg_time_forward_arctanh": 0.0428, | |
"inputs": { | |
"data": [ | |
10000, | |
1 | |
] | |
}, | |
"max_storage_mem_alloc_cpu/0": 40.0 | |
}, | |
{ | |
"avg_time_backward_arctanh": 0.0907, | |
"avg_time_forward_arctanh": 1.4483, | |
"inputs": { | |
"data": [ | |
10000, | |
100 | |
] | |
}, | |
"max_storage_mem_alloc_cpu/0": 4000.0 | |
} | |
], | |
"argmax": [ | |
{ | |
"avg_time_forward_argmax": 7.3755, | |
"inputs": { | |
"axis": 0, | |
"data": [ | |
1024, | |
1024 | |
] | |
}, | |
"max_storage_mem_alloc_cpu/0": 2.048 | |
}, | |
{ | |
"avg_time_forward_argmax": 0.084, | |
"inputs": { | |
"axis": 0, | |
"data": [ | |
10000, | |
1 | |
] | |
}, | |
"max_storage_mem_alloc_cpu/0": 0.002 | |
}, | |
{ | |
"avg_time_forward_argmax": 8.0125, | |
"inputs": { | |
"axis": 0, | |
"data": [ | |
10000, | |
100 | |
] | |
}, | |
"max_storage_mem_alloc_cpu/0": 0.4 | |
} | |
], | |
"argmax_channel": [ | |
{ | |
"avg_time_forward_argmax_channel": 0.3031, | |
"inputs": { | |
"data": [ | |
1024, | |
1024 | |
] | |
}, | |
"max_storage_mem_alloc_cpu/0": 2.048 | |
}, | |
{ | |
"avg_time_forward_argmax_channel": 0.0354, | |
"inputs": { | |
"data": [ | |
10000, | |
1 | |
] | |
}, | |
"max_storage_mem_alloc_cpu/0": 20.0 | |
}, | |
{ | |
"avg_time_forward_argmax_channel": 0.3009, | |
"inputs": { | |
"data": [ | |
10000, | |
100 | |
] | |
}, | |
"max_storage_mem_alloc_cpu/0": 40.0 | |
} | |
], | |
"argmin": [ | |
{ | |
"avg_time_forward_argmin": 13.5092, | |
"inputs": { | |
"axis": 0, | |
"data": [ | |
1024, | |
1024 | |
] | |
}, | |
"max_storage_mem_alloc_cpu/0": 4.096 | |
}, | |
{ | |
"avg_time_forward_argmin": 0.0842, | |
"inputs": { | |
"axis": 0, | |
"data": [ | |
10000, | |
1 | |
] | |
}, | |
"max_storage_mem_alloc_cpu/0": 0.002 | |
}, | |
{ | |
"avg_time_forward_argmin": 10.8974, | |
"inputs": { | |
"axis": 0, | |
"data": [ | |
10000, | |
100 | |
] | |
}, | |
"max_storage_mem_alloc_cpu/0": 0.4 | |
} | |
], | |
"argsort": [ | |
{ | |
"avg_time_forward_argsort": 23.5732, | |
"inputs": { | |
"axis": 0, | |
"data": [ | |
1024, | |
1024 | |
] | |
}, | |
"max_storage_mem_alloc_cpu/0": 2097.1521 | |
}, | |
{ | |
"avg_time_forward_argsort": 1.0202, | |
"inputs": { | |
"axis": 0, | |
"data": [ | |
10000, | |
1 | |
] | |
}, | |
"max_storage_mem_alloc_cpu/0": 20.0 | |
}, | |
{ | |
"avg_time_forward_argsort": 20.9048, | |
"inputs": { | |
"axis": 0, | |
"data": [ | |
10000, | |
100 | |
] | |
}, | |
"max_storage_mem_alloc_cpu/0": 2000.0 | |
} | |
], | |
"batch_dot": [ | |
{ | |
"avg_time_forward_batch_dot": 566.4906, | |
"inputs": { | |
"lhs": [ | |
32, | |
1024, | |
1024 | |
], | |
"rhs": [ | |
32, | |
1024, | |
1024 | |
] | |
}, | |
"max_storage_mem_alloc_cpu/0": 134217.7344 | |
}, | |
{ | |
"avg_time_forward_batch_dot": 382.0975, | |
"inputs": { | |
"lhs": [ | |
32, | |
1000, | |
10 | |
], | |
"rhs": [ | |
32, | |
1000, | |
10 | |
], | |
"transpose_b": true | |
}, | |
"max_storage_mem_alloc_cpu/0": 128000.0 | |
}, | |
{ | |
"avg_time_forward_batch_dot": 3.4849, | |
"inputs": { | |
"lhs": [ | |
32, | |
1000, | |
1 | |
], | |
"rhs": [ | |
32, | |
100, | |
1000 | |
], | |
"transpose_a": true, | |
"transpose_b": true | |
}, | |
"max_storage_mem_alloc_cpu/0": 12.8 | |
} | |
], | |
"broadcast_add": [ | |
{ | |
"avg_time_backward_broadcast_add": 0.084, | |
"avg_time_forward_broadcast_add": 0.255, | |
"inputs": { | |
"lhs": [ | |
1024, | |
1024 | |
], | |
"rhs": [ | |
1024, | |
1024 | |
] | |
}, | |
"max_storage_mem_alloc_cpu/0": 4194.3042 | |
}, | |
{ | |
"avg_time_backward_broadcast_add": 0.0463, | |
"avg_time_forward_broadcast_add": 0.0426, | |
"inputs": { | |
"lhs": [ | |
10000, | |
10 | |
], | |
"rhs": [ | |
10000, | |
10 | |
] | |
}, | |
"max_storage_mem_alloc_cpu/0": 400.0 | |
}, | |
{ | |
"avg_time_backward_broadcast_add": 0.0335, | |
"avg_time_forward_broadcast_add": 0.0383, | |
"inputs": { | |
"lhs": [ | |
10000, | |
1 | |
], | |
"rhs": [ | |
10000, | |
1 | |
] | |
}, | |
"max_storage_mem_alloc_cpu/0": 40.0 | |
} | |
], | |
"broadcast_div": [ | |
{ | |
"avg_time_backward_broadcast_div": 0.119, | |
"avg_time_forward_broadcast_div": 0.1674, | |
"inputs": { | |
"lhs": [ | |
1024, | |
1024 | |
], | |
"rhs": [ | |
1024, | |
1024 | |
] | |
}, | |
"max_storage_mem_alloc_cpu/0": 4194.3042 | |
}, | |
{ | |
"avg_time_backward_broadcast_div": 0.0439, | |
"avg_time_forward_broadcast_div": 0.0299, | |
"inputs": { | |
"lhs": [ | |
10000, | |
10 | |
], | |
"rhs": [ | |
10000, | |
10 | |
] | |
}, | |
"max_storage_mem_alloc_cpu/0": 200.0 | |
}, | |
{ | |
"avg_time_backward_broadcast_div": 0.0842, | |
"avg_time_forward_broadcast_div": 0.0487, | |
"inputs": { | |
"lhs": [ | |
10000, | |
1 | |
], | |
"rhs": [ | |
10000, | |
1 | |
] | |
}, | |
"max_storage_mem_alloc_cpu/0": 20.0 | |
} | |
], | |
"broadcast_equal": [ | |
{ | |
"avg_time_forward_broadcast_equal": 0.1657, | |
"inputs": { | |
"lhs": [ | |
1024, | |
1024 | |
], | |
"rhs": [ | |
1024, | |
1024 | |
] | |
}, | |
"max_storage_mem_alloc_cpu/0": 4194.3042 | |
}, | |
{ | |
"avg_time_forward_broadcast_equal": 0.0388, | |
"inputs": { | |
"lhs": [ | |
10000, | |
10 | |
], | |
"rhs": [ | |
10000, | |
10 | |
] | |
}, | |
"max_storage_mem_alloc_cpu/0": 400.0 | |
}, | |
{ | |
"avg_time_forward_broadcast_equal": 0.0334, | |
"inputs": { | |
"lhs": [ | |
10000, | |
1 | |
], | |
"rhs": [ | |
10000, | |
1 | |
] | |
}, | |
"max_storage_mem_alloc_cpu/0": 20.0 | |
} | |
], | |
"broadcast_greater": [ | |
{ | |
"avg_time_forward_broadcast_greater": 0.0752, | |
"inputs": { | |
"lhs": [ | |
1024, | |
1024 | |
], | |
"rhs": [ | |
1024, | |
1024 | |
] | |
}, | |
"max_storage_mem_alloc_cpu/0": 2097.1521 | |
}, | |
{ | |
"avg_time_forward_broadcast_greater": 0.0353, | |
"inputs": { | |
"lhs": [ | |
10000, | |
10 | |
], | |
"rhs": [ | |
10000, | |
10 | |
] | |
}, | |
"max_storage_mem_alloc_cpu/0": 400.0 | |
}, | |
{ | |
"avg_time_forward_broadcast_greater": 0.0346, | |
"inputs": { | |
"lhs": [ | |
10000, | |
1 | |
], | |
"rhs": [ | |
10000, | |
1 | |
] | |
}, | |
"max_storage_mem_alloc_cpu/0": 20.0 | |
} | |
], | |
"broadcast_greater_equal": [ | |
{ | |
"avg_time_forward_broadcast_greater_equal": 0.0743, | |
"inputs": { | |
"lhs": [ | |
1024, | |
1024 | |
], | |
"rhs": [ | |
1024, | |
1024 | |
] | |
}, | |
"max_storage_mem_alloc_cpu/0": 2097.1521 | |
}, | |
{ | |
"avg_time_forward_broadcast_greater_equal": 0.0378, | |
"inputs": { | |
"lhs": [ | |
10000, | |
10 | |
], | |
"rhs": [ | |
10000, | |
10 | |
] | |
}, | |
"max_storage_mem_alloc_cpu/0": 200.0 | |
}, | |
{ | |
"avg_time_forward_broadcast_greater_equal": 0.0344, | |
"inputs": { | |
"lhs": [ | |
10000, | |
1 | |
], | |
"rhs": [ | |
10000, | |
1 | |
] | |
}, | |
"max_storage_mem_alloc_cpu/0": 40.0 | |
} | |
], | |
"broadcast_hypot": [ | |
{ | |
"avg_time_backward_broadcast_hypot": 0.9103, | |
"avg_time_forward_broadcast_hypot": 0.4937, | |
"inputs": { | |
"lhs": [ | |
1024, | |
1024 | |
], | |
"rhs": [ | |
1024, | |
1024 | |
] | |
}, | |
"max_storage_mem_alloc_cpu/0": 2097.1521 | |
}, | |
{ | |
"avg_time_backward_broadcast_hypot": 0.1245, | |
"avg_time_forward_broadcast_hypot": 0.1003, | |
"inputs": { | |
"lhs": [ | |
10000, | |
10 | |
], | |
"rhs": [ | |
10000, | |
10 | |
] | |
}, | |
"max_storage_mem_alloc_cpu/0": 400.0 | |
}, | |
{ | |
"avg_time_backward_broadcast_hypot": 0.0427, | |
"avg_time_forward_broadcast_hypot": 0.0304, | |
"inputs": { | |
"lhs": [ | |
10000, | |
1 | |
], | |
"rhs": [ | |
10000, | |
1 | |
] | |
}, | |
"max_storage_mem_alloc_cpu/0": 20.0 | |
} | |
], | |
"broadcast_lesser": [ | |
{ | |
"avg_time_forward_broadcast_lesser": 0.0736, | |
"inputs": { | |
"lhs": [ | |
1024, | |
1024 | |
], | |
"rhs": [ | |
1024, | |
1024 | |
] | |
}, | |
"max_storage_mem_alloc_cpu/0": 4194.3042 | |
}, | |
{ | |
"avg_time_forward_broadcast_lesser": 0.0349, | |
"inputs": { | |
"lhs": [ | |
10000, | |
10 | |
], | |
"rhs": [ | |
10000, | |
10 | |
] | |
}, | |
"max_storage_mem_alloc_cpu/0": 200.0 | |
}, | |
{ | |
"avg_time_forward_broadcast_lesser": 0.032, | |
"inputs": { | |
"lhs": [ | |
10000, | |
1 | |
], | |
"rhs": [ | |
10000, | |
1 | |
] | |
}, | |
"max_storage_mem_alloc_cpu/0": 20.0 | |
} | |
], | |
"broadcast_lesser_equal": [ | |
{ | |
"avg_time_forward_broadcast_lesser_equal": 0.0652, | |
"inputs": { | |
"lhs": [ | |
1024, | |
1024 | |
], | |
"rhs": [ | |
1024, | |
1024 | |
] | |
}, | |
"max_storage_mem_alloc_cpu/0": 2097.1521 | |
}, | |
{ | |
"avg_time_forward_broadcast_lesser_equal": 0.0361, | |
"inputs": { | |
"lhs": [ | |
10000, | |
10 | |
], | |
"rhs": [ | |
10000, | |
10 | |
] | |
}, | |
"max_storage_mem_alloc_cpu/0": 200.0 | |
}, | |
{ | |
"avg_time_forward_broadcast_lesser_equal": 0.0314, | |
"inputs": { | |
"lhs": [ | |
10000, | |
1 | |
], | |
"rhs": [ | |
10000, | |
1 | |
] | |
}, | |
"max_storage_mem_alloc_cpu/0": 20.0 | |
} | |
], | |
"broadcast_logical_and": [ | |
{ | |
"avg_time_forward_broadcast_logical_and": 0.0642, | |
"inputs": { | |
"lhs": [ | |
1024, | |
1024 | |
], | |
"rhs": [ | |
1024, | |
1024 | |
] | |
}, | |
"max_storage_mem_alloc_cpu/0": 2097.1521 | |
}, | |
{ | |
"avg_time_forward_broadcast_logical_and": 0.0374, | |
"inputs": { | |
"lhs": [ | |
10000, | |
10 | |
], | |
"rhs": [ | |
10000, | |
10 | |
] | |
}, | |
"max_storage_mem_alloc_cpu/0": 200.0 | |
}, | |
{ | |
"avg_time_forward_broadcast_logical_and": 0.0425, | |
"inputs": { | |
"lhs": [ | |
10000, | |
1 | |
], | |
"rhs": [ | |
10000, | |
1 | |
] | |
}, | |
"max_storage_mem_alloc_cpu/0": 20.0 | |
} | |
], | |
"broadcast_logical_or": [ | |
{ | |
"avg_time_forward_broadcast_logical_or": 0.066, | |
"inputs": { | |
"lhs": [ | |
1024, | |
1024 | |
], | |
"rhs": [ | |
1024, | |
1024 | |
] | |
}, | |
"max_storage_mem_alloc_cpu/0": 2097.1521 | |
}, | |
{ | |
"avg_time_forward_broadcast_logical_or": 0.0358, | |
"inputs": { | |
"lhs": [ | |
10000, | |
10 | |
], | |
"rhs": [ | |
10000, | |
10 | |
] | |
}, | |
"max_storage_mem_alloc_cpu/0": 200.0 | |
}, | |
{ | |
"avg_time_forward_broadcast_logical_or": 0.0437, | |
"inputs": { | |
"lhs": [ | |
10000, | |
1 | |
], | |
"rhs": [ | |
10000, | |
1 | |
] | |
}, | |
"max_storage_mem_alloc_cpu/0": 20.0 | |
} | |
], | |
"broadcast_logical_xor": [ | |
{ | |
"avg_time_forward_broadcast_logical_xor": 0.1299, | |
"inputs": { | |
"lhs": [ | |
1024, | |
1024 | |
], | |
"rhs": [ | |
1024, | |
1024 | |
] | |
}, | |
"max_storage_mem_alloc_cpu/0": 2097.1521 | |
}, | |
{ | |
"avg_time_forward_broadcast_logical_xor": 0.0413, | |
"inputs": { | |
"lhs": [ | |
10000, | |
10 | |
], | |
"rhs": [ | |
10000, | |
10 | |
] | |
}, | |
"max_storage_mem_alloc_cpu/0": 200.0 | |
}, | |
{ | |
"avg_time_forward_broadcast_logical_xor": 0.0521, | |
"inputs": { | |
"lhs": [ | |
10000, | |
1 | |
], | |
"rhs": [ | |
10000, | |
1 | |
] | |
}, | |
"max_storage_mem_alloc_cpu/0": 20.0 | |
} | |
], | |
"broadcast_maximum": [ | |
{ | |
"avg_time_backward_broadcast_maximum": 0.0944, | |
"avg_time_forward_broadcast_maximum": 0.073, | |
"inputs": { | |
"lhs": [ | |
1024, | |
1024 | |
], | |
"rhs": [ | |
1024, | |
1024 | |
] | |
}, | |
"max_storage_mem_alloc_cpu/0": 2097.1521 | |
}, | |
{ | |
"avg_time_backward_broadcast_maximum": 0.0502, | |
"avg_time_forward_broadcast_maximum": 0.04, | |
"inputs": { | |
"lhs": [ | |
10000, | |
10 | |
], | |
"rhs": [ | |
10000, | |
10 | |
] | |
}, | |
"max_storage_mem_alloc_cpu/0": 200.0 | |
}, | |
{ | |
"avg_time_backward_broadcast_maximum": 0.0508, | |
"avg_time_forward_broadcast_maximum": 0.0361, | |
"inputs": { | |
"lhs": [ | |
10000, | |
1 | |
], | |
"rhs": [ | |
10000, | |
1 | |
] | |
}, | |
"max_storage_mem_alloc_cpu/0": 20.0 | |
} | |
], | |
"broadcast_minimum": [ | |
{ | |
"avg_time_backward_broadcast_minimum": 0.096, | |
"avg_time_forward_broadcast_minimum": 0.0723, | |
"inputs": { | |
"lhs": [ | |
1024, | |
1024 | |
], | |
"rhs": [ | |
1024, | |
1024 | |
] | |
}, | |
"max_storage_mem_alloc_cpu/0": 2097.1521 | |
}, | |
{ | |
"avg_time_backward_broadcast_minimum": 0.0509, | |
"avg_time_forward_broadcast_minimum": 0.038, | |
"inputs": { | |
"lhs": [ | |
10000, | |
10 | |
], | |
"rhs": [ | |
10000, | |
10 | |
] | |
}, | |
"max_storage_mem_alloc_cpu/0": 200.0 | |
}, | |
{ | |
"avg_time_backward_broadcast_minimum": 0.0512, | |
"avg_time_forward_broadcast_minimum": 0.0358, | |
"inputs": { | |
"lhs": [ | |
10000, | |
1 | |
], | |
"rhs": [ | |
10000, | |
1 | |
] | |
}, | |
"max_storage_mem_alloc_cpu/0": 20.0 | |
} | |
], | |
"broadcast_minus": [ | |
{ | |
"avg_time_forward_broadcast_minus": 0.0815, | |
"inputs": { | |
"lhs": [ | |
1024, | |
1024 | |
], | |
"rhs": [ | |
1024, | |
1024 | |
] | |
}, | |
"max_storage_mem_alloc_cpu/0": 2097.1521 | |
}, | |
{ | |
"avg_time_forward_broadcast_minus": 0.035, | |
"inputs": { | |
"lhs": [ | |
10000, | |
10 | |
], | |
"rhs": [ | |
10000, | |
10 | |
] | |
}, | |
"max_storage_mem_alloc_cpu/0": 200.0 | |
}, | |
{ | |
"avg_time_forward_broadcast_minus": 0.0321, | |
"inputs": { | |
"lhs": [ | |
10000, | |
1 | |
], | |
"rhs": [ | |
10000, | |
1 | |
] | |
}, | |
"max_storage_mem_alloc_cpu/0": 40.0 | |
} | |
], | |
"broadcast_mod": [ | |
{ | |
"avg_time_backward_broadcast_mod": 0.2037, | |
"avg_time_forward_broadcast_mod": 1.9487, | |
"inputs": { | |
"lhs": [ | |
1024, | |
1024 | |
], | |
"rhs": [ | |
1024, | |
1024 | |
] | |
}, | |
"max_storage_mem_alloc_cpu/0": 2097.1521 | |
}, | |
{ | |
"avg_time_backward_broadcast_mod": 0.0491, | |
"avg_time_forward_broadcast_mod": 0.2099, | |
"inputs": { | |
"lhs": [ | |
10000, | |
10 | |
], | |
"rhs": [ | |
10000, | |
10 | |
] | |
}, | |
"max_storage_mem_alloc_cpu/0": 200.0 | |
}, | |
{ | |
"avg_time_backward_broadcast_mod": 0.0531, | |
"avg_time_forward_broadcast_mod": 0.037, | |
"inputs": { | |
"lhs": [ | |
10000, | |
1 | |
], | |
"rhs": [ | |
10000, | |
1 | |
] | |
}, | |
"max_storage_mem_alloc_cpu/0": 20.0 | |
} | |
], | |
"broadcast_mul": [ | |
{ | |
"avg_time_backward_broadcast_mul": 0.1078, | |
"avg_time_forward_broadcast_mul": 0.0654, | |
"inputs": { | |
"lhs": [ | |
1024, | |
1024 | |
], | |
"rhs": [ | |
1024, | |
1024 | |
] | |
}, | |
"max_storage_mem_alloc_cpu/0": 2097.1521 | |
}, | |
{ | |
"avg_time_backward_broadcast_mul": 0.0453, | |
"avg_time_forward_broadcast_mul": 0.0327, | |
"inputs": { | |
"lhs": [ | |
10000, | |
10 | |
], | |
"rhs": [ | |
10000, | |
10 | |
] | |
}, | |
"max_storage_mem_alloc_cpu/0": 200.0 | |
}, | |
{ | |
"avg_time_backward_broadcast_mul": 0.0308, | |
"avg_time_forward_broadcast_mul": 0.023, | |
"inputs": { | |
"lhs": [ | |
10000, | |
1 | |
], | |
"rhs": [ | |
10000, | |
1 | |
] | |
}, | |
"max_storage_mem_alloc_cpu/0": 20.0 | |
} | |
], | |
"broadcast_not_equal": [ | |
{ | |
"avg_time_forward_broadcast_not_equal": 0.2416, | |
"inputs": { | |
"lhs": [ | |
1024, | |
1024 | |
], | |
"rhs": [ | |
1024, | |
1024 | |
] | |
}, | |
"max_storage_mem_alloc_cpu/0": 4194.3042 | |
}, | |
{ | |
"avg_time_forward_broadcast_not_equal": 0.0348, | |
"inputs": { | |
"lhs": [ | |
10000, | |
10 | |
], | |
"rhs": [ | |
10000, | |
10 | |
] | |
}, | |
"max_storage_mem_alloc_cpu/0": 400.0 | |
}, | |
{ | |
"avg_time_forward_broadcast_not_equal": 0.0367, | |
"inputs": { | |
"lhs": [ | |
10000, | |
1 | |
], | |
"rhs": [ | |
10000, | |
1 | |
] | |
}, | |
"max_storage_mem_alloc_cpu/0": 40.0 | |
} | |
], | |
"broadcast_plus": [ | |
{ | |
"avg_time_forward_broadcast_plus": 0.1386, | |
"inputs": { | |
"lhs": [ | |
1024, | |
1024 | |
], | |
"rhs": [ | |
1024, | |
1024 | |
] | |
}, | |
"max_storage_mem_alloc_cpu/0": 4194.3042 | |
}, | |
{ | |
"avg_time_forward_broadcast_plus": 0.0335, | |
"inputs": { | |
"lhs": [ | |
10000, | |
10 | |
], | |
"rhs": [ | |
10000, | |
10 | |
] | |
}, | |
"max_storage_mem_alloc_cpu/0": 400.0 | |
}, | |
{ | |
"avg_time_forward_broadcast_plus": 0.0281, | |
"inputs": { | |
"lhs": [ | |
10000, | |
1 | |
], | |
"rhs": [ | |
10000, | |
1 | |
] | |
}, | |
"max_storage_mem_alloc_cpu/0": 40.0 | |
} | |
], | |
"broadcast_power": [ | |
{ | |
"avg_time_backward_broadcast_power": 4.6383, | |
"avg_time_forward_broadcast_power": 2.0918, | |
"inputs": { | |
"lhs": [ | |
1024, | |
1024 | |
], | |
"rhs": [ | |
1024, | |
1024 | |
] | |
}, | |
"max_storage_mem_alloc_cpu/0": 2097.1521 | |
}, | |
{ | |
"avg_time_backward_broadcast_power": 0.4157, | |
"avg_time_forward_broadcast_power": 0.2459, | |
"inputs": { | |
"lhs": [ | |
10000, | |
10 | |
], | |
"rhs": [ | |
10000, | |
10 | |
] | |
}, | |
"max_storage_mem_alloc_cpu/0": 200.0 | |
}, | |
{ | |
"avg_time_backward_broadcast_power": 0.0864, | |
"avg_time_forward_broadcast_power": 0.0535, | |
"inputs": { | |
"lhs": [ | |
10000, | |
1 | |
], | |
"rhs": [ | |
10000, | |
1 | |
] | |
}, | |
"max_storage_mem_alloc_cpu/0": 20.0 | |
} | |
], | |
"broadcast_sub": [ | |
{ | |
"avg_time_backward_broadcast_sub": 0.0788, | |
"avg_time_forward_broadcast_sub": 0.0723, | |
"inputs": { | |
"lhs": [ | |
1024, | |
1024 | |
], | |
"rhs": [ | |
1024, | |
1024 | |
] | |
}, | |
"max_storage_mem_alloc_cpu/0": 4194.3042 | |
}, | |
{ | |
"avg_time_backward_broadcast_sub": 0.0481, | |
"avg_time_forward_broadcast_sub": 0.0367, | |
"inputs": { | |
"lhs": [ | |
10000, | |
10 | |
], | |
"rhs": [ | |
10000, | |
10 | |
] | |
}, | |
"max_storage_mem_alloc_cpu/0": 200.0 | |
}, | |
{ | |
"avg_time_backward_broadcast_sub": 0.0349, | |
"avg_time_forward_broadcast_sub": 0.0312, | |
"inputs": { | |
"lhs": [ | |
10000, | |
1 | |
], | |
"rhs": [ | |
10000, | |
1 | |
] | |
}, | |
"max_storage_mem_alloc_cpu/0": 40.0 | |
} | |
], | |
"cbrt": [ | |
{ | |
"avg_time_backward_cbrt": 0.0799, | |
"avg_time_forward_cbrt": 1.4149, | |
"inputs": { | |
"data": [ | |
1024, | |
1024 | |
] | |
}, | |
"max_storage_mem_alloc_cpu/0": 2097.1521 | |
}, | |
{ | |
"avg_time_backward_cbrt": 0.0366, | |
"avg_time_forward_cbrt": 0.0441, | |
"inputs": { | |
"data": [ | |
10000, | |
1 | |
] | |
}, | |
"max_storage_mem_alloc_cpu/0": 20.0 | |
}, | |
{ | |
"avg_time_backward_cbrt": 0.0737, | |
"avg_time_forward_cbrt": 0.9382, | |
"inputs": { | |
"data": [ | |
10000, | |
100 | |
] | |
}, | |
"max_storage_mem_alloc_cpu/0": 2000.0 | |
} | |
], | |
"ceil": [ | |
{ | |
"avg_time_forward_ceil": 0.123, | |
"inputs": { | |
"data": [ | |
1024, | |
1024 | |
] | |
}, | |
"max_storage_mem_alloc_cpu/0": 2097.1521 | |
}, | |
{ | |
"avg_time_forward_ceil": 0.0486, | |
"inputs": { | |
"data": [ | |
10000, | |
1 | |
] | |
}, | |
"max_storage_mem_alloc_cpu/0": 20.0 | |
}, | |
{ | |
"avg_time_forward_ceil": 0.1207, | |
"inputs": { | |
"data": [ | |
10000, | |
100 | |
] | |
}, | |
"max_storage_mem_alloc_cpu/0": 2000.0 | |
} | |
], | |
"cos": [ | |
{ | |
"avg_time_backward_cos": 0.3484, | |
"avg_time_forward_cos": 0.3891, | |
"inputs": { | |
"data": [ | |
1024, | |
1024 | |
] | |
}, | |
"max_storage_mem_alloc_cpu/0": 2097.1521 | |
}, | |
{ | |
"avg_time_backward_cos": 0.0399, | |
"avg_time_forward_cos": 0.0353, | |
"inputs": { | |
"data": [ | |
10000, | |
1 | |
] | |
}, | |
"max_storage_mem_alloc_cpu/0": 20.0 | |
}, | |
{ | |
"avg_time_backward_cos": 0.3393, | |
"avg_time_forward_cos": 0.3735, | |
"inputs": { | |
"data": [ | |
10000, | |
100 | |
] | |
}, | |
"max_storage_mem_alloc_cpu/0": 2000.0 | |
} | |
], | |
"cosh": [ | |
{ | |
"avg_time_backward_cosh": 1.0255, | |
"avg_time_forward_cosh": 0.6308, | |
"inputs": { | |
"data": [ | |
1024, | |
1024 | |
] | |
}, | |
"max_storage_mem_alloc_cpu/0": 2097.1521 | |
}, | |
{ | |
"avg_time_backward_cosh": 0.0493, | |
"avg_time_forward_cosh": 0.045, | |
"inputs": { | |
"data": [ | |
10000, | |
1 | |
] | |
}, | |
"max_storage_mem_alloc_cpu/0": 20.0 | |
}, | |
{ | |
"avg_time_backward_cosh": 1.3662, | |
"avg_time_forward_cosh": 0.7747, | |
"inputs": { | |
"data": [ | |
10000, | |
100 | |
] | |
}, | |
"max_storage_mem_alloc_cpu/0": 2000.0 | |
} | |
], | |
"degrees": [ | |
{ | |
"avg_time_backward_degrees": 0.0412, | |
"avg_time_forward_degrees": 0.0528, | |
"inputs": { | |
"data": [ | |
1024, | |
1024 | |
] | |
}, | |
"max_storage_mem_alloc_cpu/0": 2097.1521 | |
}, | |
{ | |
"avg_time_backward_degrees": 0.0279, | |
"avg_time_forward_degrees": 0.0308, | |
"inputs": { | |
"data": [ | |
10000, | |
1 | |
] | |
}, | |
"max_storage_mem_alloc_cpu/0": 20.0 | |
}, | |
{ | |
"avg_time_backward_degrees": 0.0405, | |
"avg_time_forward_degrees": 0.0477, | |
"inputs": { | |
"data": [ | |
10000, | |
100 | |
] | |
}, | |
"max_storage_mem_alloc_cpu/0": 2000.0 | |
} | |
], | |
"depth_to_space": [ | |
{ | |
"avg_time_forward_depth_to_space": 0.067, | |
"inputs": { | |
"block_size": 2, | |
"data": [ | |
1, | |
4, | |
2, | |
4 | |
] | |
}, | |
"max_storage_mem_alloc_cpu/0": 0.064 | |
}, | |
{ | |
"avg_time_forward_depth_to_space": 0.7047, | |
"inputs": { | |
"block_size": 5, | |
"data": [ | |
10, | |
25, | |
10, | |
100 | |
] | |
}, | |
"max_storage_mem_alloc_cpu/0": 500.0 | |
} | |
], | |
"dot": [ | |
{ | |
"avg_time_backward_dot": 3.7072, | |
"avg_time_forward_dot": 2.1858, | |
"inputs": { | |
"lhs": [ | |
1024, | |
1024 | |
], | |
"rhs": [ | |
1024, | |
1024 | |
] | |
}, | |
"max_storage_mem_alloc_cpu/0": 2097.1521 | |
}, | |
{ | |
"avg_time_backward_dot": 3.6739, | |
"avg_time_forward_dot": 0.4115, | |
"inputs": { | |
"lhs": [ | |
1000, | |
10 | |
], | |
"rhs": [ | |
1000, | |
10 | |
], | |
"transpose_b": true | |
}, | |
"max_storage_mem_alloc_cpu/0": 2000.0 | |
}, | |
{ | |
"avg_time_backward_dot": 0.2148, | |
"avg_time_forward_dot": 0.1696, | |
"inputs": { | |
"lhs": [ | |
1000, | |
1 | |
], | |
"rhs": [ | |
100, | |
1000 | |
], | |
"transpose_a": true, | |
"transpose_b": true | |
}, | |
"max_storage_mem_alloc_cpu/0": 0.2 | |
} | |
], | |
"elemwise_add": [ | |
{ | |
"avg_time_forward_elemwise_add": 0.0937, | |
"inputs": { | |
"lhs": [ | |
1024, | |
1024 | |
], | |
"rhs": [ | |
1024, | |
1024 | |
] | |
}, | |
"max_storage_mem_alloc_cpu/0": 2097.1521 | |
}, | |
{ | |
"avg_time_forward_elemwise_add": 0.0563, | |
"inputs": { | |
"lhs": [ | |
10000, | |
10 | |
], | |
"rhs": [ | |
10000, | |
10 | |
] | |
}, | |
"max_storage_mem_alloc_cpu/0": 200.0 | |
}, | |
{ | |
"avg_time_forward_elemwise_add": 0.055, | |
"inputs": { | |
"lhs": [ | |
10000, | |
1 | |
], | |
"rhs": [ | |
10000, | |
1 | |
] | |
}, | |
"max_storage_mem_alloc_cpu/0": 40.0 | |
} | |
], | |
"elemwise_div": [ | |
{ | |
"avg_time_forward_elemwise_div": 0.077, | |
"inputs": { | |
"lhs": [ | |
1024, | |
1024 | |
], | |
"rhs": [ | |
1024, | |
1024 | |
] | |
}, | |
"max_storage_mem_alloc_cpu/0": 2097.1521 | |
}, | |
{ | |
"avg_time_forward_elemwise_div": 0.0383, | |
"inputs": { | |
"lhs": [ | |
10000, | |
10 | |
], | |
"rhs": [ | |
10000, | |
10 | |
] | |
}, | |
"max_storage_mem_alloc_cpu/0": 200.0 | |
}, | |
{ | |
"avg_time_forward_elemwise_div": 0.032, | |
"inputs": { | |
"lhs": [ | |
10000, | |
1 | |
], | |
"rhs": [ | |
10000, | |
1 | |
] | |
}, | |
"max_storage_mem_alloc_cpu/0": 20.0 | |
} | |
], | |
"elemwise_mul": [ | |
{ | |
"avg_time_forward_elemwise_mul": 0.0772, | |
"inputs": { | |
"lhs": [ | |
1024, | |
1024 | |
], | |
"rhs": [ | |
1024, | |
1024 | |
] | |
}, | |
"max_storage_mem_alloc_cpu/0": 2097.1521 | |
}, | |
{ | |
"avg_time_forward_elemwise_mul": 0.0394, | |
"inputs": { | |
"lhs": [ | |
10000, | |
10 | |
], | |
"rhs": [ | |
10000, | |
10 | |
] | |
}, | |
"max_storage_mem_alloc_cpu/0": 200.0 | |
}, | |
{ | |
"avg_time_forward_elemwise_mul": 0.0307, | |
"inputs": { | |
"lhs": [ | |
10000, | |
1 | |
], | |
"rhs": [ | |
10000, | |
1 | |
] | |
}, | |
"max_storage_mem_alloc_cpu/0": 20.0 | |
} | |
], | |
"elemwise_sub": [ | |
{ | |
"avg_time_forward_elemwise_sub": 0.2107, | |
"inputs": { | |
"lhs": [ | |
1024, | |
1024 | |
], | |
"rhs": [ | |
1024, | |
1024 | |
] | |
}, | |
"max_storage_mem_alloc_cpu/0": 4194.3042 | |
}, | |
{ | |
"avg_time_forward_elemwise_sub": 0.043, | |
"inputs": { | |
"lhs": [ | |
10000, | |
10 | |
], | |
"rhs": [ | |
10000, | |
10 | |
] | |
}, | |
"max_storage_mem_alloc_cpu/0": 400.0 | |
}, | |
{ | |
"avg_time_forward_elemwise_sub": 0.0337, | |
"inputs": { | |
"lhs": [ | |
10000, | |
1 | |
], | |
"rhs": [ | |
10000, | |
1 | |
] | |
}, | |
"max_storage_mem_alloc_cpu/0": 40.0 | |
} | |
], | |
"erf": [ | |
{ | |
"avg_time_backward_erf": 0.7827, | |
"avg_time_forward_erf": 0.9797, | |
"inputs": { | |
"data": [ | |
1024, | |
1024 | |
] | |
}, | |
"max_storage_mem_alloc_cpu/0": 2097.1521 | |
}, | |
{ | |
"avg_time_backward_erf": 0.0358, | |
"avg_time_forward_erf": 0.0389, | |
"inputs": { | |
"data": [ | |
10000, | |
1 | |
] | |
}, | |
"max_storage_mem_alloc_cpu/0": 20.0 | |
}, | |
{ | |
"avg_time_backward_erf": 0.7422, | |
"avg_time_forward_erf": 0.9054, | |
"inputs": { | |
"data": [ | |
10000, | |
100 | |
] | |
}, | |
"max_storage_mem_alloc_cpu/0": 2000.0 | |
} | |
], | |
"erfinv": [ | |
{ | |
"avg_time_backward_erfinv": 0.7646, | |
"avg_time_forward_erfinv": 3.3437, | |
"inputs": { | |
"data": [ | |
1024, | |
1024 | |
] | |
}, | |
"max_storage_mem_alloc_cpu/0": 2097.1521 | |
}, | |
{ | |
"avg_time_backward_erfinv": 0.0356, | |
"avg_time_forward_erfinv": 0.0674, | |
"inputs": { | |
"data": [ | |
10000, | |
1 | |
] | |
}, | |
"max_storage_mem_alloc_cpu/0": 20.0 | |
}, | |
{ | |
"avg_time_backward_erfinv": 0.674, | |
"avg_time_forward_erfinv": 3.1553, | |
"inputs": { | |
"data": [ | |
10000, | |
100 | |
] | |
}, | |
"max_storage_mem_alloc_cpu/0": 2000.0 | |
} | |
], | |
"exp": [ | |
{ | |
"avg_time_forward_exp": 0.6285, | |
"inputs": { | |
"data": [ | |
1024, | |
1024 | |
] | |
}, | |
"max_storage_mem_alloc_cpu/0": 4194.3042 | |
}, | |
{ | |
"avg_time_forward_exp": 0.034, | |
"inputs": { | |
"data": [ | |
10000, | |
1 | |
] | |
}, | |
"max_storage_mem_alloc_cpu/0": 40.0 | |
}, | |
{ | |
"avg_time_forward_exp": 0.5875, | |
"inputs": { | |
"data": [ | |
10000, | |
100 | |
] | |
}, | |
"max_storage_mem_alloc_cpu/0": 4000.0 | |
} | |
], | |
"expm1": [ | |
{ | |
"avg_time_backward_expm1": 0.6689, | |
"avg_time_forward_expm1": 1.074, | |
"inputs": { | |
"data": [ | |
1024, | |
1024 | |
] | |
}, | |
"max_storage_mem_alloc_cpu/0": 4194.3042 | |
}, | |
{ | |
"avg_time_backward_expm1": 0.0325, | |
"avg_time_forward_expm1": 0.052, | |
"inputs": { | |
"data": [ | |
10000, | |
1 | |
] | |
}, | |
"max_storage_mem_alloc_cpu/0": 40.0 | |
}, | |
{ | |
"avg_time_backward_expm1": 0.6471, | |
"avg_time_forward_expm1": 1.0829, | |
"inputs": { | |
"data": [ | |
10000, | |
100 | |
] | |
}, | |
"max_storage_mem_alloc_cpu/0": 2000.0 | |
} | |
], | |
"fix": [ | |
{ | |
"avg_time_forward_fix": 0.3644, | |
"inputs": { | |
"data": [ | |
1024, | |
1024 | |
] | |
}, | |
"max_storage_mem_alloc_cpu/0": 2097.1521 | |
}, | |
{ | |
"avg_time_forward_fix": 0.0321, | |
"inputs": { | |
"data": [ | |
10000, | |
1 | |
] | |
}, | |
"max_storage_mem_alloc_cpu/0": 20.0 | |
}, | |
{ | |
"avg_time_forward_fix": 0.369, | |
"inputs": { | |
"data": [ | |
10000, | |
100 | |
] | |
}, | |
"max_storage_mem_alloc_cpu/0": 2000.0 | |
} | |
], | |
"flatten": [ | |
{ | |
"avg_time_forward_flatten": 0.0565, | |
"inputs": { | |
"data": [ | |
1024, | |
1024 | |
] | |
}, | |
"max_storage_mem_alloc_cpu/0": 4194.3042 | |
}, | |
{ | |
"avg_time_forward_flatten": 0.0441, | |
"inputs": { | |
"data": [ | |
10000, | |
1 | |
] | |
}, | |
"max_storage_mem_alloc_cpu/0": 20.0 | |
}, | |
{ | |
"avg_time_forward_flatten": 0.0564, | |
"inputs": { | |
"data": [ | |
10000, | |
100 | |
] | |
}, | |
"max_storage_mem_alloc_cpu/0": 4000.0 | |
} | |
], | |
"flip": [ | |
{ | |
"avg_time_forward_flip": 0.4088, | |
"inputs": { | |
"axis": 0, | |
"data": [ | |
1024, | |
1024 | |
] | |
}, | |
"max_storage_mem_alloc_cpu/0": 2097.1521 | |
}, | |
{ | |
"avg_time_forward_flip": 0.0356, | |
"inputs": { | |
"axis": 0, | |
"data": [ | |
10000, | |
1 | |
] | |
}, | |
"max_storage_mem_alloc_cpu/0": 20.0 | |
}, | |
{ | |
"avg_time_forward_flip": 0.2905, | |
"inputs": { | |
"axis": 0, | |
"data": [ | |
10000, | |
100 | |
] | |
}, | |
"max_storage_mem_alloc_cpu/0": 2000.0 | |
} | |
], | |
"floor": [ | |
{ | |
"avg_time_forward_floor": 0.2057, | |
"inputs": { | |
"data": [ | |
1024, | |
1024 | |
] | |
}, | |
"max_storage_mem_alloc_cpu/0": 2097.1521 | |
}, | |
{ | |
"avg_time_forward_floor": 0.0706, | |
"inputs": { | |
"data": [ | |
10000, | |
1 | |
] | |
}, | |
"max_storage_mem_alloc_cpu/0": 20.0 | |
}, | |
{ | |
"avg_time_forward_floor": 0.1519, | |
"inputs": { | |
"data": [ | |
10000, | |
100 | |
] | |
}, | |
"max_storage_mem_alloc_cpu/0": 2000.0 | |
} | |
], | |
"ftml_update": [ | |
{ | |
"avg_time_forward_ftml_update": 1.6824, | |
"inputs": { | |
"beta1": 0.1, | |
"beta2": 0.1, | |
"clip_grad": -1.0, | |
"d": [ | |
1024, | |
1024 | |
], | |
"epsilon": 1e-08, | |
"grad": [ | |
1024, | |
1024 | |
], | |
"lr": 0.1, | |
"rescale_grad": 0.4, | |
"t": 1, | |
"v": [ | |
1024, | |
1024 | |
], | |
"weight": [ | |
1024, | |
1024 | |
], | |
"z": [ | |
1024, | |
1024 | |
] | |
}, | |
"max_storage_mem_alloc_cpu/0": 2622552.75 | |
}, | |
{ | |
"avg_time_forward_ftml_update": 0.0565, | |
"inputs": { | |
"beta1": 0.5, | |
"beta2": 0.5, | |
"clip_grad": -1.0, | |
"d": [ | |
10000, | |
1 | |
], | |
"epsilon": 1e-08, | |
"grad": [ | |
10000, | |
1 | |
], | |
"lr": 0.5, | |
"rescale_grad": 0.4, | |
"t": 1, | |
"v": [ | |
10000, | |
1 | |
], | |
"weight": [ | |
10000, | |
1 | |
], | |
"z": [ | |
10000, | |
1 | |
] | |
}, | |
"max_storage_mem_alloc_cpu/0": 2622592.75 | |
}, | |
{ | |
"avg_time_forward_ftml_update": 1.7309, | |
"inputs": { | |
"beta1": 0.9, | |
"beta2": 0.9, | |
"clip_grad": -1.0, | |
"d": [ | |
10000, | |
100 | |
], | |
"epsilon": 1e-08, | |
"grad": [ | |
10000, | |
100 | |
], | |
"lr": 0.9, | |
"rescale_grad": 0.4, | |
"t": 1, | |
"v": [ | |
10000, | |
100 | |
], | |
"weight": [ | |
10000, | |
100 | |
], | |
"z": [ | |
10000, | |
100 | |
] | |
}, | |
"max_storage_mem_alloc_cpu/0": 2626572.75 | |
} | |
], | |
"ftrl_update": [ | |
{ | |
"avg_time_forward_ftrl_update": 1.3501, | |
"inputs": { | |
"grad": [ | |
1024, | |
1024 | |
], | |
"lr": 0.1, | |
"n": [ | |
1024, | |
1024 | |
], | |
"rescale_grad": 0.4, | |
"weight": [ | |
1024, | |
1024 | |
], | |
"z": [ | |
1024, | |
1024 | |
] | |
}, | |
"max_storage_mem_alloc_cpu/0": 2628767.0 | |
}, | |
{ | |
"avg_time_forward_ftrl_update": 0.0465, | |
"inputs": { | |
"grad": [ | |
10000, | |
1 | |
], | |
"lr": 0.5, | |
"n": [ | |
10000, | |
1 | |
], | |
"rescale_grad": 0.4, | |
"weight": [ | |
10000, | |
1 | |
], | |
"z": [ | |
10000, | |
1 | |
] | |
}, | |
"max_storage_mem_alloc_cpu/0": 2628807.0 | |
}, | |
{ | |
"avg_time_forward_ftrl_update": 1.2321, | |
"inputs": { | |
"grad": [ | |
10000, | |
100 | |
], | |
"lr": 0.9, | |
"n": [ | |
10000, | |
100 | |
], | |
"rescale_grad": 0.4, | |
"weight": [ | |
10000, | |
100 | |
], | |
"z": [ | |
10000, | |
100 | |
] | |
}, | |
"max_storage_mem_alloc_cpu/0": 2632787.0 | |
} | |
], | |
"gamma": [ | |
{ | |
"avg_time_backward_gamma": 4.3353, | |
"avg_time_forward_gamma": 2.5225, | |
"inputs": { | |
"data": [ | |
1024, | |
1024 | |
] | |
}, | |
"max_storage_mem_alloc_cpu/0": 2097.1521 | |
}, | |
{ | |
"avg_time_backward_gamma": 0.0709, | |
"avg_time_forward_gamma": 0.0606, | |
"inputs": { | |
"data": [ | |
10000, | |
1 | |
] | |
}, | |
"max_storage_mem_alloc_cpu/0": 20.0 | |
}, | |
{ | |
"avg_time_backward_gamma": 4.1527, | |
"avg_time_forward_gamma": 2.4293, | |
"inputs": { | |
"data": [ | |
10000, | |
100 | |
] | |
}, | |
"max_storage_mem_alloc_cpu/0": 2000.0 | |
} | |
], | |
"gammaln": [ | |
{ | |
"avg_time_backward_gammaln": 2.5238, | |
"avg_time_forward_gammaln": 24.3514, | |
"inputs": { | |
"data": [ | |
1024, | |
1024 | |
] | |
}, | |
"max_storage_mem_alloc_cpu/0": 4194.3042 | |
}, | |
{ | |
"avg_time_backward_gammaln": 0.0525, | |
"avg_time_forward_gammaln": 0.2523, | |
"inputs": { | |
"data": [ | |
10000, | |
1 | |
] | |
}, | |
"max_storage_mem_alloc_cpu/0": 20.0 | |
}, | |
{ | |
"avg_time_backward_gammaln": 3.0252, | |
"avg_time_forward_gammaln": 23.2531, | |
"inputs": { | |
"data": [ | |
10000, | |
100 | |
] | |
}, | |
"max_storage_mem_alloc_cpu/0": 2000.0 | |
} | |
], | |
"hard_sigmoid": [ | |
{ | |
"avg_time_backward_hard_sigmoid": 0.1363, | |
"avg_time_forward_hard_sigmoid": 0.1193, | |
"inputs": { | |
"alpha": 0.25, | |
"beta": 0.5, | |
"data": [ | |
1024, | |
1024 | |
] | |
}, | |
"max_storage_mem_alloc_cpu/0": 4194.3042 | |
}, | |
{ | |
"avg_time_backward_hard_sigmoid": 0.0315, | |
"avg_time_forward_hard_sigmoid": 0.0361, | |
"inputs": { | |
"alpha": 0.25, | |
"beta": 0.5, | |
"data": [ | |
10000, | |
1 | |
] | |
}, | |
"max_storage_mem_alloc_cpu/0": 40.0 | |
}, | |
{ | |
"avg_time_backward_hard_sigmoid": 0.1283, | |
"avg_time_forward_hard_sigmoid": 0.1113, | |
"inputs": { | |
"alpha": 0.25, | |
"beta": 0.5, | |
"data": [ | |
10000, | |
100 | |
] | |
}, | |
"max_storage_mem_alloc_cpu/0": 2000.0 | |
} | |
], | |
"identity": [ | |
{ | |
"inputs": { | |
"data": [ | |
1024, | |
1024 | |
] | |
}, | |
"max_storage_mem_alloc_cpu/0": 4194.3042 | |
}, | |
{ | |
"inputs": { | |
"data": [ | |
10000, | |
1 | |
] | |
}, | |
"max_storage_mem_alloc_cpu/0": 40.0 | |
}, | |
{ | |
"inputs": { | |
"data": [ | |
10000, | |
100 | |
] | |
}, | |
"max_storage_mem_alloc_cpu/0": 4000.0 | |
} | |
], | |
"log": [ | |
{ | |
"avg_time_backward_log": 0.088, | |
"avg_time_forward_log": 0.8636, | |
"inputs": { | |
"data": [ | |
1024, | |
1024 | |
] | |
}, | |
"max_storage_mem_alloc_cpu/0": 4194.3042 | |
}, | |
{ | |
"avg_time_backward_log": 0.0305, | |
"avg_time_forward_log": 0.0429, | |
"inputs": { | |
"data": [ | |
10000, | |
1 | |
] | |
}, | |
"max_storage_mem_alloc_cpu/0": 40.0 | |
}, | |
{ | |
"avg_time_backward_log": 0.0903, | |
"avg_time_forward_log": 0.8536, | |
"inputs": { | |
"data": [ | |
10000, | |
100 | |
] | |
}, | |
"max_storage_mem_alloc_cpu/0": 4000.0 | |
} | |
], | |
"log10": [ | |
{ | |
"avg_time_backward_log10": 0.087, | |
"avg_time_forward_log10": 1.0409, | |
"inputs": { | |
"data": [ | |
1024, | |
1024 | |
] | |
}, | |
"max_storage_mem_alloc_cpu/0": 4194.3042 | |
}, | |
{ | |
"avg_time_backward_log10": 0.0301, | |
"avg_time_forward_log10": 0.0433, | |
"inputs": { | |
"data": [ | |
10000, | |
1 | |
] | |
}, | |
"max_storage_mem_alloc_cpu/0": 40.0 | |
}, | |
{ | |
"avg_time_backward_log10": 0.0821, | |
"avg_time_forward_log10": 1.0784, | |
"inputs": { | |
"data": [ | |
10000, | |
100 | |
] | |
}, | |
"max_storage_mem_alloc_cpu/0": 2000.0 | |
} | |
], | |
"log1p": [ | |
{ | |
"avg_time_backward_log1p": 0.0844, | |
"avg_time_forward_log1p": 1.1085, | |
"inputs": { | |
"data": [ | |
1024, | |
1024 | |
] | |
}, | |
"max_storage_mem_alloc_cpu/0": 4194.3042 | |
}, | |
{ | |
"avg_time_backward_log1p": 0.0308, | |
"avg_time_forward_log1p": 0.0575, | |
"inputs": { | |
"data": [ | |
10000, | |
1 | |
] | |
}, | |
"max_storage_mem_alloc_cpu/0": 20.0 | |
}, | |
{ | |
"avg_time_backward_log1p": 0.08, | |
"avg_time_forward_log1p": 1.0414, | |
"inputs": { | |
"data": [ | |
10000, | |
100 | |
] | |
}, | |
"max_storage_mem_alloc_cpu/0": 2000.0 | |
} | |
], | |
"log2": [ | |
{ | |
"avg_time_backward_log2": 0.0683, | |
"avg_time_forward_log2": 0.8256, | |
"inputs": { | |
"data": [ | |
1024, | |
1024 | |
] | |
}, | |
"max_storage_mem_alloc_cpu/0": 2097.1521 | |
}, | |
{ | |
"avg_time_backward_log2": 0.0313, | |
"avg_time_forward_log2": 0.0392, | |
"inputs": { | |
"data": [ | |
10000, | |
1 | |
] | |
}, | |
"max_storage_mem_alloc_cpu/0": 40.0 | |
}, | |
{ | |
"avg_time_backward_log2": 0.0656, | |
"avg_time_forward_log2": 0.7936, | |
"inputs": { | |
"data": [ | |
10000, | |
100 | |
] | |
}, | |
"max_storage_mem_alloc_cpu/0": 2000.0 | |
} | |
], | |
"log_softmax": [ | |
{ | |
"avg_time_backward_log_softmax": 0.8013, | |
"avg_time_forward_log_softmax": 1.6769, | |
"inputs": { | |
"axis": -1, | |
"data": [ | |
1024, | |
1024 | |
], | |
"temperature": 0.5 | |
}, | |
"max_storage_mem_alloc_cpu/0": 4194.3042 | |
}, | |
{ | |
"avg_time_backward_log_softmax": 0.0393, | |
"avg_time_forward_log_softmax": 0.0482, | |
"inputs": { | |
"axis": -1, | |
"data": [ | |
10000, | |
1 | |
], | |
"temperature": 0.5 | |
}, | |
"max_storage_mem_alloc_cpu/0": 20.0 | |
}, | |
{ | |
"avg_time_backward_log_softmax": 0.7768, | |
"avg_time_forward_log_softmax": 1.6099, | |
"inputs": { | |
"axis": -1, | |
"data": [ | |
10000, | |
100 | |
], | |
"temperature": 0.5 | |
}, | |
"max_storage_mem_alloc_cpu/0": 4000.0 | |
} | |
], | |
"logical_not": [ | |
{ | |
"avg_time_forward_logical_not": 0.0535, | |
"inputs": { | |
"data": [ | |
1024, | |
1024 | |
] | |
}, | |
"max_storage_mem_alloc_cpu/0": 2097.1521 | |
}, | |
{ | |
"avg_time_forward_logical_not": 0.0241, | |
"inputs": { | |
"data": [ | |
10000, | |
1 | |
] | |
}, | |
"max_storage_mem_alloc_cpu/0": 20.0 | |
}, | |
{ | |
"avg_time_forward_logical_not": 0.055, | |
"inputs": { | |
"data": [ | |
10000, | |
100 | |
] | |
}, | |
"max_storage_mem_alloc_cpu/0": 4000.0 | |
} | |
], | |
"make_loss": [ | |
{ | |
"avg_time_forward_make_loss": 0.3873, | |
"inputs": { | |
"data": [ | |
1024, | |
1024 | |
] | |
}, | |
"max_storage_mem_alloc_cpu/0": 2097.1521 | |
}, | |
{ | |
"avg_time_forward_make_loss": 0.0159, | |
"inputs": { | |
"data": [ | |
10000, | |
1 | |
] | |
}, | |
"max_storage_mem_alloc_cpu/0": 20.0 | |
}, | |
{ | |
"avg_time_forward_make_loss": 0.3598, | |
"inputs": { | |
"data": [ | |
10000, | |
100 | |
] | |
}, | |
"max_storage_mem_alloc_cpu/0": 2000.0 | |
} | |
], | |
"max": [ | |
{ | |
"avg_time_backward_max": 0.7578, | |
"avg_time_forward_max": 0.6867, | |
"inputs": { | |
"axis": 0, | |
"data": [ | |
1024, | |
1024 | |
] | |
}, | |
"max_storage_mem_alloc_cpu/0": 2.048 | |
}, | |
{ | |
"avg_time_backward_max": 0.0316, | |
"avg_time_forward_max": 0.1001, | |
"inputs": { | |
"axis": 0, | |
"data": [ | |
10000, | |
1 | |
] | |
}, | |
"max_storage_mem_alloc_cpu/0": 0.002 | |
}, | |
{ | |
"avg_time_backward_max": 0.6468, | |
"avg_time_forward_max": 0.5863, | |
"inputs": { | |
"axis": 0, | |
"data": [ | |
10000, | |
100 | |
] | |
}, | |
"max_storage_mem_alloc_cpu/0": 0.2 | |
} | |
], | |
"max_axis": [ | |
{ | |
"avg_time_forward_max_axis": 0.558, | |
"inputs": { | |
"axis": 0, | |
"data": [ | |
1024, | |
1024 | |
] | |
}, | |
"max_storage_mem_alloc_cpu/0": 2.048 | |
}, | |
{ | |
"avg_time_forward_max_axis": 0.1477, | |
"inputs": { | |
"axis": 0, | |
"data": [ | |
10000, | |
1 | |
] | |
}, | |
"max_storage_mem_alloc_cpu/0": 0.002 | |
}, | |
{ | |
"avg_time_forward_max_axis": 0.6615, | |
"inputs": { | |
"axis": 0, | |
"data": [ | |
10000, | |
100 | |
] | |
}, | |
"max_storage_mem_alloc_cpu/0": 0.2 | |
} | |
], | |
"mean": [ | |
{ | |
"avg_time_backward_mean": 1.3769, | |
"avg_time_forward_mean": 0.7853, | |
"inputs": { | |
"axis": 0, | |
"data": [ | |
1024, | |
1024 | |
] | |
}, | |
"max_storage_mem_alloc_cpu/0": 2.048 | |
}, | |
{ | |
"avg_time_backward_mean": 0.0538, | |
"avg_time_forward_mean": 0.183, | |
"inputs": { | |
"axis": 0, | |
"data": [ | |
10000, | |
1 | |
] | |
}, | |
"max_storage_mem_alloc_cpu/0": 0.002 | |
}, | |
{ | |
"avg_time_backward_mean": 1.4845, | |
"avg_time_forward_mean": 0.8149, | |
"inputs": { | |
"axis": 0, | |
"data": [ | |
10000, | |
100 | |
] | |
}, | |
"max_storage_mem_alloc_cpu/0": 0.2 | |
} | |
], | |
"min": [ | |
{ | |
"avg_time_backward_min": 0.7709, | |
"avg_time_forward_min": 0.7182, | |
"inputs": { | |
"axis": 0, | |
"data": [ | |
1024, | |
1024 | |
] | |
}, | |
"max_storage_mem_alloc_cpu/0": 4.096 | |
}, | |
{ | |
"avg_time_backward_min": 0.0395, | |
"avg_time_forward_min": 0.1402, | |
"inputs": { | |
"axis": 0, | |
"data": [ | |
10000, | |
1 | |
] | |
}, | |
"max_storage_mem_alloc_cpu/0": 0.002 | |
}, | |
{ | |
"avg_time_backward_min": 0.5194, | |
"avg_time_forward_min": 0.5003, | |
"inputs": { | |
"axis": 0, | |
"data": [ | |
10000, | |
100 | |
] | |
}, | |
"max_storage_mem_alloc_cpu/0": 0.2 | |
} | |
], | |
"min_axis": [ | |
{ | |
"avg_time_forward_min_axis": 0.7148, | |
"inputs": { | |
"axis": 0, | |
"data": [ | |
1024, | |
1024 | |
] | |
}, | |
"max_storage_mem_alloc_cpu/0": 4.096 | |
}, | |
{ | |
"avg_time_forward_min_axis": 0.1367, | |
"inputs": { | |
"axis": 0, | |
"data": [ | |
10000, | |
1 | |
] | |
}, | |
"max_storage_mem_alloc_cpu/0": 0.004 | |
}, | |
{ | |
"avg_time_forward_min_axis": 0.6664, | |
"inputs": { | |
"axis": 0, | |
"data": [ | |
10000, | |
100 | |
] | |
}, | |
"max_storage_mem_alloc_cpu/0": 0.4 | |
} | |
], | |
"mp_sgd_mom_update": [ | |
{ | |
"avg_time_forward_mp_sgd_mom_update": 0.3911, | |
"inputs": { | |
"grad": [ | |
1024, | |
1024 | |
], | |
"lazy_update": 0, | |
"lr": 0.1, | |
"mom": [ | |
1024, | |
1024 | |
], | |
"rescale_grad": 0.4, | |
"weight": [ | |
1024, | |
1024 | |
], | |
"weight32": [ | |
1024, | |
1024 | |
] | |
}, | |
"max_storage_mem_alloc_cpu/0": 2634981.25 | |
}, | |
{ | |
"avg_time_forward_mp_sgd_mom_update": 0.0414, | |
"inputs": { | |
"grad": [ | |
10000, | |
1 | |
], | |
"lazy_update": 0, | |
"lr": 0.5, | |
"mom": [ | |
10000, | |
1 | |
], | |
"rescale_grad": 0.4, | |
"weight": [ | |
10000, | |
1 | |
], | |
"weight32": [ | |
10000, | |
1 | |
] | |
}, | |
"max_storage_mem_alloc_cpu/0": 2635021.25 | |
}, | |
{ | |
"avg_time_forward_mp_sgd_mom_update": 0.2421, | |
"inputs": { | |
"grad": [ | |
10000, | |
100 | |
], | |
"lazy_update": 0, | |
"lr": 0.9, | |
"mom": [ | |
10000, | |
100 | |
], | |
"rescale_grad": 0.4, | |
"weight": [ | |
10000, | |
100 | |
], | |
"weight32": [ | |
10000, | |
100 | |
] | |
}, | |
"max_storage_mem_alloc_cpu/0": 2639001.25 | |
} | |
], | |
"mp_sgd_update": [ | |
{ | |
"avg_time_forward_mp_sgd_update": 0.2726, | |
"inputs": { | |
"grad": [ | |
1024, | |
1024 | |
], | |
"lazy_update": 0, | |
"lr": 0.1, | |
"rescale_grad": 0.4, | |
"weight": [ | |
1024, | |
1024 | |
], | |
"weight32": [ | |
1024, | |
1024 | |
] | |
}, | |
"max_storage_mem_alloc_cpu/0": 2643292.75 | |
}, | |
{ | |
"avg_time_forward_mp_sgd_update": 0.0379, | |
"inputs": { | |
"grad": [ | |
10000, | |
1 | |
], | |
"lazy_update": 0, | |
"lr": 0.5, | |
"rescale_grad": 0.4, | |
"weight": [ | |
10000, | |
1 | |
], | |
"weight32": [ | |
10000, | |
1 | |
] | |
}, | |
"max_storage_mem_alloc_cpu/0": 2641235.5 | |
}, | |
{ | |
"avg_time_forward_mp_sgd_update": 0.1705, | |
"inputs": { | |
"grad": [ | |
10000, | |
100 | |
], | |
"lazy_update": 0, | |
"lr": 0.9, | |
"rescale_grad": 0.4, | |
"weight": [ | |
10000, | |
100 | |
], | |
"weight32": [ | |
10000, | |
100 | |
] | |
}, | |
"max_storage_mem_alloc_cpu/0": 2645215.5 | |
} | |
], | |
"nanprod": [ | |
{ | |
"avg_time_backward_nanprod": 0.4656, | |
"avg_time_forward_nanprod": 0.5312, | |
"inputs": { | |
"axis": 0, | |
"data": [ | |
1024, | |
1024 | |
] | |
}, | |
"max_storage_mem_alloc_cpu/0": 2.048 | |
}, | |
{ | |
"avg_time_backward_nanprod": 0.0402, | |
"avg_time_forward_nanprod": 0.1463, | |
"inputs": { | |
"axis": 0, | |
"data": [ | |
10000, | |
1 | |
] | |
}, | |
"max_storage_mem_alloc_cpu/0": 0.002 | |
}, | |
{ | |
"avg_time_backward_nanprod": 0.8231, | |
"avg_time_forward_nanprod": 0.58, | |
"inputs": { | |
"axis": 0, | |
"data": [ | |
10000, | |
100 | |
] | |
}, | |
"max_storage_mem_alloc_cpu/0": 0.4 | |
} | |
], | |
"nansum": [ | |
{ | |
"avg_time_backward_nansum": 0.8373, | |
"avg_time_forward_nansum": 0.8334, | |
"inputs": { | |
"axis": 0, | |
"data": [ | |
1024, | |
1024 | |
] | |
}, | |
"max_storage_mem_alloc_cpu/0": 4.096 | |
}, | |
{ | |
"avg_time_backward_nansum": 0.0367, | |
"avg_time_forward_nansum": 0.1526, | |
"inputs": { | |
"axis": 0, | |
"data": [ | |
10000, | |
1 | |
] | |
}, | |
"max_storage_mem_alloc_cpu/0": 0.002 | |
}, | |
{ | |
"avg_time_backward_nansum": 0.8021, | |
"avg_time_forward_nansum": 0.7588, | |
"inputs": { | |
"axis": 0, | |
"data": [ | |
10000, | |
100 | |
] | |
}, | |
"max_storage_mem_alloc_cpu/0": 0.4 | |
} | |
], | |
"negative": [ | |
{ | |
"avg_time_forward_negative": 0.0544, | |
"inputs": { | |
"data": [ | |
1024, | |
1024 | |
] | |
}, | |
"max_storage_mem_alloc_cpu/0": 4194.3042 | |
}, | |
{ | |
"avg_time_forward_negative": 0.0204, | |
"inputs": { | |
"data": [ | |
10000, | |
1 | |
] | |
}, | |
"max_storage_mem_alloc_cpu/0": 40.0 | |
}, | |
{ | |
"avg_time_forward_negative": 0.0572, | |
"inputs": { | |
"data": [ | |
10000, | |
100 | |
] | |
}, | |
"max_storage_mem_alloc_cpu/0": 4000.0 | |
} | |
], | |
"ones_like": [ | |
{ | |
"avg_time_forward_ones_like": 0.0462, | |
"inputs": { | |
"data": [ | |
1024, | |
1024 | |
] | |
}, | |
"max_storage_mem_alloc_cpu/0": 4194.3042 | |
}, | |
{ | |
"avg_time_forward_ones_like": 0.025, | |
"inputs": { | |
"data": [ | |
10000, | |
1 | |
] | |
}, | |
"max_storage_mem_alloc_cpu/0": 20.0 | |
}, | |
{ | |
"avg_time_forward_ones_like": 0.0404, | |
"inputs": { | |
"data": [ | |
10000, | |
100 | |
] | |
}, | |
"max_storage_mem_alloc_cpu/0": 2000.0 | |
} | |
], | |
"prod": [ | |
{ | |
"avg_time_backward_prod": 0.7644, | |
"avg_time_forward_prod": 0.7692, | |
"inputs": { | |
"axis": 0, | |
"data": [ | |
1024, | |
1024 | |
] | |
}, | |
"max_storage_mem_alloc_cpu/0": 2.048 | |
}, | |
{ | |
"avg_time_backward_prod": 0.0402, | |
"avg_time_forward_prod": 0.1388, | |
"inputs": { | |
"axis": 0, | |
"data": [ | |
10000, | |
1 | |
] | |
}, | |
"max_storage_mem_alloc_cpu/0": 0.002 | |
}, | |
{ | |
"avg_time_backward_prod": 0.7376, | |
"avg_time_forward_prod": 0.6659, | |
"inputs": { | |
"axis": 0, | |
"data": [ | |
10000, | |
100 | |
] | |
}, | |
"max_storage_mem_alloc_cpu/0": 0.2 | |
} | |
], | |
"radians": [ | |
{ | |
"avg_time_backward_radians": 0.0522, | |
"avg_time_forward_radians": 0.2718, | |
"inputs": { | |
"data": [ | |
1024, | |
1024 | |
] | |
}, | |
"max_storage_mem_alloc_cpu/0": 4194.3042 | |
}, | |
{ | |
"avg_time_backward_radians": 0.0274, | |
"avg_time_forward_radians": 0.0325, | |
"inputs": { | |
"data": [ | |
10000, | |
1 | |
] | |
}, | |
"max_storage_mem_alloc_cpu/0": 40.0 | |
}, | |
{ | |
"avg_time_backward_radians": 0.057, | |
"avg_time_forward_radians": 0.2365, | |
"inputs": { | |
"data": [ | |
10000, | |
100 | |
] | |
}, | |
"max_storage_mem_alloc_cpu/0": 4000.0 | |
} | |
], | |
"random_exponential": [ | |
{ | |
"avg_time_forward_random_exponential": 2.453, | |
"inputs": { | |
"shape": [ | |
1024, | |
1024 | |
] | |
}, | |
"max_storage_mem_alloc_cpu/0": 4194.3042 | |
}, | |
{ | |
"avg_time_forward_random_exponential": 0.0503, | |
"inputs": { | |
"shape": [ | |
10000, | |
1 | |
] | |
}, | |
"max_storage_mem_alloc_cpu/0": 20.0 | |
}, | |
{ | |
"avg_time_forward_random_exponential": 2.3081, | |
"inputs": { | |
"shape": [ | |
10000, | |
100 | |
] | |
}, | |
"max_storage_mem_alloc_cpu/0": 4000.0 | |
} | |
], | |
"random_gamma": [ | |
{ | |
"avg_time_forward_random_gamma": 7.1039, | |
"inputs": { | |
"shape": [ | |
1024, | |
1024 | |
] | |
}, | |
"max_storage_mem_alloc_cpu/0": 4194.3042 | |
}, | |
{ | |
"avg_time_forward_random_gamma": 0.0933, | |
"inputs": { | |
"shape": [ | |
10000, | |
1 | |
] | |
}, | |
"max_storage_mem_alloc_cpu/0": 20.0 | |
}, | |
{ | |
"avg_time_forward_random_gamma": 6.6858, | |
"inputs": { | |
"shape": [ | |
10000, | |
100 | |
] | |
}, | |
"max_storage_mem_alloc_cpu/0": 4000.0 | |
} | |
], | |
"random_generalized_negative_binomial": [ | |
{ | |
"avg_time_forward_random_generalized_negative_binomial": 7.7819, | |
"inputs": { | |
"shape": [ | |
1024, | |
1024 | |
] | |
}, | |
"max_storage_mem_alloc_cpu/0": 4194.3042 | |
}, | |
{ | |
"avg_time_forward_random_generalized_negative_binomial": 0.2137, | |
"inputs": { | |
"shape": [ | |
10000, | |
1 | |
] | |
}, | |
"max_storage_mem_alloc_cpu/0": 20.0 | |
}, | |
{ | |
"avg_time_forward_random_generalized_negative_binomial": 9.118, | |
"inputs": { | |
"shape": [ | |
10000, | |
100 | |
] | |
}, | |
"max_storage_mem_alloc_cpu/0": 4000.0 | |
} | |
], | |
"random_negative_binomial": [ | |
{ | |
"avg_time_forward_random_negative_binomial": 6.806, | |
"inputs": { | |
"k": 1, | |
"p": 1, | |
"shape": [ | |
1024, | |
1024 | |
] | |
}, | |
"max_storage_mem_alloc_cpu/0": 2097.1521 | |
}, | |
{ | |
"avg_time_forward_random_negative_binomial": 0.2612, | |
"inputs": { | |
"k": 1, | |
"p": 1, | |
"shape": [ | |
10000, | |
1 | |
] | |
}, | |
"max_storage_mem_alloc_cpu/0": 20.0 | |
}, | |
{ | |
"avg_time_forward_random_negative_binomial": 8.1712, | |
"inputs": { | |
"k": 1, | |
"p": 1, | |
"shape": [ | |
10000, | |
100 | |
] | |
}, | |
"max_storage_mem_alloc_cpu/0": 4000.0 | |
} | |
], | |
"random_normal": [ | |
{ | |
"avg_time_forward_random_normal": 1.5839, | |
"inputs": { | |
"shape": [ | |
1024, | |
1024 | |
] | |
}, | |
"max_storage_mem_alloc_cpu/0": 2097.1521 | |
}, | |
{ | |
"avg_time_forward_random_normal": 0.047, | |
"inputs": { | |
"shape": [ | |
10000, | |
1 | |
] | |
}, | |
"max_storage_mem_alloc_cpu/0": 20.0 | |
}, | |
{ | |
"avg_time_forward_random_normal": 1.5105, | |
"inputs": { | |
"shape": [ | |
10000, | |
100 | |
] | |
}, | |
"max_storage_mem_alloc_cpu/0": 2000.0 | |
} | |
], | |
"random_pdf_dirichlet": [ | |
{ | |
"avg_time_forward_random_pdf_dirichlet": 0.0327, | |
"inputs": { | |
"alpha": [ | |
0.0, | |
2.5 | |
], | |
"sample": [ | |
2 | |
] | |
}, | |
"max_storage_mem_alloc_cpu/0": 0.002 | |
} | |
], | |
"random_pdf_exponential": [ | |
{ | |
"avg_time_forward_random_pdf_exponential": 0.0349, | |
"inputs": { | |
"lam": [ | |
1.0, | |
8.5 | |
], | |
"sample": [ | |
2 | |
] | |
}, | |
"max_storage_mem_alloc_cpu/0": 0.004 | |
} | |
], | |
"random_pdf_gamma": [ | |
{ | |
"avg_time_forward_random_pdf_gamma": 0.0354, | |
"inputs": { | |
"alpha": [ | |
0.0, | |
2.5 | |
], | |
"beta": [ | |
1.0, | |
0.7 | |
], | |
"sample": [ | |
2 | |
] | |
}, | |
"max_storage_mem_alloc_cpu/0": 0.004 | |
} | |
], | |
"random_pdf_generalized_negative_binomial": [ | |
{ | |
"avg_time_forward_random_pdf_generalized_negative_binomial": 0.0416, | |
"inputs": { | |
"alpha": [ | |
0.0, | |
2.5 | |
], | |
"mu": [ | |
2.0, | |
2.5 | |
], | |
"sample": [ | |
2 | |
] | |
}, | |
"max_storage_mem_alloc_cpu/0": 0.008 | |
} | |
], | |
"random_pdf_negative_binomial": [ | |
{ | |
"avg_time_forward_random_pdf_negative_binomial": 0.041, | |
"inputs": { | |
"k": [ | |
20, | |
49 | |
], | |
"p": [ | |
0.4, | |
0.77 | |
], | |
"sample": [ | |
2 | |
] | |
}, | |
"max_storage_mem_alloc_cpu/0": 0.004 | |
} | |
], | |
"random_pdf_normal": [ | |
{ | |
"avg_time_forward_random_pdf_normal": 0.0364, | |
"inputs": { | |
"mu": [ | |
2.0, | |
2.5 | |
], | |
"sample": [ | |
2 | |
], | |
"sigma": [ | |
1.0, | |
3.7 | |
] | |
}, | |
"max_storage_mem_alloc_cpu/0": 0.004 | |
} | |
], | |
"random_pdf_poisson": [ | |
{ | |
"avg_time_forward_random_pdf_poisson": 0.0404, | |
"inputs": { | |
"lam": [ | |
1.0, | |
8.5 | |
], | |
"sample": [ | |
2 | |
] | |
}, | |
"max_storage_mem_alloc_cpu/0": 0.004 | |
} | |
], | |
"random_pdf_uniform": [ | |
{ | |
"avg_time_forward_random_pdf_uniform": 0.0348, | |
"inputs": { | |
"high": [ | |
1.0, | |
3.7 | |
], | |
"low": [ | |
0.0, | |
2.5 | |
], | |
"sample": [ | |
2 | |
] | |
}, | |
"max_storage_mem_alloc_cpu/0": 0.004 | |
} | |
], | |
"random_poisson": [ | |
{ | |
"avg_time_forward_random_poisson": 1.5076, | |
"inputs": { | |
"shape": [ | |
1024, | |
1024 | |
] | |
}, | |
"max_storage_mem_alloc_cpu/0": 2097.1521 | |
}, | |
{ | |
"avg_time_forward_random_poisson": 0.048, | |
"inputs": { | |
"shape": [ | |
10000, | |
1 | |
] | |
}, | |
"max_storage_mem_alloc_cpu/0": 20.0 | |
}, | |
{ | |
"avg_time_forward_random_poisson": 1.5337, | |
"inputs": { | |
"shape": [ | |
10000, | |
100 | |
] | |
}, | |
"max_storage_mem_alloc_cpu/0": 2000.0 | |
} | |
], | |
"random_randint": [ | |
{ | |
"avg_time_forward_random_randint": 1.3664, | |
"inputs": { | |
"high": 5, | |
"low": 0, | |
"shape": [ | |
1024, | |
1024 | |
] | |
}, | |
"max_storage_mem_alloc_cpu/0": 2097.1521 | |
}, | |
{ | |
"avg_time_forward_random_randint": 0.0426, | |
"inputs": { | |
"high": 5, | |
"low": 0, | |
"shape": [ | |
10000, | |
1 | |
] | |
}, | |
"max_storage_mem_alloc_cpu/0": 20.0 | |
}, | |
{ | |
"avg_time_forward_random_randint": 1.3065, | |
"inputs": { | |
"high": 5, | |
"low": 0, | |
"shape": [ | |
10000, | |
100 | |
] | |
}, | |
"max_storage_mem_alloc_cpu/0": 4000.0 | |
} | |
], | |
"random_uniform": [ | |
{ | |
"avg_time_forward_random_uniform": 0.5823, | |
"inputs": { | |
"high": 5, | |
"low": 0, | |
"shape": [ | |
1024, | |
1024 | |
] | |
}, | |
"max_storage_mem_alloc_cpu/0": 4194.3042 | |
}, | |
{ | |
"avg_time_forward_random_uniform": 0.0356, | |
"inputs": { | |
"high": 5, | |
"low": 0, | |
"shape": [ | |
10000, | |
1 | |
] | |
}, | |
"max_storage_mem_alloc_cpu/0": 40.0 | |
}, | |
{ | |
"avg_time_forward_random_uniform": 0.5671, | |
"inputs": { | |
"high": 5, | |
"low": 0, | |
"shape": [ | |
10000, | |
100 | |
] | |
}, | |
"max_storage_mem_alloc_cpu/0": 4000.0 | |
} | |
], | |
"rcbrt": [ | |
{ | |
"avg_time_backward_rcbrt": 1.6738, | |
"avg_time_forward_rcbrt": 1.4382, | |
"inputs": { | |
"data": [ | |
1024, | |
1024 | |
] | |
}, | |
"max_storage_mem_alloc_cpu/0": 4194.3042 | |
}, | |
{ | |
"avg_time_backward_rcbrt": 0.0419, | |
"avg_time_forward_rcbrt": 0.0437, | |
"inputs": { | |
"data": [ | |
10000, | |
1 | |
] | |
}, | |
"max_storage_mem_alloc_cpu/0": 40.0 | |
}, | |
{ | |
"avg_time_backward_rcbrt": 1.3627, | |
"avg_time_forward_rcbrt": 1.4097, | |
"inputs": { | |
"data": [ | |
10000, | |
100 | |
] | |
}, | |
"max_storage_mem_alloc_cpu/0": 4000.0 | |
} | |
], | |
"reciprocal": [ | |
{ | |
"avg_time_backward_reciprocal": 0.0806, | |
"avg_time_forward_reciprocal": 0.1234, | |
"inputs": { | |
"data": [ | |
1024, | |
1024 | |
] | |
}, | |
"max_storage_mem_alloc_cpu/0": 4194.3042 | |
}, | |
{ | |
"avg_time_backward_reciprocal": 0.0322, | |
"avg_time_forward_reciprocal": 0.0328, | |
"inputs": { | |
"data": [ | |
10000, | |
1 | |
] | |
}, | |
"max_storage_mem_alloc_cpu/0": 40.0 | |
}, | |
{ | |
"avg_time_backward_reciprocal": 0.0715, | |
"avg_time_forward_reciprocal": 0.3225, | |
"inputs": { | |
"data": [ | |
10000, | |
100 | |
] | |
}, | |
"max_storage_mem_alloc_cpu/0": 4000.0 | |
} | |
], | |
"relu": [ | |
{ | |
"avg_time_backward_relu": 0.1031, | |
"avg_time_forward_relu": 0.2552, | |
"inputs": { | |
"data": [ | |
1024, | |
1024 | |
] | |
}, | |
"max_storage_mem_alloc_cpu/0": 4194.3042 | |
}, | |
{ | |
"avg_time_backward_relu": 0.0482, | |
"avg_time_forward_relu": 0.0435, | |
"inputs": { | |
"data": [ | |
10000, | |
1 | |
] | |
}, | |
"max_storage_mem_alloc_cpu/0": 20.0 | |
}, | |
{ | |
"avg_time_backward_relu": 0.1008, | |
"avg_time_forward_relu": 0.1457, | |
"inputs": { | |
"data": [ | |
10000, | |
100 | |
] | |
}, | |
"max_storage_mem_alloc_cpu/0": 2000.0 | |
} | |
], | |
"rint": [ | |
{ | |
"avg_time_forward_rint": 0.288, | |
"inputs": { | |
"data": [ | |
1024, | |
1024 | |
] | |
}, | |
"max_storage_mem_alloc_cpu/0": 2097.1521 | |
}, | |
{ | |
"avg_time_forward_rint": 0.0348, | |
"inputs": { | |
"data": [ | |
10000, | |
1 | |
] | |
}, | |
"max_storage_mem_alloc_cpu/0": 20.0 | |
}, | |
{ | |
"avg_time_forward_rint": 0.2774, | |
"inputs": { | |
"data": [ | |
10000, | |
100 | |
] | |
}, | |
"max_storage_mem_alloc_cpu/0": 2000.0 | |
} | |
], | |
"rmsprop_update": [ | |
{ | |
"avg_time_forward_rmsprop_update": 0.495, | |
"inputs": { | |
"epsilon": 1e-08, | |
"gamma1": 0.1, | |
"grad": [ | |
1024, | |
1024 | |
], | |
"lr": 0.1, | |
"n": [ | |
1024, | |
1024 | |
], | |
"rescale_grad": 0.4, | |
"weight": [ | |
1024, | |
1024 | |
] | |
}, | |
"max_storage_mem_alloc_cpu/0": 2649507.0 | |
}, | |
{ | |
"avg_time_forward_rmsprop_update": 0.0379, | |
"inputs": { | |
"epsilon": 1e-08, | |
"gamma1": 0.5, | |
"grad": [ | |
10000, | |
1 | |
], | |
"lr": 0.5, | |
"n": [ | |
10000, | |
1 | |
], | |
"rescale_grad": 0.4, | |
"weight": [ | |
10000, | |
1 | |
] | |
}, | |
"max_storage_mem_alloc_cpu/0": 2647450.0 | |
}, | |
{ | |
"avg_time_forward_rmsprop_update": 0.5192, | |
"inputs": { | |
"epsilon": 1e-08, | |
"gamma1": 0.9, | |
"grad": [ | |
10000, | |
100 | |
], | |
"lr": 0.9, | |
"n": [ | |
10000, | |
100 | |
], | |
"rescale_grad": 0.4, | |
"weight": [ | |
10000, | |
100 | |
] | |
}, | |
"max_storage_mem_alloc_cpu/0": 2651430.0 | |
} | |
], | |
"rmspropalex_update": [ | |
{ | |
"avg_time_forward_rmspropalex_update": 1.0621, | |
"inputs": { | |
"delta": [ | |
1024, | |
1024 | |
], | |
"epsilon": 1e-08, | |
"g": [ | |
1024, | |
1024 | |
], | |
"gamma1": 0.1, | |
"gamma2": 0.1, | |
"grad": [ | |
1024, | |
1024 | |
], | |
"lr": 0.1, | |
"n": [ | |
1024, | |
1024 | |
], | |
"rescale_grad": 0.4, | |
"weight": [ | |
1024, | |
1024 | |
] | |
}, | |
"max_storage_mem_alloc_cpu/0": 2653624.25 | |
}, | |
{ | |
"avg_time_forward_rmspropalex_update": 0.061, | |
"inputs": { | |
"delta": [ | |
10000, | |
1 | |
], | |
"epsilon": 1e-08, | |
"g": [ | |
10000, | |
1 | |
], | |
"gamma1": 0.5, | |
"gamma2": 0.5, | |
"grad": [ | |
10000, | |
1 | |
], | |
"lr": 0.5, | |
"n": [ | |
10000, | |
1 | |
], | |
"rescale_grad": 0.4, | |
"weight": [ | |
10000, | |
1 | |
] | |
}, | |
"max_storage_mem_alloc_cpu/0": 2653664.25 | |
}, | |
{ | |
"avg_time_forward_rmspropalex_update": 2.0506, | |
"inputs": { | |
"delta": [ | |
10000, | |
100 | |
], | |
"epsilon": 1e-08, | |
"g": [ | |
10000, | |
100 | |
], | |
"gamma1": 0.9, | |
"gamma2": 0.9, | |
"grad": [ | |
10000, | |
100 | |
], | |
"lr": 0.9, | |
"n": [ | |
10000, | |
100 | |
], | |
"rescale_grad": 0.4, | |
"weight": [ | |
10000, | |
100 | |
] | |
}, | |
"max_storage_mem_alloc_cpu/0": 2657644.25 | |
} | |
], | |
"round": [ | |
{ | |
"avg_time_forward_round": 0.3479, | |
"inputs": { | |
"data": [ | |
1024, | |
1024 | |
] | |
}, | |
"max_storage_mem_alloc_cpu/0": 2097.1521 | |
}, | |
{ | |
"avg_time_forward_round": 0.0369, | |
"inputs": { | |
"data": [ | |
10000, | |
1 | |
] | |
}, | |
"max_storage_mem_alloc_cpu/0": 20.0 | |
}, | |
{ | |
"avg_time_forward_round": 0.3433, | |
"inputs": { | |
"data": [ | |
10000, | |
100 | |
] | |
}, | |
"max_storage_mem_alloc_cpu/0": 2000.0 | |
} | |
], | |
"rsqrt": [ | |
{ | |
"avg_time_backward_rsqrt": 0.6861, | |
"avg_time_forward_rsqrt": 0.7197, | |
"inputs": { | |
"data": [ | |
1024, | |
1024 | |
] | |
}, | |
"max_storage_mem_alloc_cpu/0": 2097.1521 | |
}, | |
{ | |
"avg_time_backward_rsqrt": 0.0325, | |
"avg_time_forward_rsqrt": 0.0387, | |
"inputs": { | |
"data": [ | |
10000, | |
1 | |
] | |
}, | |
"max_storage_mem_alloc_cpu/0": 40.0 | |
}, | |
{ | |
"avg_time_backward_rsqrt": 0.6652, | |
"avg_time_forward_rsqrt": 0.7411, | |
"inputs": { | |
"data": [ | |
10000, | |
100 | |
] | |
}, | |
"max_storage_mem_alloc_cpu/0": 2000.0 | |
} | |
], | |
"sample_exponential": [ | |
{ | |
"avg_time_forward_sample_exponential": 3.1745, | |
"inputs": { | |
"lam": [ | |
1.0, | |
8.5 | |
], | |
"shape": [ | |
1024, | |
1024 | |
] | |
}, | |
"max_storage_mem_alloc_cpu/0": 4194.3042 | |
}, | |
{ | |
"avg_time_forward_sample_exponential": 0.2575, | |
"inputs": { | |
"lam": [ | |
1.0, | |
8.5 | |
], | |
"shape": [ | |
10000, | |
1 | |
] | |
}, | |
"max_storage_mem_alloc_cpu/0": 40.0 | |
}, | |
{ | |
"avg_time_forward_sample_exponential": 4.7243, | |
"inputs": { | |
"lam": [ | |
1.0, | |
8.5 | |
], | |
"shape": [ | |
10000, | |
100 | |
] | |
}, | |
"max_storage_mem_alloc_cpu/0": 8000.0 | |
} | |
], | |
"sample_gamma": [ | |
{ | |
"avg_time_forward_sample_gamma": 13.948, | |
"inputs": { | |
"alpha": [ | |
0.0, | |
2.5 | |
], | |
"beta": [ | |
1.0, | |
0.7 | |
], | |
"shape": [ | |
1024, | |
1024 | |
] | |
}, | |
"max_storage_mem_alloc_cpu/0": 8388.6084 | |
}, | |
{ | |
"avg_time_forward_sample_gamma": 0.3941, | |
"inputs": { | |
"alpha": [ | |
0.0, | |
2.5 | |
], | |
"beta": [ | |
1.0, | |
0.7 | |
], | |
"shape": [ | |
10000, | |
1 | |
] | |
}, | |
"max_storage_mem_alloc_cpu/0": 80.0 | |
}, | |
{ | |
"avg_time_forward_sample_gamma": 16.8534, | |
"inputs": { | |
"alpha": [ | |
0.0, | |
2.5 | |
], | |
"beta": [ | |
1.0, | |
0.7 | |
], | |
"shape": [ | |
10000, | |
100 | |
] | |
}, | |
"max_storage_mem_alloc_cpu/0": 8000.0 | |
} | |
], | |
"sample_generalized_negative_binomial": [ | |
{ | |
"avg_time_forward_sample_generalized_negative_binomial": 32.5925, | |
"inputs": { | |
"alpha": [ | |
0.0, | |
2.5 | |
], | |
"mu": [ | |
2.0, | |
2.5 | |
], | |
"shape": [ | |
1024, | |
1024 | |
] | |
}, | |
"max_storage_mem_alloc_cpu/0": 8388.6084 | |
}, | |
{ | |
"avg_time_forward_sample_generalized_negative_binomial": 0.3845, | |
"inputs": { | |
"alpha": [ | |
0.0, | |
2.5 | |
], | |
"mu": [ | |
2.0, | |
2.5 | |
], | |
"shape": [ | |
10000, | |
1 | |
] | |
}, | |
"max_storage_mem_alloc_cpu/0": 80.0 | |
}, | |
{ | |
"avg_time_forward_sample_generalized_negative_binomial": 28.6354, | |
"inputs": { | |
"alpha": [ | |
0.0, | |
2.5 | |
], | |
"mu": [ | |
2.0, | |
2.5 | |
], | |
"shape": [ | |
10000, | |
100 | |
] | |
}, | |
"max_storage_mem_alloc_cpu/0": 8000.0 | |
} | |
], | |
"sample_negative_binomial": [ | |
{ | |
"avg_time_forward_sample_negative_binomial": 314.4601, | |
"inputs": { | |
"k": [ | |
20, | |
49 | |
], | |
"p": [ | |
0.4, | |
0.77 | |
], | |
"shape": [ | |
1024, | |
1024 | |
] | |
}, | |
"max_storage_mem_alloc_cpu/0": 4194.3042 | |
}, | |
{ | |
"avg_time_forward_sample_negative_binomial": 3.0867, | |
"inputs": { | |
"k": [ | |
20, | |
49 | |
], | |
"p": [ | |
0.4, | |
0.77 | |
], | |
"shape": [ | |
10000, | |
1 | |
] | |
}, | |
"max_storage_mem_alloc_cpu/0": 80.0 | |
}, | |
{ | |
"avg_time_forward_sample_negative_binomial": 299.933, | |
"inputs": { | |
"k": [ | |
20, | |
49 | |
], | |
"p": [ | |
0.4, | |
0.77 | |
], | |
"shape": [ | |
10000, | |
100 | |
] | |
}, | |
"max_storage_mem_alloc_cpu/0": 4000.0 | |
} | |
], | |
"sample_normal": [ | |
{ | |
"avg_time_forward_sample_normal": 3.1211, | |
"inputs": { | |
"mu": [ | |
2.0, | |
2.5 | |
], | |
"shape": [ | |
1024, | |
1024 | |
], | |
"sigma": [ | |
1.0, | |
3.7 | |
] | |
}, | |
"max_storage_mem_alloc_cpu/0": 4194.3042 | |
}, | |
{ | |
"avg_time_forward_sample_normal": 0.1975, | |
"inputs": { | |
"mu": [ | |
2.0, | |
2.5 | |
], | |
"shape": [ | |
10000, | |
1 | |
], | |
"sigma": [ | |
1.0, | |
3.7 | |
] | |
}, | |
"max_storage_mem_alloc_cpu/0": 40.0 | |
}, | |
{ | |
"avg_time_forward_sample_normal": 4.9562, | |
"inputs": { | |
"mu": [ | |
2.0, | |
2.5 | |
], | |
"shape": [ | |
10000, | |
100 | |
], | |
"sigma": [ | |
1.0, | |
3.7 | |
] | |
}, | |
"max_storage_mem_alloc_cpu/0": 8000.0 | |
} | |
], | |
"sample_poisson": [ | |
{ | |
"avg_time_forward_sample_poisson": 9.7635, | |
"inputs": { | |
"lam": [ | |
1.0, | |
8.5 | |
], | |
"shape": [ | |
1024, | |
1024 | |
] | |
}, | |
"max_storage_mem_alloc_cpu/0": 8388.6084 | |
}, | |
{ | |
"avg_time_forward_sample_poisson": 0.1201, | |
"inputs": { | |
"lam": [ | |
1.0, | |
8.5 | |
], | |
"shape": [ | |
10000, | |
1 | |
] | |
}, | |
"max_storage_mem_alloc_cpu/0": 40.0 | |
}, | |
{ | |
"avg_time_forward_sample_poisson": 9.0652, | |
"inputs": { | |
"lam": [ | |
1.0, | |
8.5 | |
], | |
"shape": [ | |
10000, | |
100 | |
] | |
}, | |
"max_storage_mem_alloc_cpu/0": 8000.0 | |
} | |
], | |
"sample_uniform": [ | |
{ | |
"avg_time_forward_sample_uniform": 1.1059, | |
"inputs": { | |
"high": [ | |
1.0, | |
3.7 | |
], | |
"low": [ | |
0.0, | |
2.5 | |
], | |
"shape": [ | |
1024, | |
1024 | |
] | |
}, | |
"max_storage_mem_alloc_cpu/0": 8388.6084 | |
}, | |
{ | |
"avg_time_forward_sample_uniform": 0.0488, | |
"inputs": { | |
"high": [ | |
1.0, | |
3.7 | |
], | |
"low": [ | |
0.0, | |
2.5 | |
], | |
"shape": [ | |
10000, | |
1 | |
] | |
}, | |
"max_storage_mem_alloc_cpu/0": 80.0 | |
}, | |
{ | |
"avg_time_forward_sample_uniform": 1.0615, | |
"inputs": { | |
"high": [ | |
1.0, | |
3.7 | |
], | |
"low": [ | |
0.0, | |
2.5 | |
], | |
"shape": [ | |
10000, | |
100 | |
] | |
}, | |
"max_storage_mem_alloc_cpu/0": 8000.0 | |
} | |
], | |
"sgd_mom_update": [ | |
{ | |
"avg_time_forward_sgd_mom_update": 0.2869, | |
"inputs": { | |
"grad": [ | |
1024, | |
1024 | |
], | |
"lazy_update": 0, | |
"lr": 0.1, | |
"mom": [ | |
1024, | |
1024 | |
], | |
"rescale_grad": 0.4, | |
"weight": [ | |
1024, | |
1024 | |
] | |
}, | |
"max_storage_mem_alloc_cpu/0": 2661935.75 | |
}, | |
{ | |
"avg_time_forward_sgd_mom_update": 0.0452, | |
"inputs": { | |
"grad": [ | |
10000, | |
1 | |
], | |
"lazy_update": 0, | |
"lr": 0.5, | |
"mom": [ | |
10000, | |
1 | |
], | |
"rescale_grad": 0.4, | |
"weight": [ | |
10000, | |
1 | |
] | |
}, | |
"max_storage_mem_alloc_cpu/0": 2659878.5 | |
}, | |
{ | |
"avg_time_forward_sgd_mom_update": 0.1963, | |
"inputs": { | |
"grad": [ | |
10000, | |
100 | |
], | |
"lazy_update": 0, | |
"lr": 0.9, | |
"mom": [ | |
10000, | |
100 | |
], | |
"rescale_grad": 0.4, | |
"weight": [ | |
10000, | |
100 | |
] | |
}, | |
"max_storage_mem_alloc_cpu/0": 2663858.5 | |
} | |
], | |
"sgd_update": [ | |
{ | |
"avg_time_forward_sgd_update": 0.2671, | |
"inputs": { | |
"grad": [ | |
1024, | |
1024 | |
], | |
"lazy_update": 0, | |
"lr": 0.1, | |
"rescale_grad": 0.4, | |
"weight": [ | |
1024, | |
1024 | |
] | |
}, | |
"max_storage_mem_alloc_cpu/0": 2668150.0 | |
}, | |
{ | |
"avg_time_forward_sgd_update": 0.0435, | |
"inputs": { | |
"grad": [ | |
10000, | |
1 | |
], | |
"lazy_update": 0, | |
"lr": 0.5, | |
"rescale_grad": 0.4, | |
"weight": [ | |
10000, | |
1 | |
] | |
}, | |
"max_storage_mem_alloc_cpu/0": 2666092.75 | |
}, | |
{ | |
"avg_time_forward_sgd_update": 0.173, | |
"inputs": { | |
"grad": [ | |
10000, | |
100 | |
], | |
"lazy_update": 0, | |
"lr": 0.9, | |
"rescale_grad": 0.4, | |
"weight": [ | |
10000, | |
100 | |
] | |
}, | |
"max_storage_mem_alloc_cpu/0": 2670072.75 | |
} | |
], | |
"shape_array": [ | |
{ | |
"avg_time_forward_shape_array": 0.0153, | |
"inputs": { | |
"data": [ | |
1024, | |
1024 | |
] | |
}, | |
"max_storage_mem_alloc_cpu/0": 0.008 | |
}, | |
{ | |
"avg_time_forward_shape_array": 0.0149, | |
"inputs": { | |
"data": [ | |
10000, | |
1 | |
] | |
}, | |
"max_storage_mem_alloc_cpu/0": 0.008 | |
}, | |
{ | |
"avg_time_forward_shape_array": 0.0146, | |
"inputs": { | |
"data": [ | |
10000, | |
100 | |
] | |
}, | |
"max_storage_mem_alloc_cpu/0": 0.008 | |
} | |
], | |
"shuffle": [ | |
{ | |
"avg_time_forward_shuffle": 1.0521, | |
"inputs": { | |
"data": [ | |
1024, | |
1024 | |
] | |
}, | |
"max_storage_mem_alloc_cpu/0": 2097.1521 | |
}, | |
{ | |
"avg_time_forward_shuffle": 0.411, | |
"inputs": { | |
"data": [ | |
10000, | |
1 | |
] | |
}, | |
"max_storage_mem_alloc_cpu/0": 20.0 | |
}, | |
{ | |
"avg_time_forward_shuffle": 1.2007, | |
"inputs": { | |
"data": [ | |
10000, | |
100 | |
] | |
}, | |
"max_storage_mem_alloc_cpu/0": 2000.0 | |
} | |
], | |
"sigmoid": [ | |
{ | |
"avg_time_backward_sigmoid": 0.0712, | |
"avg_time_forward_sigmoid": 0.8389, | |
"inputs": { | |
"data": [ | |
1024, | |
1024 | |
] | |
}, | |
"max_storage_mem_alloc_cpu/0": 2097.1521 | |
}, | |
{ | |
"avg_time_backward_sigmoid": 0.0247, | |
"avg_time_forward_sigmoid": 0.0419, | |
"inputs": { | |
"data": [ | |
10000, | |
1 | |
] | |
}, | |
"max_storage_mem_alloc_cpu/0": 20.0 | |
}, | |
{ | |
"avg_time_backward_sigmoid": 0.0586, | |
"avg_time_forward_sigmoid": 0.6822, | |
"inputs": { | |
"data": [ | |
10000, | |
100 | |
] | |
}, | |
"max_storage_mem_alloc_cpu/0": 2000.0 | |
} | |
], | |
"sign": [ | |
{ | |
"avg_time_backward_sign": 0.0463, | |
"avg_time_forward_sign": 0.2619, | |
"inputs": { | |
"data": [ | |
1024, | |
1024 | |
] | |
}, | |
"max_storage_mem_alloc_cpu/0": 2097.1521 | |
}, | |
{ | |
"avg_time_backward_sign": 0.0194, | |
"avg_time_forward_sign": 0.0669, | |
"inputs": { | |
"data": [ | |
10000, | |
1 | |
] | |
}, | |
"max_storage_mem_alloc_cpu/0": 20.0 | |
}, | |
{ | |
"avg_time_backward_sign": 0.0445, | |
"avg_time_forward_sign": 0.2508, | |
"inputs": { | |
"data": [ | |
10000, | |
100 | |
] | |
}, | |
"max_storage_mem_alloc_cpu/0": 2000.0 | |
} | |
], | |
"signsgd_update": [ | |
{ | |
"avg_time_forward_signsgd_update": 0.265, | |
"inputs": { | |
"grad": [ | |
1024, | |
1024 | |
], | |
"lr": 0.1, | |
"rescale_grad": 0.4, | |
"weight": [ | |
1024, | |
1024 | |
] | |
}, | |
"max_storage_mem_alloc_cpu/0": 2674364.25 | |
}, | |
{ | |
"avg_time_forward_signsgd_update": 0.0339, | |
"inputs": { | |
"grad": [ | |
10000, | |
1 | |
], | |
"lr": 0.5, | |
"rescale_grad": 0.4, | |
"weight": [ | |
10000, | |
1 | |
] | |
}, | |
"max_storage_mem_alloc_cpu/0": 2672307.0 | |
}, | |
{ | |
"avg_time_forward_signsgd_update": 0.2048, | |
"inputs": { | |
"grad": [ | |
10000, | |
100 | |
], | |
"lr": 0.9, | |
"rescale_grad": 0.4, | |
"weight": [ | |
10000, | |
100 | |
] | |
}, | |
"max_storage_mem_alloc_cpu/0": 2676287.0 | |
} | |
], | |
"signum_update": [ | |
{ | |
"avg_time_forward_signum_update": 0.5366, | |
"inputs": { | |
"grad": [ | |
1024, | |
1024 | |
], | |
"lr": 0.1, | |
"mom": [ | |
1024, | |
1024 | |
], | |
"rescale_grad": 0.4, | |
"weight": [ | |
1024, | |
1024 | |
] | |
}, | |
"max_storage_mem_alloc_cpu/0": 2680578.5 | |
}, | |
{ | |
"avg_time_forward_signum_update": 0.0361, | |
"inputs": { | |
"grad": [ | |
10000, | |
1 | |
], | |
"lr": 0.5, | |
"mom": [ | |
10000, | |
1 | |
], | |
"rescale_grad": 0.4, | |
"weight": [ | |
10000, | |
1 | |
] | |
}, | |
"max_storage_mem_alloc_cpu/0": 2678521.5 | |
}, | |
{ | |
"avg_time_forward_signum_update": 0.3554, | |
"inputs": { | |
"grad": [ | |
10000, | |
100 | |
], | |
"lr": 0.9, | |
"mom": [ | |
10000, | |
100 | |
], | |
"rescale_grad": 0.4, | |
"weight": [ | |
10000, | |
100 | |
] | |
}, | |
"max_storage_mem_alloc_cpu/0": 2682501.5 | |
} | |
], | |
"sin": [ | |
{ | |
"avg_time_backward_sin": 0.511, | |
"avg_time_forward_sin": 0.508, | |
"inputs": { | |
"data": [ | |
1024, | |
1024 | |
] | |
}, | |
"max_storage_mem_alloc_cpu/0": 2097.1521 | |
}, | |
{ | |
"avg_time_backward_sin": 0.0247, | |
"avg_time_forward_sin": 0.0291, | |
"inputs": { | |
"data": [ | |
10000, | |
1 | |
] | |
}, | |
"max_storage_mem_alloc_cpu/0": 20.0 | |
}, | |
{ | |
"avg_time_backward_sin": 0.4947, | |
"avg_time_forward_sin": 0.4925, | |
"inputs": { | |
"data": [ | |
10000, | |
100 | |
] | |
}, | |
"max_storage_mem_alloc_cpu/0": 2000.0 | |
} | |
], | |
"sinh": [ | |
{ | |
"avg_time_backward_sinh": 1.0077, | |
"avg_time_forward_sinh": 1.6091, | |
"inputs": { | |
"data": [ | |
1024, | |
1024 | |
] | |
}, | |
"max_storage_mem_alloc_cpu/0": 2097.1521 | |
}, | |
{ | |
"avg_time_backward_sinh": 0.0469, | |
"avg_time_forward_sinh": 0.0533, | |
"inputs": { | |
"data": [ | |
10000, | |
1 | |
] | |
}, | |
"max_storage_mem_alloc_cpu/0": 20.0 | |
}, | |
{ | |
"avg_time_backward_sinh": 0.9642, | |
"avg_time_forward_sinh": 1.5352, | |
"inputs": { | |
"data": [ | |
10000, | |
100 | |
] | |
}, | |
"max_storage_mem_alloc_cpu/0": 2000.0 | |
} | |
], | |
"size_array": [ | |
{ | |
"avg_time_forward_size_array": 0.012, | |
"inputs": { | |
"data": [ | |
1024, | |
1024 | |
] | |
}, | |
"max_storage_mem_alloc_cpu/0": 0.008 | |
}, | |
{ | |
"avg_time_forward_size_array": 0.0125, | |
"inputs": { | |
"data": [ | |
10000, | |
1 | |
] | |
}, | |
"max_storage_mem_alloc_cpu/0": 0.008 | |
}, | |
{ | |
"avg_time_forward_size_array": 0.0126, | |
"inputs": { | |
"data": [ | |
10000, | |
100 | |
] | |
}, | |
"max_storage_mem_alloc_cpu/0": 0.008 | |
} | |
], | |
"softmax": [ | |
{ | |
"avg_time_backward_softmax": 0.1423, | |
"avg_time_forward_softmax": 1.7546, | |
"inputs": { | |
"axis": -1, | |
"data": [ | |
1024, | |
1024 | |
], | |
"temperature": 0.5 | |
}, | |
"max_storage_mem_alloc_cpu/0": 4194.3042 | |
}, | |
{ | |
"avg_time_backward_softmax": 0.0362, | |
"avg_time_forward_softmax": 0.0493, | |
"inputs": { | |
"axis": -1, | |
"data": [ | |
10000, | |
1 | |
], | |
"temperature": 0.5 | |
}, | |
"max_storage_mem_alloc_cpu/0": 20.0 | |
}, | |
{ | |
"avg_time_backward_softmax": 0.1479, | |
"avg_time_forward_softmax": 1.6998, | |
"inputs": { | |
"axis": -1, | |
"data": [ | |
10000, | |
100 | |
], | |
"temperature": 0.5 | |
}, | |
"max_storage_mem_alloc_cpu/0": 2000.0 | |
} | |
], | |
"softsign": [ | |
{ | |
"avg_time_backward_softsign": 0.0612, | |
"avg_time_forward_softsign": 0.0718, | |
"inputs": { | |
"data": [ | |
1024, | |
1024 | |
] | |
}, | |
"max_storage_mem_alloc_cpu/0": 2097.1521 | |
}, | |
{ | |
"avg_time_backward_softsign": 0.0277, | |
"avg_time_forward_softsign": 0.0248, | |
"inputs": { | |
"data": [ | |
10000, | |
1 | |
] | |
}, | |
"max_storage_mem_alloc_cpu/0": 20.0 | |
}, | |
{ | |
"avg_time_backward_softsign": 0.0645, | |
"avg_time_forward_softsign": 0.0729, | |
"inputs": { | |
"data": [ | |
10000, | |
100 | |
] | |
}, | |
"max_storage_mem_alloc_cpu/0": 2000.0 | |
} | |
], | |
"sort": [ | |
{ | |
"avg_time_forward_sort": 22.9446, | |
"inputs": { | |
"axis": 0, | |
"data": [ | |
1024, | |
1024 | |
] | |
}, | |
"max_storage_mem_alloc_cpu/0": 4194.3042 | |
}, | |
{ | |
"avg_time_forward_sort": 1.0443, | |
"inputs": { | |
"axis": 0, | |
"data": [ | |
10000, | |
1 | |
] | |
}, | |
"max_storage_mem_alloc_cpu/0": 60.0 | |
}, | |
{ | |
"avg_time_forward_sort": 21.0701, | |
"inputs": { | |
"axis": 0, | |
"data": [ | |
10000, | |
100 | |
] | |
}, | |
"max_storage_mem_alloc_cpu/0": 4000.0 | |
} | |
], | |
"space_to_depth": [ | |
{ | |
"avg_time_forward_space_to_depth": 0.0505, | |
"inputs": { | |
"block_size": 2, | |
"data": [ | |
1, | |
4, | |
2, | |
4 | |
] | |
}, | |
"max_storage_mem_alloc_cpu/0": 0.064 | |
}, | |
{ | |
"avg_time_forward_space_to_depth": 0.3008, | |
"inputs": { | |
"block_size": 5, | |
"data": [ | |
10, | |
25, | |
10, | |
100 | |
] | |
}, | |
"max_storage_mem_alloc_cpu/0": 1000.0 | |
} | |
], | |
"sqrt": [ | |
{ | |
"avg_time_backward_sqrt": 0.0639, | |
"avg_time_forward_sqrt": 0.7595, | |
"inputs": { | |
"data": [ | |
1024, | |
1024 | |
] | |
}, | |
"max_storage_mem_alloc_cpu/0": 2097.1521 | |
}, | |
{ | |
"avg_time_backward_sqrt": 0.028, | |
"avg_time_forward_sqrt": 0.0448, | |
"inputs": { | |
"data": [ | |
10000, | |
1 | |
] | |
}, | |
"max_storage_mem_alloc_cpu/0": 20.0 | |
}, | |
{ | |
"avg_time_backward_sqrt": 0.061, | |
"avg_time_forward_sqrt": 0.6866, | |
"inputs": { | |
"data": [ | |
10000, | |
100 | |
] | |
}, | |
"max_storage_mem_alloc_cpu/0": 2000.0 | |
} | |
], | |
"square": [ | |
{ | |
"avg_time_backward_square": 0.0604, | |
"avg_time_forward_square": 0.0612, | |
"inputs": { | |
"data": [ | |
1024, | |
1024 | |
] | |
}, | |
"max_storage_mem_alloc_cpu/0": 2097.1521 | |
}, | |
{ | |
"avg_time_backward_square": 0.0334, | |
"avg_time_forward_square": 0.0316, | |
"inputs": { | |
"data": [ | |
10000, | |
1 | |
] | |
}, | |
"max_storage_mem_alloc_cpu/0": 20.0 | |
}, | |
{ | |
"avg_time_backward_square": 0.0602, | |
"avg_time_forward_square": 0.0716, | |
"inputs": { | |
"data": [ | |
10000, | |
100 | |
] | |
}, | |
"max_storage_mem_alloc_cpu/0": 2000.0 | |
} | |
], | |
"stop_gradient": [ | |
{ | |
"inputs": { | |
"data": [ | |
1024, | |
1024 | |
] | |
}, | |
"max_storage_mem_alloc_cpu/0": 2097.1521 | |
}, | |
{ | |
"inputs": { | |
"data": [ | |
10000, | |
1 | |
] | |
}, | |
"max_storage_mem_alloc_cpu/0": 20.0 | |
}, | |
{ | |
"inputs": { | |
"data": [ | |
10000, | |
100 | |
] | |
}, | |
"max_storage_mem_alloc_cpu/0": 2000.0 | |
} | |
], | |
"sum": [ | |
{ | |
"avg_time_backward_sum": 0.8277, | |
"avg_time_forward_sum": 0.8687, | |
"inputs": { | |
"axis": 0, | |
"data": [ | |
1024, | |
1024 | |
] | |
}, | |
"max_storage_mem_alloc_cpu/0": 4.096 | |
}, | |
{ | |
"avg_time_backward_sum": 0.0373, | |
"avg_time_forward_sum": 0.1687, | |
"inputs": { | |
"axis": 0, | |
"data": [ | |
10000, | |
1 | |
] | |
}, | |
"max_storage_mem_alloc_cpu/0": 0.002 | |
}, | |
{ | |
"avg_time_backward_sum": 0.7963, | |
"avg_time_forward_sum": 0.8004, | |
"inputs": { | |
"axis": 0, | |
"data": [ | |
10000, | |
100 | |
] | |
}, | |
"max_storage_mem_alloc_cpu/0": 0.2 | |
} | |
], | |
"sum_axis": [ | |
{ | |
"avg_time_forward_sum_axis": 0.8525, | |
"inputs": { | |
"axis": 0, | |
"data": [ | |
1024, | |
1024 | |
] | |
}, | |
"max_storage_mem_alloc_cpu/0": 2.048 | |
}, | |
{ | |
"avg_time_forward_sum_axis": 0.1649, | |
"inputs": { | |
"axis": 0, | |
"data": [ | |
10000, | |
1 | |
] | |
}, | |
"max_storage_mem_alloc_cpu/0": 0.004 | |
}, | |
{ | |
"avg_time_forward_sum_axis": 0.6077, | |
"inputs": { | |
"axis": 0, | |
"data": [ | |
10000, | |
100 | |
] | |
}, | |
"max_storage_mem_alloc_cpu/0": 0.2 | |
} | |
], | |
"swapaxes": [ | |
{ | |
"avg_time_forward_swapaxes": 0.8545, | |
"inputs": { | |
"data": [ | |
1024, | |
1024 | |
], | |
"dim1": 0, | |
"dim2": 1 | |
}, | |
"max_storage_mem_alloc_cpu/0": 2097.1521 | |
}, | |
{ | |
"avg_time_forward_swapaxes": 0.0395, | |
"inputs": { | |
"data": [ | |
10000, | |
1 | |
], | |
"dim1": 0, | |
"dim2": 1 | |
}, | |
"max_storage_mem_alloc_cpu/0": 20.0 | |
}, | |
{ | |
"avg_time_forward_swapaxes": 0.8611, | |
"inputs": { | |
"data": [ | |
10000, | |
100 | |
], | |
"dim1": 0, | |
"dim2": 1 | |
}, | |
"max_storage_mem_alloc_cpu/0": 4000.0 | |
} | |
], | |
"tan": [ | |
{ | |
"avg_time_backward_tan": 0.0818, | |
"avg_time_forward_tan": 1.1838, | |
"inputs": { | |
"data": [ | |
1024, | |
1024 | |
] | |
}, | |
"max_storage_mem_alloc_cpu/0": 2097.1521 | |
}, | |
{ | |
"avg_time_backward_tan": 0.0257, | |
"avg_time_forward_tan": 0.0341, | |
"inputs": { | |
"data": [ | |
10000, | |
1 | |
] | |
}, | |
"max_storage_mem_alloc_cpu/0": 20.0 | |
}, | |
{ | |
"avg_time_backward_tan": 0.079, | |
"avg_time_forward_tan": 1.1179, | |
"inputs": { | |
"data": [ | |
10000, | |
100 | |
] | |
}, | |
"max_storage_mem_alloc_cpu/0": 2000.0 | |
} | |
], | |
"tanh": [ | |
{ | |
"avg_time_backward_tanh": 0.0791, | |
"avg_time_forward_tanh": 1.4415, | |
"inputs": { | |
"data": [ | |
1024, | |
1024 | |
] | |
}, | |
"max_storage_mem_alloc_cpu/0": 4194.3042 | |
}, | |
{ | |
"avg_time_backward_tanh": 0.0279, | |
"avg_time_forward_tanh": 0.0483, | |
"inputs": { | |
"data": [ | |
10000, | |
1 | |
] | |
}, | |
"max_storage_mem_alloc_cpu/0": 20.0 | |
}, | |
{ | |
"avg_time_backward_tanh": 0.08, | |
"avg_time_forward_tanh": 1.3766, | |
"inputs": { | |
"data": [ | |
10000, | |
100 | |
] | |
}, | |
"max_storage_mem_alloc_cpu/0": 2000.0 | |
} | |
], | |
"topk": [ | |
{ | |
"avg_time_forward_topk": 17.5967, | |
"inputs": { | |
"axis": 0, | |
"data": [ | |
1024, | |
1024 | |
], | |
"k": 1 | |
}, | |
"max_storage_mem_alloc_cpu/0": 2.048 | |
}, | |
{ | |
"avg_time_forward_topk": 0.2751, | |
"inputs": { | |
"axis": 0, | |
"data": [ | |
10000, | |
1 | |
], | |
"k": 1 | |
}, | |
"max_storage_mem_alloc_cpu/0": 0.002 | |
}, | |
{ | |
"avg_time_forward_topk": 15.5901, | |
"inputs": { | |
"axis": 0, | |
"data": [ | |
10000, | |
100 | |
], | |
"k": 1 | |
}, | |
"max_storage_mem_alloc_cpu/0": 0.2 | |
} | |
], | |
"transpose": [ | |
{ | |
"avg_time_forward_transpose": 0.2382, | |
"inputs": { | |
"data": [ | |
1024, | |
1024 | |
] | |
}, | |
"max_storage_mem_alloc_cpu/0": 2097.1521 | |
}, | |
{ | |
"avg_time_forward_transpose": 0.0415, | |
"inputs": { | |
"data": [ | |
10000, | |
1 | |
] | |
}, | |
"max_storage_mem_alloc_cpu/0": 20.0 | |
}, | |
{ | |
"avg_time_forward_transpose": 0.216, | |
"inputs": { | |
"data": [ | |
10000, | |
100 | |
] | |
}, | |
"max_storage_mem_alloc_cpu/0": 2000.0 | |
} | |
], | |
"trunc": [ | |
{ | |
"avg_time_forward_trunc": 0.4083, | |
"inputs": { | |
"data": [ | |
1024, | |
1024 | |
] | |
}, | |
"max_storage_mem_alloc_cpu/0": 4194.3042 | |
}, | |
{ | |
"avg_time_forward_trunc": 0.1285, | |
"inputs": { | |
"data": [ | |
10000, | |
1 | |
] | |
}, | |
"max_storage_mem_alloc_cpu/0": 20.0 | |
}, | |
{ | |
"avg_time_forward_trunc": 0.3011, | |
"inputs": { | |
"data": [ | |
10000, | |
100 | |
] | |
}, | |
"max_storage_mem_alloc_cpu/0": 4000.0 | |
} | |
], | |
"zeros_like": [ | |
{ | |
"avg_time_forward_zeros_like": 0.0393, | |
"inputs": { | |
"data": [ | |
1024, | |
1024 | |
] | |
}, | |
"max_storage_mem_alloc_cpu/0": 4194.3042 | |
}, | |
{ | |
"avg_time_forward_zeros_like": 0.0161, | |
"inputs": { | |
"data": [ | |
10000, | |
1 | |
] | |
}, | |
"max_storage_mem_alloc_cpu/0": 20.0 | |
}, | |
{ | |
"avg_time_forward_zeros_like": 0.0367, | |
"inputs": { | |
"data": [ | |
10000, | |
100 | |
] | |
}, | |
"max_storage_mem_alloc_cpu/0": 2000.0 | |
} | |
] | |
} |
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment