ChaiBapchya/mxnet_opperf_cpu_array_manipulation.md

## mxnet_opperf_cpu_array_manipulation.md

      
    Raw
  

              mxnet_opperf_cpu_array_manipulation.md
            
          
    Runtime Features


BLAS_APPLE : ✖ BLAS_APPLE
BLAS_ATLAS : ✖ BLAS_ATLAS
BLAS_MKL : ✖ BLAS_MKL
BLAS_OPEN : ✔ BLAS_OPEN
CAFFE : ✖ CAFFE
CPU_AVX : ✔ CPU_AVX
CPU_AVX2 : ✖ CPU_AVX2
CPU_SSE : ✔ CPU_SSE
CPU_SSE2 : ✔ CPU_SSE2
CPU_SSE3 : ✔ CPU_SSE3
CPU_SSE4A : ✖ CPU_SSE4A
CPU_SSE4_1 : ✔ CPU_SSE4_1
CPU_SSE4_2 : ✔ CPU_SSE4_2
CUDA : ✖ CUDA
CUDA_RTC : ✖ CUDA_RTC
CUDNN : ✖ CUDNN
CXX14 : ✖ CXX14
DEBUG : ✖ DEBUG
DIST_KVSTORE : ✖ DIST_KVSTORE
F16C : ✔ F16C
INT64_TENSOR_SIZE : ✖ INT64_TENSOR_SIZE
JEMALLOC : ✖ JEMALLOC
LAPACK : ✔ LAPACK
MKLDNN : ✔ MKLDNN
NCCL : ✖ NCCL
OPENCV : ✔ OPENCV
OPENMP : ✔ OPENMP
PROFILER : ✖ PROFILER
SIGNAL_HANDLER : ✔ SIGNAL_HANDLER
SSE : ✖ SSE
TENSORRT : ✖ TENSORRT
TVM_OP : ✖ TVM_OP

Benchmark Results


Operator
Inputs
Max Mem Usage (Storage) (Bytes)
Avg Forward Time (ms)
Avg. Backward Time (ms)


Activation
{'data': (1024, 1024), 'act_type': 'relu'}
1564139.25
0.0566
0.0686


Activation
{'data': (1024, 1024), 'act_type': 'sigmoid'}
1566236.375
0.506
0.8792


Activation
{'data': (1024, 1024), 'act_type': 'softrelu'}
1568333.5
0.1783
0.8208


Activation
{'data': (1024, 1024), 'act_type': 'softsign'}
1570430.625
0.3635
0.0677


Activation
{'data': (1024, 1024), 'act_type': 'tanh'}
1572527.75
0.2232
1.7112


BatchNorm
{'data': (32, 3, 256, 256), 'gamma': (3,), 'beta': (3,), 'moving_mean': (3,), 'moving_var': (3,), 'eps': 1e-08, 'axis': 1}
416853.1562
2.6452
1.378


BatchNorm
{'data': (32, 3, 10000, 10), 'gamma': (3,), 'beta': (3,), 'moving_mean': (3,), 'moving_var': (3,), 'eps': 1e-08, 'axis': 1}
442670.25
6.2914
1.9779


BilinearSampler
{'data': (32, 2, 256, 256), 'grid': (32, 2, 256, 256)}
25165.8242
103.9467
116.2548


BlockGrad
{'data': (1024, 1024)}
4194.3042
0.4
---


BlockGrad
{'data': (10000, 1)}
40.0
0.0101
---


BlockGrad
{'data': (10000, 100)}
2000.0
0.3253
---


CTCLoss
{'data': (1024, 100, 100), 'label': (100, 100)}
5406962.5
60.1745
---


Convolution
{'data': (32, 3, 64, 64), 'weight': (64, 3, 3, 3), 'bias': (64,), 'kernel': (3, 3), 'stride': (1, 1), 'dilate': (1, 1), 'pad': (0, 0), 'num_filter': 64, 'layout': 'NCHW'}
4707539.5
2.1701
4.1398


Correlation
{'data1': (32, 3, 256, 256), 'data2': (32, 3, 256, 256), 'kernel_size': 3, 'max_displacement': 2, 'stride1': 2, 'stride2': 2}
475636.0625
491.1135
2688.7466


Correlation
{'data1': (32, 3, 10000, 10), 'data2': (32, 3, 10000, 10), 'kernel_size': 3, 'max_displacement': 2, 'stride1': 2, 'stride2': 2}
491383.3438
353.6813
1713.1298


Custom
{'args': [(1024, 1024)], 'op_type': 'CustomAddOne'}
5530295.0
0.0097
0.0099


Custom
{'args': [(10000, 1)], 'op_type': 'CustomAddOne'}
5530295.0
0.0113
0.0093


Custom
{'args': [(10000, 10)], 'op_type': 'CustomAddOne'}
5530295.0
0.0101
0.0106


Deconvolution
{'data': (32, 3, 64, 64), 'weight': (3, 64, 3, 3), 'bias': (64,), 'kernel': (3, 3), 'stride': (1, 1), 'dilate': (1, 1), 'pad': (0, 0), 'num_filter': 64, 'no_bias': False, 'layout': 'NCHW'}
5386482.0
25.3685
24.9116


Dropout
{'data': (32, 3, 256, 256), 'p': 1, 'mode': 'always', 'axes': [0, 1]}
485106.5938
1.4163
0.94


Dropout
{'data': (10000, 10), 'p': 1, 'mode': 'always', 'axes': [0, 1]}
472792.625
0.1068
0.0812


ElementWiseSum
{'args': (1024, 1024)}
2097.1521
0.0934
---


Embedding
{'data': (1024, 1024), 'weight': (3, 4), 'input_dim': 3, 'output_dim': 4, 'dtype': 'float32', 'sparse_grad': False}
497758.4375
0.4161
---


Embedding
{'data': (10000, 1), 'weight': (3, 4), 'input_dim': 3, 'output_dim': 4, 'dtype': 'int32', 'sparse_grad': False}
489529.8125
0.0385
---


Embedding
{'data': (10000, 100), 'weight': (3, 4), 'input_dim': 3, 'output_dim': 4, 'dtype': 'float32', 'sparse_grad': False}
505449.8125
0.4026
---


FullyConnected
{'data': (32, 3, 256, 256), 'weight': (64, 196608), 'bias': (64,), 'num_hidden': 64, 'flatten': True}
497462.125
1.7125
4.8088


FullyConnected
{'data': (32, 3, 10000, 10), 'weight': (64, 10), 'bias': (64,), 'num_hidden': 64, 'flatten': False}
743222.125
44.2965
76.7286


GridGenerator
{'data': (32, 2, 256, 256), 'transform_type': 'warp', 'target_shape': (256, 6)}
17039.3594
10.911
8.744


GridGenerator
{'data': (256, 6), 'transform_type': 'affine', 'target_shape': (256, 6)}
1582.08
0.5582
1.5901


GroupNorm
{'data': (32, 3, 256, 256), 'gamma': (1,), 'beta': (1,), 'num_groups': 1, 'eps': 1e-08}
658091.0
11.7595
175.6284


GroupNorm
{'data': (32, 10, 10000, 10), 'gamma': (10,), 'beta': (10,), 'num_groups': 10, 'eps': 1e-08}
895288.6875
52.0266
255.2737


InstanceNorm
{'data': (32, 3, 256, 256), 'gamma': (3,), 'beta': (3,), 'eps': 1e-08}
869036.5
11.2944
184.461


InstanceNorm
{'data': (32, 3, 10000, 10), 'gamma': (3,), 'beta': (3,), 'eps': 1e-08}
894853.625
24.1254
281.2478


L2Normalization
{'data': (32, 3, 256, 256), 'eps': 1e-08, 'mode': 'channel'}
917596.25
4.2096
9.7036


L2Normalization
{'data': (32, 3, 256, 256), 'eps': 1e-08, 'mode': 'instance'}
925984.9375
0.7249
1.4459


L2Normalization
{'data': (32, 3, 256, 256), 'eps': 1e-08, 'mode': 'spatial'}
938568.0
0.6508
7.5462


LRN
{'data': (32, 3, 256, 256), 'alpha': 0.001, 'beta': 0.2, 'nsize': 3}
963733.625
22.1928
53.8383


LRN
{'data': (32, 3, 10000, 10), 'alpha': 0.001, 'beta': 0.2, 'nsize': 3}
989550.6875
46.5718
94.5082


LayerNorm
{'data': (32, 3, 256, 256), 'gamma': (32,), 'beta': (32,), 'axis': 0, 'eps': 1e-08}
1008885.875
13.4946
22.2397


LayerNorm
{'data': (32, 3, 10000, 10), 'gamma': (32,), 'beta': (32,), 'axis': 0, 'eps': 1e-08}
1048485.875
20.7723
33.9548


LeakyReLU
{'data': (1024, 1024), 'act_type': 'leaky'}
1576722.125
0.0538
0.0649


LeakyReLU
{'data': (1024, 1024), 'act_type': 'elu'}
1578819.25
0.1205
0.76


LeakyReLU
{'data': (1024, 1024), 'act_type': 'selu'}
1580916.375
0.7919
0.3222


LeakyReLU
{'data': (1024, 1024), 'act_type': 'gelu'}
1583013.625
0.296
2.1755


LinearRegressionOutput
{'data': (32, 3, 256, 256), 'label': (32, 3, 256, 256), 'grad_scale': 0.5}
1065834.625
1.5131
---


LinearRegressionOutput
{'data': (32, 3, 10000, 10), 'label': (32, 3, 10000, 10), 'grad_scale': 0.5}
1091651.625
5.034
---


LogisticRegressionOutput
{'data': (32, 3, 256, 256), 'label': (32, 3, 256, 256), 'grad_scale': 0.5}
1110200.375
4.4785
---


LogisticRegressionOutput
{'data': (32, 3, 10000, 10), 'label': (32, 3, 10000, 10), 'grad_scale': 0.5}
1136017.5
9.3099
---


MAERegressionOutput
{'data': (32, 3, 256, 256), 'label': (32, 3, 256, 256), 'grad_scale': 0.5}
1154566.25
1.2484
---


MAERegressionOutput
{'data': (32, 3, 10000, 10), 'label': (32, 3, 10000, 10), 'grad_scale': 0.5}
1180383.375
4.9788
---


MakeLoss
{'data': (1024, 1024), 'grad_scale': 0.5, 'normalization': 'batch'}
5392774.0
0.0569
0.0411


MakeLoss
{'data': (10000, 1), 'grad_scale': 0.5, 'normalization': 'batch'}
5390716.5
0.0356
0.031


MakeLoss
{'data': (10000, 100), 'grad_scale': 0.5, 'normalization': 'batch'}
5394696.5
0.0611
0.0398


Pooling
{'data': (32, 3, 64, 64), 'kernel': (3, 3), 'pool_type': 'sum', 'global_pool': 1, 'stride': (1, 1), 'pad': (0, 0)}
1749846.0
0.5543
0.2783


RNN
{'data': (32, 4, 4), 'parameters': (7,), 'state': (1, 4, 1), 'state_size': 1, 'num_layers': 1, 'mode': 'rnn_relu', 'p': 0.5}
1161186.0
0.9437
1.3606


RNN
{'data': (512, 10000, 10), 'parameters': (104,), 'state': (2, 10000, 4), 'state_size': 4, 'num_layers': 2, 'mode': 'rnn_tanh', 'p': 0.5}
1284088.375
434.3607
753.2392


ROIPooling
{'data': (32, 3, 64, 64), 'rois': (32, 5), 'pooled_size': (2, 2), 'spatial_scale': 0.5}
1749851.875
0.541
0.0435


SVMOutput
{'data': (32, 3, 256, 256), 'label': (32, 3, 256), 'margin': 0.5, 'regularization_coefficient': 0.5}
1321837.125
2.5928
29.7243


SVMOutput
{'data': (32, 3, 10000, 10), 'label': (32, 3, 10000), 'margin': 0.5, 'regularization_coefficient': 0.5}
1347654.25
5.5232
64.2004


SequenceLast
{'data': (1024, 1024), 'axis': 0}
5530301.0
0.049
---


SequenceLast
{'data': (10000, 1), 'axis': 0}
5530299.0
0.0483
---


SequenceLast
{'data': (10000, 100), 'axis': 0}
5530299.5
0.0477
---


SequenceMask
{'data': (1024, 1024), 'axis': 0}
5536590.5
0.5358
0.2289


SequenceMask
{'data': (10000, 1), 'axis': 0}
5534533.5
0.0406
0.0344


SequenceMask
{'data': (10000, 100), 'axis': 0}
5538513.5
0.3242
0.223


SequenceReverse
{'data': (1024, 1024), 'axis': 0}
5542805.0
1.085
1.0869


SequenceReverse
{'data': (10000, 1), 'axis': 0}
5540747.5
0.0704
0.0329


SequenceReverse
{'data': (10000, 100), 'axis': 0}
5544727.5
1.3211
1.2445


Softmax
{'data': (1024, 1024), 'label': (1024, 1024), 'grad_scale': 0.5, 'normalization': 'batch'}
1587207.875
0.101
---


Softmax
{'data': (10000, 1), 'label': (10000, 1), 'grad_scale': 0.5, 'normalization': 'batch'}
1585150.75
0.1121
---


Softmax
{'data': (10000, 100), 'label': (10000, 100), 'grad_scale': 0.5, 'normalization': 'batch'}
1589130.75
0.2125
---


SoftmaxActivation
{'data': (1024, 1024)}
1593422.125
0.7383
1.1597


SoftmaxActivation
{'data': (10000, 1)}
1591365.0
0.0376
0.0678


SoftmaxActivation
{'data': (10000, 100)}
1595345.0
0.7143
1.1253


SoftmaxOutput
{'data': (32, 3, 256, 256), 'label': (32, 3, 256), 'grad_scale': 0.5, 'normalization': 'batch'}
1366203.0
4.5316
0.5743


SoftmaxOutput
{'data': (32, 3, 10000, 10), 'label': (32, 3, 10000), 'grad_scale': 0.5, 'normalization': 'batch'}
1392020.0
9.957
1.5397


SpatialTransformer
{'data': (32, 3, 256, 6), 'loc': (32, 6), 'target_shape': (32, 6), 'transform_type': 'affine', 'sampler_type': 'bilinear'}
1372961.0
2.8351
0.5095


SpatialTransformer
{'data': (256, 3, 10000, 6), 'loc': (256, 6), 'target_shape': (256, 6), 'transform_type': 'affine', 'sampler_type': 'bilinear'}
1379257.0
182.0544
53.164


SwapAxis
{'data': (1024, 1024), 'dim1': 0, 'dim2': 1}
248309.8594
1.4315
1.5278


SwapAxis
{'data': (10000, 1), 'dim1': 0, 'dim2': 1}
246252.7188
0.0397
0.0377


SwapAxis
{'data': (10000, 100), 'dim1': 0, 'dim2': 1}
250232.7188
1.2698
1.264


UpSampling
{'args': (32, 3, 256, 256), 'scale': 2, 'sample_type': 'nearest'}
5549906.5
16.3392
3.3291


UpSampling
{'args': (32, 3, 10000, 1), 'scale': 4, 'sample_type': 'nearest'}
5561015.0
10.6011
2.3157


abs
{'data': (1024, 1024)}
2097.1521
0.0434
0.2535


abs
{'data': (10000, 1)}
20.0
0.0139
0.0177


abs
{'data': (10000, 100)}
4000.0
0.0526
0.3456


adam_update
{'weight': (1024, 1024), 'grad': (1024, 1024), 'mean': (1024, 1024), 'var': (1024, 1024), 'lr': 0.1, 'beta1': 0.1, 'beta2': 0.1, 'epsilon': 1e-08, 'wd': 0.1, 'rescale_grad': 0.4, 'lazy_update': 0}
4698086.0
0.4284
---


adam_update
{'weight': (10000, 1), 'grad': (10000, 1), 'mean': (10000, 1), 'var': (10000, 1), 'lr': 0.5, 'beta1': 0.5, 'beta2': 0.5, 'epsilon': 1e-08, 'wd': 0.5, 'rescale_grad': 0.4, 'lazy_update': 0}
4696028.5
0.0406
---


adam_update
{'weight': (10000, 100), 'grad': (10000, 100), 'mean': (10000, 100), 'var': (10000, 100), 'lr': 0.9, 'beta1': 0.9, 'beta2': 0.9, 'epsilon': 1e-08, 'wd': 0.9, 'rescale_grad': 0.4, 'lazy_update': 0}
4700008.5
0.7697
---


add_n
{'args': [(1024, 1024)]}
5398911.5
0.0352
---


add_n
{'args': [(10000, 1)]}
5398911.5
0.0364
---


add_n
{'args': [(10000, 10)]}
5398911.5
0.0353
---


all_finite
{'data': (1024, 1024)}
5542727.5
0.092
---


all_finite
{'data': (10000, 1)}
5542727.5
0.0499
---


all_finite
{'data': (10000, 100)}
5542727.5
0.0926
---


amp_cast
{'data': (1024, 1024), 'dtype': 'float32'}
2097.1521
0.0847
0.0733


amp_cast
{'data': (10000, 1), 'dtype': 'int32'}
20.0
0.352
0.3344


amp_cast
{'data': (10000, 100), 'dtype': 'float32'}
2000.0
0.0872
0.0756


amp_multicast
{'args': [(1024, 1024)], 'num_outputs': 1}
0.004
0.0913
0.0808


amp_multicast
{'args': [(10000, 1)], 'num_outputs': 1}
0.008
0.0538
0.0504


arccos
{'data': (1024, 1024)}
4194.3042
1.1368
0.9571


arccos
{'data': (10000, 1)}
20.0
0.0376
0.0342


arccos
{'data': (10000, 100)}
4000.0
0.813
0.6066


arccosh
{'data': (1024, 1024)}
4194.3042
0.6973
0.7103


arccosh
{'data': (10000, 1)}
40.0
0.0459
0.0333


arccosh
{'data': (10000, 100)}
4000.0
0.7189
0.6841


arcsin
{'data': (1024, 1024)}
4194.3042
0.7536
0.6546


arcsin
{'data': (10000, 1)}
20.0
0.0469
0.0325


arcsin
{'data': (10000, 100)}
4000.0
0.7938
0.7792


arcsinh
{'data': (1024, 1024)}
4194.3042
1.7109
0.5158


arcsinh
{'data': (10000, 1)}
40.0
0.0449
0.035


arcsinh
{'data': (10000, 100)}
2000.0
1.5514
0.4768


arctan
{'data': (1024, 1024)}
2097.1521
0.9648
0.083


arctan
{'data': (10000, 1)}
20.0
0.0377
0.036


arctan
{'data': (10000, 100)}
4000.0
0.9344
0.0813


arctanh
{'data': (1024, 1024)}
2097.1521
1.3064
0.0814


arctanh
{'data': (10000, 1)}
20.0
0.042
0.0356


arctanh
{'data': (10000, 100)}
2000.0
1.2438
0.0809


argmax
{'data': (1024, 1024), 'axis': 0}
223297.25
7.6787
---


argmax
{'data': (10000, 1), 'axis': 0}
223295.2031
0.0867
---


argmax
{'data': (10000, 100), 'axis': 0}
223295.6094
6.6238
---


argmax_channel
{'data': (1024, 1024)}
2.048
0.2884
---


argmax_channel
{'data': (10000, 1)}
20.0
0.0385
---


argmax_channel
{'data': (10000, 100)}
20.0
0.303
---


argmin
{'data': (1024, 1024), 'axis': 0}
223301.5469
7.6812
---


argmin
{'data': (10000, 1), 'axis': 0}
223299.5
0.0861
---


argmin
{'data': (10000, 100), 'axis': 0}
223299.9062
6.6216
---


argsort
{'data': (1024, 1024), 'axis': 0, 'dtype': 'float32'}
229591.1562
35.6613
---


argsort
{'data': (10000, 1), 'axis': 0, 'dtype': 'int32'}
227534.0
1.1665
---


argsort
{'data': (10000, 100), 'axis': 0, 'dtype': 'float32'}
231514.0
31.1643
---


batch_dot
{'lhs': (32, 1024, 1024), 'rhs': (32, 1024, 1024)}
134217.7344
550.3716
---


batch_dot
{'lhs': (32, 1000, 10), 'rhs': (32, 1000, 10), 'transpose_b': True}
128000.0
380.5898
---


batch_dot
{'lhs': (32, 1000, 1), 'rhs': (32, 100, 1000), 'transpose_a': True, 'transpose_b': True}
12.8
4.2182
---


broadcast_add
{'lhs': (1024, 1024), 'rhs': (1024, 1024)}
2097.1521
0.0959
0.1302


broadcast_add
{'lhs': (10000, 10), 'rhs': (10000, 10)}
400.0
0.0573
0.0481


broadcast_add
{'lhs': (10000, 1), 'rhs': (10000, 1)}
20.0
0.0328
0.036


broadcast_axes
{'data': (1, 1024), 'axis': 0, 'size': 2}
293046.875
0.031
---


broadcast_axes
{'data': (1, 1), 'axis': 0, 'size': 2}
293042.7812
0.0306
---


broadcast_axes
{'data': (1, 100), 'axis': 0, 'size': 2}
293043.5938
0.0311
---


broadcast_axis
{'data': (1, 1024), 'axis': 0, 'size': 2}
293055.4688
0.0358
---


broadcast_axis
{'data': (1, 1), 'axis': 0, 'size': 2}
293051.375
0.0314
---


broadcast_axis
{'data': (1, 100), 'axis': 0, 'size': 2}
293052.1875
0.0331
---


broadcast_div
{'lhs': (1024, 1024), 'rhs': (1024, 1024)}
4194.3042
0.0857
0.1299


broadcast_div
{'lhs': (10000, 10), 'rhs': (10000, 10)}
400.0
0.0335
0.0459


broadcast_div
{'lhs': (10000, 1), 'rhs': (10000, 1)}
20.0
0.0268
0.0423


broadcast_equal
{'lhs': (1024, 1024), 'rhs': (1024, 1024)}
2097.1521
0.0742
---


broadcast_equal
{'lhs': (10000, 10), 'rhs': (10000, 10)}
200.0
0.0326
---


broadcast_equal
{'lhs': (10000, 1), 'rhs': (10000, 1)}
20.0
0.0263
---


broadcast_greater
{'lhs': (1024, 1024), 'rhs': (1024, 1024)}
2097.1521
0.0766
---


broadcast_greater
{'lhs': (10000, 10), 'rhs': (10000, 10)}
200.0
0.0328
---


broadcast_greater
{'lhs': (10000, 1), 'rhs': (10000, 1)}
20.0
0.0283
---


broadcast_greater_equal
{'lhs': (1024, 1024), 'rhs': (1024, 1024)}
2097.1521
0.0753
---


broadcast_greater_equal
{'lhs': (10000, 10), 'rhs': (10000, 10)}
200.0
0.0326
---


broadcast_greater_equal
{'lhs': (10000, 1), 'rhs': (10000, 1)}
20.0
0.0271
---


broadcast_hypot
{'lhs': (1024, 1024), 'rhs': (1024, 1024)}
2097.1521
0.4796
1.0338


broadcast_hypot
{'lhs': (10000, 10), 'rhs': (10000, 10)}
400.0
0.0735
0.1301


broadcast_hypot
{'lhs': (10000, 1), 'rhs': (10000, 1)}
20.0
0.0393
0.0522


broadcast_lesser
{'lhs': (1024, 1024), 'rhs': (1024, 1024)}
2097.1521
0.0756
---


broadcast_lesser
{'lhs': (10000, 10), 'rhs': (10000, 10)}
400.0
0.0374
---


broadcast_lesser
{'lhs': (10000, 1), 'rhs': (10000, 1)}
20.0
0.0317
---


broadcast_lesser_equal
{'lhs': (1024, 1024), 'rhs': (1024, 1024)}
4194.3042
0.0995
---


broadcast_lesser_equal
{'lhs': (10000, 10), 'rhs': (10000, 10)}
400.0
0.0371
---


broadcast_lesser_equal
{'lhs': (10000, 1), 'rhs': (10000, 1)}
40.0
0.0311
---


broadcast_like
{'lhs': (1024, 1024), 'rhs': (1024, 1024)}
299343.25
0.7073
---


broadcast_like
{'lhs': (10000, 10), 'rhs': (10000, 10)}
297646.0938
0.1026
---


broadcast_like
{'lhs': (10000, 1), 'rhs': (10000, 1)}
297486.0938
0.0417
---


broadcast_logical_and
{'lhs': (1024, 1024), 'rhs': (1024, 1024)}
2097.1521
0.0797
---


broadcast_logical_and
{'lhs': (10000, 10), 'rhs': (10000, 10)}
200.0
0.0378
---


broadcast_logical_and
{'lhs': (10000, 1), 'rhs': (10000, 1)}
40.0
0.0393
---


broadcast_logical_or
{'lhs': (1024, 1024), 'rhs': (1024, 1024)}
4194.3042
0.1025
---


broadcast_logical_or
{'lhs': (10000, 10), 'rhs': (10000, 10)}
200.0
0.0363
---


broadcast_logical_or
{'lhs': (10000, 1), 'rhs': (10000, 1)}
40.0
0.0415
---


broadcast_logical_xor
{'lhs': (1024, 1024), 'rhs': (1024, 1024)}
2097.1521
0.1622
---


broadcast_logical_xor
{'lhs': (10000, 10), 'rhs': (10000, 10)}
400.0
0.0418
---


broadcast_logical_xor
{'lhs': (10000, 1), 'rhs': (10000, 1)}
20.0
0.0463
---


broadcast_maximum
{'lhs': (1024, 1024), 'rhs': (1024, 1024)}
2097.1521
0.0892
0.1256


broadcast_maximum
{'lhs': (10000, 10), 'rhs': (10000, 10)}
200.0
0.0355
0.0498


broadcast_maximum
{'lhs': (10000, 1), 'rhs': (10000, 1)}
20.0
0.0315
0.0506


broadcast_minimum
{'lhs': (1024, 1024), 'rhs': (1024, 1024)}
4194.3042
0.1152
0.1305


broadcast_minimum
{'lhs': (10000, 10), 'rhs': (10000, 10)}
200.0
0.0375
0.051


broadcast_minimum
{'lhs': (10000, 1), 'rhs': (10000, 1)}
20.0
0.0325
0.052


broadcast_minus
{'lhs': (1024, 1024), 'rhs': (1024, 1024)}
2097.1521
0.0753
---


broadcast_minus
{'lhs': (10000, 10), 'rhs': (10000, 10)}
400.0
0.0389
---


broadcast_minus
{'lhs': (10000, 1), 'rhs': (10000, 1)}
40.0
0.0299
---


broadcast_mod
{'lhs': (1024, 1024), 'rhs': (1024, 1024)}
4194.3042
1.9691
0.2051


broadcast_mod
{'lhs': (10000, 10), 'rhs': (10000, 10)}
200.0
0.2143
0.0547


broadcast_mod
{'lhs': (10000, 1), 'rhs': (10000, 1)}
40.0
0.047
0.045


broadcast_mul
{'lhs': (1024, 1024), 'rhs': (1024, 1024)}
2097.1521
0.0667
0.1019


broadcast_mul
{'lhs': (10000, 10), 'rhs': (10000, 10)}
200.0
0.0382
0.0509


broadcast_mul
{'lhs': (10000, 1), 'rhs': (10000, 1)}
20.0
0.0321
0.0404


broadcast_not_equal
{'lhs': (1024, 1024), 'rhs': (1024, 1024)}
4194.3042
0.0955
---


broadcast_not_equal
{'lhs': (10000, 10), 'rhs': (10000, 10)}
400.0
0.0375
---


broadcast_not_equal
{'lhs': (10000, 1), 'rhs': (10000, 1)}
40.0
0.0326
---


broadcast_plus
{'lhs': (1024, 1024), 'rhs': (1024, 1024)}
2097.1521
0.0713
---


broadcast_plus
{'lhs': (10000, 10), 'rhs': (10000, 10)}
400.0
0.0376
---


broadcast_plus
{'lhs': (10000, 1), 'rhs': (10000, 1)}
20.0
0.0289
---


broadcast_power
{'lhs': (1024, 1024), 'rhs': (1024, 1024)}
4194.3042
2.4365
5.7618


broadcast_power
{'lhs': (10000, 10), 'rhs': (10000, 10)}
200.0
0.3054
0.6323


broadcast_power
{'lhs': (10000, 1), 'rhs': (10000, 1)}
20.0
0.0534
0.0925


broadcast_sub
{'lhs': (1024, 1024), 'rhs': (1024, 1024)}
2097.1521
0.0696
0.0863


broadcast_sub
{'lhs': (10000, 10), 'rhs': (10000, 10)}
200.0
0.0386
0.0498


broadcast_sub
{'lhs': (10000, 1), 'rhs': (10000, 1)}
20.0
0.0304
0.0364


broadcast_to
{'data': (1, 1024), 'shape': (1024, 1024)}
303757.5312
0.711
---


broadcast_to
{'data': (1, 1), 'shape': (10000, 1)}
301700.375
0.0369
---


broadcast_to
{'data': (1, 100), 'shape': (10000, 100)}
305680.375
0.4137
---


cast
{'data': (1024, 1024), 'dtype': 'float32'}
2097.1521
---
0.7171


cast
{'data': (10000, 1), 'dtype': 'int32'}
40.0
---
0.0616


cast
{'data': (10000, 100), 'dtype': 'float32'}
2000.0
---
1.3491


cast_storage
{'data': (1024, 1024), 'stype': 'default'}
5549019.0
0.4026
---


cast_storage
{'data': (10000, 1), 'stype': 'csr'}
5547122.0
0.1022
---


cast_storage
{'data': (10000, 100), 'stype': 'row_sparse'}
5551102.0
0.6637
---


cbrt
{'data': (1024, 1024)}
4194.3042
1.3934
0.0854


cbrt
{'data': (10000, 1)}
20.0
0.045
0.0364


cbrt
{'data': (10000, 100)}
4000.0
1.335
0.0852


ceil
{'data': (1024, 1024)}
341655.75
0.1613
---


ceil
{'data': (10000, 1)}
339598.5938
0.0319
---


ceil
{'data': (10000, 100)}
343578.5938
0.153
---


choose_element_0index
{'data': (1024, 1024), 'index': (1, 1024), 'axis': 0}
4.096
0.0371
---


choose_element_0index
{'data': (10000, 1), 'index': (1, 1), 'axis': 0}
0.004
0.0365
---


choose_element_0index
{'data': (10000, 100), 'index': (1, 100), 'axis': 0}
0.4
0.0362
---


clip
{'data': (1024, 1024), 'a_min': 0.1, 'a_max': 0.9}
5555353.5
0.2876
0.2823


clip
{'data': (10000, 1), 'a_min': 0.1, 'a_max': 0.9}
5553296.5
0.0356
0.0323


clip
{'data': (10000, 100), 'a_min': 0.1, 'a_max': 0.9}
5557276.5
0.2733
0.2671


col2im
{'data': (32, 64, 256), 'output_size': (64, 16, 1), 'kernel': (1, 1, 1), 'stride': (2, 2, 2)}
1387898.625
22.5543
21.4321


col2im
{'data': (32, 64, 256), 'output_size': (32, 8, 1), 'kernel': (1, 1, 1), 'stride': (1, 1, 1)}
1385801.375
21.7266
20.5925


concat
{'args0': '<NDArray 100x100 @cpu(0)>', 'args1': '<NDArray 100x100 @cpu(0)>', 'args2': '<NDArray 100x100 @cpu(0)>'}
372830.125
---
---


cos
{'data': (1024, 1024)}
2097.1521
0.5013
0.5392


cos
{'data': (10000, 1)}
20.0
0.0382
0.0324


cos
{'data': (10000, 100)}
2000.0
0.4034
0.369


cosh
{'data': (1024, 1024)}
2097.1521
0.6531
1.0635


cosh
{'data': (10000, 1)}
20.0
0.0364
0.0355


cosh
{'data': (10000, 100)}
4000.0
0.7704
1.3442


ctc_loss
{'data': (1024, 100, 100), 'label': (100, 100)}
5433657.5
62.4658
---


cumsum
{'a': (1024, 1024), 'axis': 0, 'dtype': 'float32'}
5561568.0
1.1224
---


cumsum
{'a': (1024, 1024), 'axis': 0, 'dtype': 'int32'}
5563665.0
1.576
---


cumsum
{'a': (1024, 1024), 'axis': 0, 'dtype': 'float32'}
5565762.0
1.1237
---


degrees
{'data': (1024, 1024)}
4194.3042
0.063
0.051


degrees
{'data': (10000, 1)}
20.0
0.0275
0.0263


degrees
{'data': (10000, 100)}
2000.0
0.0502
0.0455


depth_to_space
{'data': (1, 4, 2, 4), 'block_size': 2}
248232.9062
0.0708
---


depth_to_space
{'data': (10, 25, 10, 100), 'block_size': 5}
249232.8438
0.8308
---


diag
{'data': (1024, 1024), 'k': 1}
274096.3125
0.0311
0.3142


diag
{'data': (10000, 1), 'k': 1}
---
0.0294
0.0456


diag
{'data': (10000, 100), 'k': 1}
274094.6875
0.0375
0.3087


dot
{'lhs': (1024, 1024), 'rhs': (1024, 1024)}
4194.3042
5.3503
7.048


dot
{'lhs': (1000, 10), 'rhs': (1000, 10), 'transpose_b': True}
2000.0
0.423
4.6241


dot
{'lhs': (1000, 1), 'rhs': (100, 1000), 'transpose_a': True, 'transpose_b': True}
0.2
0.1632
0.2758


elemwise_add
{'lhs': (1024, 1024), 'rhs': (1024, 1024)}
4194.3042
0.1915
---


elemwise_add
{'lhs': (10000, 10), 'rhs': (10000, 10)}
400.0
0.0545
---


elemwise_add
{'lhs': (10000, 1), 'rhs': (10000, 1)}
40.0
0.0526
---


elemwise_div
{'lhs': (1024, 1024), 'rhs': (1024, 1024)}
4194.3042
0.17
---


elemwise_div
{'lhs': (10000, 10), 'rhs': (10000, 10)}
400.0
0.0354
---


elemwise_div
{'lhs': (10000, 1), 'rhs': (10000, 1)}
40.0
0.033
---


elemwise_mul
{'lhs': (1024, 1024), 'rhs': (1024, 1024)}
4194.3042
0.2112
---


elemwise_mul
{'lhs': (10000, 10), 'rhs': (10000, 10)}
400.0
0.0441
---


elemwise_mul
{'lhs': (10000, 1), 'rhs': (10000, 1)}
40.0
0.0315
---


elemwise_sub
{'lhs': (1024, 1024), 'rhs': (1024, 1024)}
4194.3042
0.0944
---


elemwise_sub
{'lhs': (10000, 10), 'rhs': (10000, 10)}
400.0
0.0507
---


elemwise_sub
{'lhs': (10000, 1), 'rhs': (10000, 1)}
40.0
0.0308
---


erf
{'data': (1024, 1024)}
2097.1521
0.9296
0.7832


erf
{'data': (10000, 1)}
20.0
0.0399
0.0349


erf
{'data': (10000, 100)}
2000.0
0.9264
0.7504


erfinv
{'data': (1024, 1024)}
2097.1521
2.5779
0.5366


erfinv
{'data': (10000, 1)}
40.0
0.0601
0.0363


erfinv
{'data': (10000, 100)}
2000.0
2.9925
0.6673


exp
{'data': (1024, 1024)}
2097.1521
0.3765
---


exp
{'data': (10000, 1)}
20.0
0.0361
---


exp
{'data': (10000, 100)}
2000.0
0.3574
---


expand_dims
{'data': (1024, 1024), 'axis': 0}
309971.8438
0.0476
---


expand_dims
{'data': (10000, 1), 'axis': 0}
307914.6875
0.0466
---


expand_dims
{'data': (10000, 100), 'axis': 0}
311894.6875
0.0785
---


expm1
{'data': (1024, 1024)}
2097.1521
0.7082
0.4433


expm1
{'data': (10000, 1)}
20.0
0.0385
0.0339


expm1
{'data': (10000, 100)}
4000.0
0.795
0.5046


fill_element_0index
{'lhs': (1024, 1024), 'mhs': (1024,), 'rhs': (1024,)}
5569956.5
0.105
---


fill_element_0index
{'lhs': (10000, 1), 'mhs': (10000,), 'rhs': (10000,)}
5567899.0
0.0326
---


fill_element_0index
{'lhs': (10000, 100), 'mhs': (10000,), 'rhs': (10000,)}
5571879.0
0.1058
---


fix
{'data': (1024, 1024)}
347870.0625
0.3468
---


fix
{'data': (10000, 1)}
345812.9062
0.0329
---


fix
{'data': (10000, 100)}
349792.9062
0.337
---


flatten
{'data': (1024, 1024)}
4194.3042
0.3266
---


flatten
{'data': (10000, 1)}
40.0
0.0516
---


flatten
{'data': (10000, 100)}
4000.0
0.0738
---


flip
{'data': (1024, 1024), 'axis': 0}
255024.2969
0.4731
---


flip
{'data': (10000, 1), 'axis': 0}
252967.1406
0.0365
---


flip
{'data': (10000, 100), 'axis': 0}
256947.1406
0.4615
---


floor
{'data': (1024, 1024)}
354084.3438
0.1206
---


floor
{'data': (10000, 1)}
352027.2188
0.0289
---


floor
{'data': (10000, 100)}
356007.2188
0.111
---


ftml_update
{'weight': (1024, 1024), 'grad': (1024, 1024), 'd': (1024, 1024), 'v': (1024, 1024), 'z': (1024, 1024), 'lr': 0.1, 'beta1': 0.1, 'beta2': 0.1, 'epsilon': 1e-08, 't': 1, 'wd': 0.1, 'rescale_grad': 0.4, 'clip_grad': -1.0}
4704300.0
1.5838
---


ftml_update
{'weight': (10000, 1), 'grad': (10000, 1), 'd': (10000, 1), 'v': (10000, 1), 'z': (10000, 1), 'lr': 0.5, 'beta1': 0.5, 'beta2': 0.5, 'epsilon': 1e-08, 't': 1, 'wd': 0.5, 'rescale_grad': 0.4, 'clip_grad': -1.0}
4702243.0
0.0561
---


ftml_update
{'weight': (10000, 100), 'grad': (10000, 100), 'd': (10000, 100), 'v': (10000, 100), 'z': (10000, 100), 'lr': 0.9, 'beta1': 0.9, 'beta2': 0.9, 'epsilon': 1e-08, 't': 1, 'wd': 0.9, 'rescale_grad': 0.4, 'clip_grad': -1.0}
4706223.0
1.5817
---


ftrl_update
{'weight': (1024, 1024), 'grad': (1024, 1024), 'z': (1024, 1024), 'n': (1024, 1024), 'lr': 0.1, 'wd': 0.1, 'rescale_grad': 0.4}
4710514.5
1.2678
---


ftrl_update
{'weight': (10000, 1), 'grad': (10000, 1), 'z': (10000, 1), 'n': (10000, 1), 'lr': 0.5, 'wd': 0.5, 'rescale_grad': 0.4}
4708457.5
0.0425
---


ftrl_update
{'weight': (10000, 100), 'grad': (10000, 100), 'z': (10000, 100), 'n': (10000, 100), 'lr': 0.9, 'wd': 0.9, 'rescale_grad': 0.4}
4712437.5
1.2607
---


gamma
{'data': (1024, 1024)}
4194.3042
3.2805
5.8549


gamma
{'data': (10000, 1)}
20.0
0.0714
0.099


gamma
{'data': (10000, 100)}
4000.0
3.2982
5.7524


gammaln
{'data': (1024, 1024)}
4194.3042
25.025
3.1289


gammaln
{'data': (10000, 1)}
20.0
0.2754
0.0555


gammaln
{'data': (10000, 100)}
2000.0
24.0462
2.2594


gather_nd
{'data': (1024, 1024), 'indices': (1, 1)}
235738.75
0.0534
---


gather_nd
{'data': (10000, 1), 'indices': (1, 1)}
235736.7031
0.0534
---


gather_nd
{'data': (10000, 100), 'indices': (1, 1)}
235737.1094
0.0539
---


hard_sigmoid
{'data': (1024, 1024)}
1599636.5
0.1093
0.128


hard_sigmoid
{'data': (10000, 1)}
1597579.375
0.032
0.0269


hard_sigmoid
{'data': (10000, 100)}
1601559.375
0.1088
0.1215


identity
{'data': (1024, 1024)}
2097.1521
0.4072
---


identity
{'data': (10000, 1)}
20.0
0.0244
---


identity
{'data': (10000, 100)}
2000.0
0.3965
---


im2col
{'data': (32, 3, 256, 256), 'kernel': (3,), 'stride': (1,), 'dilate': (1,), 'pad': (1,)}
1385195.25
1.4523
1.353


im2col
{'data': (32, 3, 10000, 10), 'kernel': (3, 3), 'stride': (1, 1), 'dilate': (1, 1), 'pad': (1, 1)}
1730647.75
342.6642
133.3937


khatri_rao
{'args': [(32, 32), (32, 32)]}
0.008
0.0505
---


khatri_rao
{'args': [(64, 64), (64, 64)]}
0.008
0.0522
---


lamb_update_phase1
{'weight': (1024, 1024), 'grad': (1024, 1024), 'mean': (1024, 1024), 'var': (1024, 1024), 'beta1': 0.1, 'beta2': 0.1, 'epsilon': 1e-08, 't': 1, 'wd': 0.1, 'rescale_grad': 0.4}
4716729.0
0.9193
---


lamb_update_phase1
{'weight': (10000, 1), 'grad': (10000, 1), 'mean': (10000, 1), 'var': (10000, 1), 'beta1': 0.5, 'beta2': 0.5, 'epsilon': 1e-08, 't': 1, 'wd': 0.5, 'rescale_grad': 0.4}
4714671.5
0.0428
---


lamb_update_phase1
{'weight': (10000, 100), 'grad': (10000, 100), 'mean': (10000, 100), 'var': (10000, 100), 'beta1': 0.9, 'beta2': 0.9, 'epsilon': 1e-08, 't': 1, 'wd': 0.9, 'rescale_grad': 0.4}
4718651.5
2.0237
---


lamb_update_phase2
{'weight': (1024, 1024), 'g': (1024, 1024), 'r1': (1, 1024), 'r2': (1, 1024), 'lr': 0.1}
4722943.0
0.236
---


lamb_update_phase2
{'weight': (10000, 1), 'g': (10000, 1), 'r1': (1, 1), 'r2': (1, 1), 'lr': 0.5}
4720886.0
0.0414
---


lamb_update_phase2
{'weight': (10000, 100), 'g': (10000, 100), 'r1': (1, 100), 'r2': (1, 100), 'lr': 0.9}
4724866.0
0.2239
---


linalg_det
{'A': (1024, 1024)}
5586581.5
15.4087
39.9563


linalg_extractdiag
{'A': (1024, 1024)}
5584488.5
0.0517
0.0562


linalg_extracttrian
{'A': (1024, 1024)}
5587635.0
0.6429
0.6641


linalg_gelqf
{'A': (1024, 1024)}
5599168.5
182.5085
---


linalg_gemm
{'A': (1024, 1024), 'B': (1024, 1024), 'C': (1024, 1024), 'axis': 0}
5601265.5
3.8929
6.7858


linalg_gemm2
{'A': (1024, 1024), 'B': (1024, 1024), 'axis': 0}
5605460.0
2.8579
5.8484


linalg_inverse
{'A': (1024, 1024)}
5609654.0
58.9799
8.6117


linalg_makediag
{'A': (1024, 1024)}
12050008.0
558.6019
4.2716


linalg_maketrian
{'A': (1024, 1035)}
9914966.0
1.304
1.1431


linalg_potrf
{'A': [[1, 0], [0, 1]]}
5584482.0
0.0351
---


linalg_potrf
{'A': [[2, -1, 0], [-1, 2, -1], [0, -1, 2]]}
5584482.0
0.0917
---


linalg_potri
{'A': (1024, 1024)}
9917110.0
20.7041
12.2103


linalg_slogdet
{'A': (1024, 1024)}
9917112.0
17.7089
---


linalg_sumlogdiag
{'A': (1024, 1024)}
9915013.0
0.0604
0.4945


linalg_syrk
{'A': (1024, 1024)}
9921304.0
3.4626
5.865


linalg_trmm
{'A': (1024, 1024), 'B': (1024, 1024)}
9925499.0
2.5562
5.658


linalg_trsm
{'A': (1024, 1024), 'B': (1024, 1024)}
9929693.0
3.4078
7.5423


log
{'data': (1024, 1024)}
2097.1521
0.6218
0.044


log
{'data': (10000, 1)}
40.0
0.0416
0.031


log
{'data': (10000, 100)}
4000.0
0.8078
0.067


log10
{'data': (1024, 1024)}
2097.1521
0.9951
0.0784


log10
{'data': (10000, 1)}
20.0
0.0414
0.0299


log10
{'data': (10000, 100)}
2000.0
0.9594
0.0769


log1p
{'data': (1024, 1024)}
2097.1521
0.9963
0.0909


log1p
{'data': (10000, 1)}
20.0
0.0398
0.0332


log1p
{'data': (10000, 100)}
2000.0
0.9476
0.0786


log2
{'data': (1024, 1024)}
2097.1521
0.829
0.0758


log2
{'data': (10000, 1)}
40.0
0.039
0.0291


log2
{'data': (10000, 100)}
2000.0
0.7962
0.0781


log_softmax
{'data': (1024, 1024), 'axis': 0, 'dtype': 'float16'}
1602705.0
3.269
2.0835


log_softmax
{'data': (10000, 1), 'axis': 0, 'dtype': 'float32'}
1601696.5
0.5576
0.268


log_softmax
{'data': (10000, 100), 'axis': 0, 'dtype': 'float64'}
1609676.5
5.5442
4.5522


logical_not
{'data': (1024, 1024)}
4194.3042
0.1906
---


logical_not
{'data': (10000, 1)}
20.0
0.0294
---


logical_not
{'data': (10000, 100)}
2000.0
0.0566
---


make_loss
{'data': (1024, 1024)}
2097.1521
0.4085
---


make_loss
{'data': (10000, 1)}
20.0
0.0203
---


make_loss
{'data': (10000, 100)}
2000.0
0.3917
---


max
{'data': (1024, 1024), 'axis': 0}
223247.9219
0.7283
0.7584


max
{'data': (10000, 1), 'axis': 0}
223245.875
0.1406
0.038


max
{'data': (10000, 100), 'axis': 0}
223246.2812
0.6682
0.7202


max_axis
{'data': (1024, 1024), 'axis': 0}
223252.2188
0.7647
---


max_axis
{'data': (10000, 1), 'axis': 0}
223250.1719
0.1397
---


max_axis
{'data': (10000, 100), 'axis': 0}
223250.5781
0.7048
---


mean
{'data': (1024, 1024), 'axis': 0}
223256.5156
0.8271
1.4


mean
{'data': (10000, 1), 'axis': 0}
223254.4688
0.242
0.0494


mean
{'data': (10000, 100), 'axis': 0}
223254.875
1.2087
1.3809


min
{'data': (1024, 1024), 'axis': 0}
223262.8594
0.7243
0.7596


min
{'data': (10000, 1), 'axis': 0}
223260.8125
0.1407
0.0372


min
{'data': (10000, 100), 'axis': 0}
223261.2188
0.6795
0.7245


min_axis
{'data': (1024, 1024), 'axis': 0}
223267.1562
0.7068
---


min_axis
{'data': (10000, 1), 'axis': 0}
223265.125
0.139
---


min_axis
{'data': (10000, 100), 'axis': 0}
223265.5156
0.6728
---


moments
{'data': (1024, 1024), 'axes': [0, 1]}
9927596.0
34.0742
---


moments
{'data': (10000, 1), 'axes': [0, 1]}
9927596.0
0.4383
---


moments
{'data': (10000, 100), 'axes': [0, 1]}
9927596.0
30.9714
---


mp_nag_mom_update
{'weight': (1024, 1024), 'grad': (1024, 1024), 'mom': (1024, 1024), 'weight32': (1024, 1024), 'lr': 0.1, 'wd': 0.1, 'rescale_grad': 0.4}
4729157.5
0.3158
---


mp_nag_mom_update
{'weight': (10000, 1), 'grad': (10000, 1), 'mom': (10000, 1), 'weight32': (10000, 1), 'lr': 0.5, 'wd': 0.5, 'rescale_grad': 0.4}
4727100.5
0.0407
---


mp_nag_mom_update
{'weight': (10000, 100), 'grad': (10000, 100), 'mom': (10000, 100), 'weight32': (10000, 100), 'lr': 0.9, 'wd': 0.9, 'rescale_grad': 0.4}
4731080.5
0.2828
---


mp_sgd_mom_update
{'weight': (1024, 1024), 'grad': (1024, 1024), 'mom': (1024, 1024), 'weight32': (1024, 1024), 'lr': 0.1, 'wd': 0.1, 'rescale_grad': 0.4, 'lazy_update': 0}
4735371.5
0.2167
---


mp_sgd_mom_update
{'weight': (10000, 1), 'grad': (10000, 1), 'mom': (10000, 1), 'weight32': (10000, 1), 'lr': 0.5, 'wd': 0.5, 'rescale_grad': 0.4, 'lazy_update': 0}
4733314.5
0.0405
---


mp_sgd_mom_update
{'weight': (10000, 100), 'grad': (10000, 100), 'mom': (10000, 100), 'weight32': (10000, 100), 'lr': 0.9, 'wd': 0.9, 'rescale_grad': 0.4, 'lazy_update': 0}
4737294.5
0.2028
---


mp_sgd_update
{'weight': (1024, 1024), 'grad': (1024, 1024), 'weight32': (1024, 1024), 'lr': 0.1, 'wd': 0.1, 'rescale_grad': 0.4, 'lazy_update': 0}
4741586.0
0.49
---


mp_sgd_update
{'weight': (10000, 1), 'grad': (10000, 1), 'weight32': (10000, 1), 'lr': 0.5, 'wd': 0.5, 'rescale_grad': 0.4, 'lazy_update': 0}
4739529.0
0.0438
---


mp_sgd_update
{'weight': (10000, 100), 'grad': (10000, 100), 'weight32': (10000, 100), 'lr': 0.9, 'wd': 0.9, 'rescale_grad': 0.4, 'lazy_update': 0}
4743509.0
0.1801
---


multi_all_finite
{'args': [(1024, 1024)], 'num_arrays': 1}
5398911.5
0.0459
---


multi_all_finite
{'args': [(10000, 1)], 'num_arrays': 1}
5398911.5
0.0968
---


multi_all_finite
{'args': [(10000, 10)], 'num_arrays': 1}
5398911.5
0.0438
---


multi_lars
{'lrs': (1024, 1024), 'weights_sum_sq': (1024, 1024), 'grads_sum_sq': (1024, 1024), 'wds': (1024, 1024), 'eta': 0.5, 'eps': 1e-08, 'rescale_grad': 0.4}
5576170.5
0.9196
---


multi_lars
{'lrs': (10000, 1), 'weights_sum_sq': (10000, 1), 'grads_sum_sq': (10000, 1), 'wds': (10000, 1), 'eta': 0.5, 'eps': 1e-08, 'rescale_grad': 0.4}
5574113.5
0.039
---


multi_lars
{'lrs': (10000, 100), 'weights_sum_sq': (10000, 100), 'grads_sum_sq': (10000, 100), 'wds': (10000, 100), 'eta': 0.5, 'eps': 1e-08, 'rescale_grad': 0.4}
5578093.5
0.9514
---


multi_mp_sgd_mom_update
{'args0': '<NDArray 5x5 @cpu(0)>', 'args1': '<NDArray 5x5 @cpu(0)>', 'args2': '<NDArray 5x5 @cpu(0)>', 'args3': '<NDArray 5x5 @cpu(0)>', 'lrs': 0.1, 'wds': 0.2, 'out': '<NDArray 5x5 @cpu(0)>'}
---
0.0403
---


multi_mp_sgd_update
{'args0': '<NDArray 5x5 @cpu(0)>', 'args1': '<NDArray 5x5 @cpu(0)>', 'args2': '<NDArray 5x5 @cpu(0)>', 'lrs': 0.1, 'wds': 0.2, 'out': '<NDArray 5x5 @cpu(0)>'}
---
0.0332
---


multi_sgd_mom_update
{'args0': '<NDArray 5x5 @cpu(0)>', 'args1': '<NDArray 5x5 @cpu(0)>', 'args2': '<NDArray 5x5 @cpu(0)>', 'lrs': 0.1, 'wds': 0.2, 'out': '<NDArray 5x5 @cpu(0)>'}
---
0.0352
---


multi_sgd_update
{'args0': '<NDArray 5x5 @cpu(0)>', 'args1': '<NDArray 5x5 @cpu(0)>', 'lrs': 0.1, 'wds': 0.2, 'out': '<NDArray 5x5 @cpu(0)>'}
---
0.0364
---


multi_sum_sq
{'args': [(1024, 1024)], 'num_arrays': 1}
5398911.5
0.0298
---


multi_sum_sq
{'args': [(10000, 1)], 'num_arrays': 1}
5398911.5
0.0308
---


multi_sum_sq
{'args': [(10000, 10)], 'num_arrays': 1}
5398911.5
0.0309
---


nag_mom_update
{'weight': (1024, 1024), 'grad': (1024, 1024), 'mom': (1024, 1024), 'lr': 0.1, 'wd': 0.1, 'rescale_grad': 0.4}
4747800.5
0.2863
---


nag_mom_update
{'weight': (10000, 1), 'grad': (10000, 1), 'mom': (10000, 1), 'lr': 0.5, 'wd': 0.5, 'rescale_grad': 0.4}
4745743.0
0.0655
---


nag_mom_update
{'weight': (10000, 100), 'grad': (10000, 100), 'mom': (10000, 100), 'lr': 0.9, 'wd': 0.9, 'rescale_grad': 0.4}
4749723.0
0.3044
---


nanprod
{'data': (1024, 1024), 'axis': 0}
223271.4531
0.7834
0.7836


nanprod
{'data': (10000, 1), 'axis': 0}
223269.4219
0.14
0.0384


nanprod
{'data': (10000, 100), 'axis': 0}
223269.8125
0.6744
0.7459


nansum
{'data': (1024, 1024), 'axis': 0}
223275.7656
0.796
0.8368


nansum
{'data': (10000, 1), 'axis': 0}
223273.7188
0.1565
0.0383


nansum
{'data': (10000, 100), 'axis': 0}
223274.1094
0.7841
0.7963


negative
{'data': (1024, 1024)}
2097.1521
0.0551
---


negative
{'data': (10000, 1)}
20.0
0.0271
---


negative
{'data': (10000, 100)}
2000.0
0.0585
---


norm
{'data': (1024, 1024), 'axis': 0}
223280.0625
0.9442
0.8252


norm
{'data': (10000, 1), 'axis': 0}
223278.0156
0.1732
0.0399


norm
{'data': (10000, 100), 'axis': 0}
223278.4062
0.8574
0.7871


one_hot
{'indices': (1, 1), 'depth': 0, 'dtype': 'float32'}
---
0.0137
---


one_hot
{'indices': (1, 1), 'depth': 0, 'dtype': 'int32'}
---
0.0136
---


one_hot
{'indices': (1, 1), 'depth': 0, 'dtype': 'float32'}
---
0.0139
---


ones_like
{'data': (1024, 1024)}
2097.1521
0.0543
---


ones_like
{'data': (10000, 1)}
20.0
0.0246
---


ones_like
{'data': (10000, 100)}
2000.0
0.0536
---


pad
{'data': (1, 4, 2, 4), 'mode': 'constant', 'pad_width': (0, 0, 0, 0, 1, 1, 1, 1)}
309895.2812
---
---


pad
{'data': (10, 25, 10, 100), 'mode': 'constant', 'pad_width': (0, 0, 0, 0, 1, 1, 1, 1)}
311119.0625
---
---


pick
{'data': (1024, 1024), 'index': (1, 1024), 'axis': 0}
235743.0469
0.0411
0.286


pick
{'data': (10000, 1), 'index': (1, 1), 'axis': 0}
235741.0
0.0372
0.0492


pick
{'data': (10000, 100), 'index': (1, 100), 'axis': 0}
235741.4062
0.0384
0.2822


preloaded_multi_mp_sgd_mom_update
{'args0': '<NDArray 5x5 @cpu(0)>', 'args1': '<NDArray 5x5 @cpu(0)>', 'args2': '<NDArray 5x5 @cpu(0)>', 'args3': '<NDArray 5x5 @cpu(0)>', 'args4': '<NDArray 1 @cpu(0)>', 'args5': '<NDArray 1 @cpu(0)>', 'out': '<NDArray 5x5 @cpu(0)>'}
---
0.0353
---


preloaded_multi_mp_sgd_update
{'args0': '<NDArray 5x5 @cpu(0)>', 'args1': '<NDArray 5x5 @cpu(0)>', 'args2': '<NDArray 5x5 @cpu(0)>', 'args3': '<NDArray 1 @cpu(0)>', 'args4': '<NDArray 1 @cpu(0)>', 'out': '<NDArray 5x5 @cpu(0)>'}
---
0.0341
---


preloaded_multi_sgd_mom_update
{'args0': '<NDArray 5x5 @cpu(0)>', 'args1': '<NDArray 5x5 @cpu(0)>', 'args2': '<NDArray 5x5 @cpu(0)>', 'args3': '<NDArray 1 @cpu(0)>', 'args4': '<NDArray 1 @cpu(0)>', 'out': '<NDArray 5x5 @cpu(0)>'}
---
0.0341
---


preloaded_multi_sgd_update
{'args0': '<NDArray 5x5 @cpu(0)>', 'args1': '<NDArray 5x5 @cpu(0)>', 'args4': '<NDArray 1 @cpu(0)>', 'args5': '<NDArray 1 @cpu(0)>', 'out': '<NDArray 5x5 @cpu(0)>'}
---
0.0328
---


prod
{'data': (1024, 1024), 'axis': 0}
223284.3594
0.7798
0.7656


prod
{'data': (10000, 1), 'axis': 0}
223282.3125
0.1594
0.0406


prod
{'data': (10000, 100), 'axis': 0}
223282.7031
0.7659
0.9068


radians
{'data': (1024, 1024)}
2097.1521
0.0504
0.0417


radians
{'data': (10000, 1)}
20.0
0.0301
0.0287


radians
{'data': (10000, 100)}
2000.0
0.0507
0.0403


random_exponential
{'shape': (1024, 1024), 'dtype': 'float16'}
2097.1521
2.5186
---


random_exponential
{'shape': (10000, 1), 'dtype': 'float32'}
20.0
0.0514
---


random_exponential
{'shape': (10000, 100), 'dtype': 'float64'}
8000.0
2.2788
---


random_gamma
{'shape': (1024, 1024), 'dtype': 'float16'}
2097.1521
6.947
---


random_gamma
{'shape': (10000, 1), 'dtype': 'float32'}
20.0
0.0989
---


random_gamma
{'shape': (10000, 100), 'dtype': 'float64'}
8000.0
8.3352
---


random_generalized_negative_binomial
{'shape': (1024, 1024), 'dtype': 'float16'}
2097.1521
9.38
---


random_generalized_negative_binomial
{'shape': (10000, 1), 'dtype': 'float32'}
20.0
0.1396
---


random_generalized_negative_binomial
{'shape': (10000, 100), 'dtype': 'float64'}
8000.0
8.9768
---


random_negative_binomial
{'k': 1, 'p': 1, 'shape': (1024, 1024), 'dtype': 'float16'}
2097.1521
6.5638
---


random_negative_binomial
{'k': 1, 'p': 1, 'shape': (10000, 1), 'dtype': 'float32'}
40.0
0.1053
---


random_negative_binomial
{'k': 1, 'p': 1, 'shape': (10000, 100), 'dtype': 'float64'}
8000.0
7.8058
---


random_normal
{'shape': (1024, 1024), 'dtype': 'float16'}
1048.576
3.0816
---


random_normal
{'shape': (10000, 1), 'dtype': 'float32'}
20.0
0.0512
---


random_normal
{'shape': (10000, 100), 'dtype': 'float64'}
4000.0
2.6179
---


random_pdf_dirichlet
{'sample': (2,), 'alpha': [0.0, 2.5]}
0.002
0.0372
---


random_pdf_exponential
{'sample': (2,), 'lam': [1.0, 8.5]}
0.004
0.0355
---


random_pdf_gamma
{'sample': (2,), 'alpha': [0.0, 2.5], 'beta': [1.0, 0.7]}
0.008
0.0411
---


random_pdf_generalized_negative_binomial
{'sample': (2,), 'mu': [2.0, 2.5], 'alpha': [0.0, 2.5]}
0.004
0.034
---


random_pdf_negative_binomial
{'sample': (2,), 'k': [20, 49], 'p': [0.4, 0.77]}
0.008
0.0347
---


random_pdf_normal
{'sample': (2,), 'mu': [2.0, 2.5], 'sigma': [1.0, 3.7]}
0.008
0.0355
---


random_pdf_poisson
{'sample': (2,), 'lam': [1.0, 8.5]}
0.004
0.0356
---


random_pdf_uniform
{'sample': (2,), 'low': [0.0, 2.5], 'high': [1.0, 3.7]}
0.004
0.0335
---


random_poisson
{'shape': (1024, 1024), 'dtype': 'float16'}
1048.576
2.0767
---


random_poisson
{'shape': (10000, 1), 'dtype': 'float32'}
20.0
0.0537
---


random_poisson
{'shape': (10000, 100), 'dtype': 'float64'}
4000.0
1.5296
---


random_randint
{'low': 0, 'high': 5, 'shape': (1024, 1024), 'dtype': 'int32'}
4194.3042
1.3714
---


random_randint
{'low': 0, 'high': 5, 'shape': (10000, 1), 'dtype': 'int64'}
40.0
0.0434
---


random_randint
{'low': 0, 'high': 5, 'shape': (10000, 100), 'dtype': 'int32'}
2000.0
1.3065
---


random_uniform
{'low': 0, 'high': 5, 'shape': (1024, 1024), 'dtype': 'float16'}
2097.1521
0.9942
---


random_uniform
{'low': 0, 'high': 5, 'shape': (10000, 1), 'dtype': 'float32'}
20.0
0.0394
---


random_uniform
{'low': 0, 'high': 5, 'shape': (10000, 100), 'dtype': 'float64'}
4000.0
0.9265
---


ravel_multi_index
{'data': (2, 1024), 'shape': (1024, 1024)}
235747.3438
0.0381
---


ravel_multi_index
{'data': (2, 1024), 'shape': (10000, 1)}
235749.3906
0.0368
---


ravel_multi_index
{'data': (2, 1024), 'shape': (10000, 100)}
235751.4375
0.0378
---


rcbrt
{'data': (1024, 1024)}
4194.3042
1.3384
1.397


rcbrt
{'data': (10000, 1)}
20.0
0.0464
0.0381


rcbrt
{'data': (10000, 100)}
4000.0
1.2793
1.3368


reciprocal
{'data': (1024, 1024)}
2097.1521
0.0621
0.0617


reciprocal
{'data': (10000, 1)}
40.0
0.0297
0.0315


reciprocal
{'data': (10000, 100)}
2000.0
0.0643
0.0583


relu
{'data': (1024, 1024)}
2097.1521
0.081
0.1016


relu
{'data': (10000, 1)}
20.0
0.0316
0.0388


relu
{'data': (10000, 100)}
2000.0
0.0813
0.0965


repeat
{'data': (1024, 1024), 'repeats': 2, 'axis': 0}
323090.0
2.0962
2.9878


repeat
{'data': (10000, 1), 'repeats': 2, 'axis': 0}
318975.6875
0.0654
0.0628


repeat
{'data': (10000, 100), 'repeats': 2, 'axis': 0}
326935.6875
3.1059
4.5136


reset_arrays
{'args': [(1024, 1024)], 'num_arrays': 1}
---
0.0096
---


reset_arrays
{'args': [(10000, 1)], 'num_arrays': 1}
---
0.0098
---


reset_arrays
{'args': [(10000, 10)], 'num_arrays': 1}
---
0.0097
---


reshape
{'data': (1024, 1024), 'shape': (1024, 1024), 'target_shape': (32, 6)}
282483.0938
0.1375
---


reshape
{'data': (10000, 1), 'shape': (10000, 1), 'target_shape': (32, 6)}
280425.9375
0.0547
---


reshape
{'data': (10000, 100), 'shape': (10000, 100), 'target_shape': (32, 6)}
284405.9375
0.1234
---


reshape_like
{'lhs': (1024, 1024), 'rhs': (1024, 1024)}
288697.375
0.5433
---


reshape_like
{'lhs': (10000, 10), 'rhs': (10000, 10)}
287000.25
0.0601
---


reshape_like
{'lhs': (10000, 1), 'rhs': (10000, 1)}
286840.25
0.0195
---


reverse
{'data': (1024, 1024), 'axis': 0}
261238.6094
0.4752
0.4646


reverse
{'data': (10000, 1), 'axis': 0}
259181.4531
0.0375
0.0312


reverse
{'data': (10000, 100), 'axis': 0}
263161.4375
0.4917
0.4575


rint
{'data': (1024, 1024)}
360298.6562
0.1924
---


rint
{'data': (10000, 1)}
358241.5
0.0287
---


rint
{'data': (10000, 100)}
362221.5
0.3056
---


rmsprop_update
{'weight': (1024, 1024), 'grad': (1024, 1024), 'n': (1024, 1024), 'lr': 0.1, 'gamma1': 0.1, 'epsilon': 1e-08, 'wd': 0.1, 'rescale_grad': 0.4}
4754014.5
0.4045
---


rmsprop_update
{'weight': (10000, 1), 'grad': (10000, 1), 'n': (10000, 1), 'lr': 0.5, 'gamma1': 0.5, 'epsilon': 1e-08, 'wd': 0.5, 'rescale_grad': 0.4}
4751957.5
0.0384
---


rmsprop_update
{'weight': (10000, 100), 'grad': (10000, 100), 'n': (10000, 100), 'lr': 0.9, 'gamma1': 0.9, 'epsilon': 1e-08, 'wd': 0.9, 'rescale_grad': 0.4}
4755937.5
0.4541
---


rmspropalex_update
{'weight': (1024, 1024), 'grad': (1024, 1024), 'n': (1024, 1024), 'g': (1024, 1024), 'delta': (1024, 1024), 'lr': 0.1, 'gamma1': 0.1, 'gamma2': 0.1, 'epsilon': 1e-08, 'wd': 0.1, 'rescale_grad': 0.4}
4760229.0
0.6353
---


rmspropalex_update
{'weight': (10000, 1), 'grad': (10000, 1), 'n': (10000, 1), 'g': (10000, 1), 'delta': (10000, 1), 'lr': 0.5, 'gamma1': 0.5, 'gamma2': 0.5, 'epsilon': 1e-08, 'wd': 0.5, 'rescale_grad': 0.4}
4758172.0
0.0677
---


rmspropalex_update
{'weight': (10000, 100), 'grad': (10000, 100), 'n': (10000, 100), 'g': (10000, 100), 'delta': (10000, 100), 'lr': 0.9, 'gamma1': 0.9, 'gamma2': 0.9, 'epsilon': 1e-08, 'wd': 0.9, 'rescale_grad': 0.4}
4762152.0
1.1841
---


round
{'data': (1024, 1024)}
366512.9688
0.3535
---


round
{'data': (10000, 1)}
364455.8125
0.0269
---


round
{'data': (10000, 100)}
368435.8125
0.3296
---


rsqrt
{'data': (1024, 1024)}
2097.1521
0.6999
0.6844


rsqrt
{'data': (10000, 1)}
40.0
0.0391
0.0317


rsqrt
{'data': (10000, 100)}
2000.0
0.6705
0.6459


sample_exponential
{'lam': [1.0, 8.5], 'shape': (1024, 1024), 'dtype': 'float16'}
2097.1521
3.6229
---


sample_exponential
{'lam': [1.0, 8.5], 'shape': (10000, 1), 'dtype': 'float32'}
40.0
0.0661
---


sample_exponential
{'lam': [1.0, 8.5], 'shape': (10000, 100), 'dtype': 'float64'}
8000.0
3.2448
---


sample_gamma
{'alpha': [0.0, 2.5], 'shape': (1024, 1024), 'dtype': 'float16', 'beta': [1.0, 0.7]}
4194.3042
13.3519
---


sample_gamma
{'alpha': [0.0, 2.5], 'shape': (10000, 1), 'dtype': 'float32', 'beta': [1.0, 0.7]}
40.0
0.1437
---


sample_gamma
{'alpha': [0.0, 2.5], 'shape': (10000, 100), 'dtype': 'float64', 'beta': [1.0, 0.7]}
16000.0
13.662
---


sample_generalized_negative_binomial
{'mu': [2.0, 2.5], 'shape': (1024, 1024), 'dtype': 'float16', 'alpha': [0.0, 2.5]}
4194.3042
29.5582
---


sample_generalized_negative_binomial
{'mu': [2.0, 2.5], 'shape': (10000, 1), 'dtype': 'float32', 'alpha': [0.0, 2.5]}
40.0
0.3035
---


sample_generalized_negative_binomial
{'mu': [2.0, 2.5], 'shape': (10000, 100), 'dtype': 'float64', 'alpha': [0.0, 2.5]}
8000.0
26.279
---


sample_multinomial
{'data': (32, 32), 'shape': (1024, 1024), 'dtype': 'float16'}
79218.9922
176.1481
---


sample_multinomial
{'data': (32, 32), 'shape': (10000, 1), 'dtype': 'float32'}
46944.5625
1.7724
---


sample_multinomial
{'data': (32, 32), 'shape': (10000, 100), 'dtype': 'float64'}
302304.5625
180.7354
---


sample_negative_binomial
{'k': [20, 49], 'shape': (1024, 1024), 'dtype': 'float16', 'p': [0.4, 0.77]}
180596.0156
330.7904
---


sample_negative_binomial
{'k': [20, 49], 'shape': (10000, 1), 'dtype': 'float32', 'p': [0.4, 0.77]}
178578.8594
3.191
---


sample_negative_binomial
{'k': [20, 49], 'shape': (10000, 100), 'dtype': 'float64', 'p': [0.4, 0.77]}
194538.8594
315.403
---


sample_normal
{'mu': [2.0, 2.5], 'shape': (1024, 1024), 'dtype': 'float16', 'sigma': [1.0, 3.7]}
192830.3281
8.1756
---


sample_normal
{'mu': [2.0, 2.5], 'shape': (10000, 1), 'dtype': 'float32', 'sigma': [1.0, 3.7]}
190813.1719
0.1012
---


sample_normal
{'mu': [2.0, 2.5], 'shape': (10000, 100), 'dtype': 'float64', 'sigma': [1.0, 3.7]}
206773.1719
7.9631
---


sample_poisson
{'lam': [1.0, 8.5], 'shape': (1024, 1024), 'dtype': 'float16'}
205064.625
9.5478
---


sample_poisson
{'lam': [1.0, 8.5], 'shape': (10000, 1), 'dtype': 'float32'}
203047.4688
0.1273
---


sample_poisson
{'lam': [1.0, 8.5], 'shape': (10000, 100), 'dtype': 'float64'}
219007.4688
7.2977
---


sample_uniform
{'low': [0.0, 2.5], 'shape': (1024, 1024), 'dtype': 'float16', 'high': [1.0, 3.7]}
217298.9375
1.9348
---


sample_uniform
{'low': [0.0, 2.5], 'shape': (10000, 1), 'dtype': 'float32', 'high': [1.0, 3.7]}
215281.7812
0.0429
---


sample_uniform
{'low': [0.0, 2.5], 'shape': (10000, 100), 'dtype': 'float64', 'high': [1.0, 3.7]}
231241.7812
1.7466
---


sgd_mom_update
{'weight': (1024, 1024), 'grad': (1024, 1024), 'mom': (1024, 1024), 'lr': 0.1, 'wd': 0.1, 'rescale_grad': 0.4, 'lazy_update': 0}
4766443.0
0.2115
---


sgd_mom_update
{'weight': (10000, 1), 'grad': (10000, 1), 'mom': (10000, 1), 'lr': 0.5, 'wd': 0.5, 'rescale_grad': 0.4, 'lazy_update': 0}
4764386.0
0.0422
---


sgd_mom_update
{'weight': (10000, 100), 'grad': (10000, 100), 'mom': (10000, 100), 'lr': 0.9, 'wd': 0.9, 'rescale_grad': 0.4, 'lazy_update': 0}
4768366.0
0.201
---


sgd_update
{'weight': (1024, 1024), 'grad': (1024, 1024), 'lr': 0.1, 'wd': 0.1, 'rescale_grad': 0.4, 'lazy_update': 0}
4772657.5
0.1861
---


sgd_update
{'weight': (10000, 1), 'grad': (10000, 1), 'lr': 0.5, 'wd': 0.5, 'rescale_grad': 0.4, 'lazy_update': 0}
4770600.5
0.0307
---


sgd_update
{'weight': (10000, 100), 'grad': (10000, 100), 'lr': 0.9, 'wd': 0.9, 'rescale_grad': 0.4, 'lazy_update': 0}
4774580.5
0.1693
---


shape_array
{'data': (1024, 1024)}
286820.25
0.0127
---


shape_array
{'data': (10000, 1)}
286820.2812
0.0124
---


shape_array
{'data': (10000, 100)}
286820.2812
0.0121
---


shuffle
{'data': (1024, 1024)}
2097.1521
1.1539
---


shuffle
{'data': (10000, 1)}
20.0
0.3627
---


shuffle
{'data': (10000, 100)}
2000.0
1.2481
---


sigmoid
{'data': (1024, 1024)}
4194.3042
0.9294
0.0746


sigmoid
{'data': (10000, 1)}
20.0
0.0371
0.03


sigmoid
{'data': (10000, 100)}
4000.0
0.7201
0.0713


sign
{'data': (1024, 1024)}
2097.1521
0.2711
0.0529


sign
{'data': (10000, 1)}
20.0
0.0828
0.0253


sign
{'data': (10000, 100)}
2000.0
0.3402
0.1345


signsgd_update
{'weight': (1024, 1024), 'grad': (1024, 1024), 'lr': 0.1, 'wd': 0.1, 'rescale_grad': 0.4}
4778872.0
0.2178
---


signsgd_update
{'weight': (10000, 1), 'grad': (10000, 1), 'lr': 0.5, 'wd': 0.5, 'rescale_grad': 0.4}
4776814.5
0.0362
---


signsgd_update
{'weight': (10000, 100), 'grad': (10000, 100), 'lr': 0.9, 'wd': 0.9, 'rescale_grad': 0.4}
4780794.5
0.2093
---


signum_update
{'weight': (1024, 1024), 'grad': (1024, 1024), 'mom': (1024, 1024), 'lr': 0.1, 'wd': 0.1, 'rescale_grad': 0.4}
4785086.0
0.3259
---


signum_update
{'weight': (10000, 1), 'grad': (10000, 1), 'mom': (10000, 1), 'lr': 0.5, 'wd': 0.5, 'rescale_grad': 0.4}
4783029.0
0.0391
---


signum_update
{'weight': (10000, 100), 'grad': (10000, 100), 'mom': (10000, 100), 'lr': 0.9, 'wd': 0.9, 'rescale_grad': 0.4}
4787009.0
0.3102
---


sin
{'data': (1024, 1024)}
2097.1521
0.5525
0.5599


sin
{'data': (10000, 1)}
40.0
0.0829
0.0316


sin
{'data': (10000, 100)}
4000.0
0.5252
0.5173


sinh
{'data': (1024, 1024)}
4194.3042
1.5371
0.9416


sinh
{'data': (10000, 1)}
20.0
0.0453
0.0379


sinh
{'data': (10000, 100)}
4000.0
1.4809
0.9119


size_array
{'data': (1024, 1024)}
286820.2812
0.0124
---


size_array
{'data': (10000, 1)}
286820.2812
0.0124
---


size_array
{'data': (10000, 100)}
286820.2812
0.0125
---


slice
{'data': (1024, 1024), 'begin': 0, 'end': 1}
235755.5312
0.0577
0.0556


slice
{'data': (10000, 1), 'begin': 0, 'end': 1}
235753.5
0.0557
0.0377


slice
{'data': (10000, 100), 'begin': 0, 'end': 1}
235753.8906
0.0554
0.0558


slice_axis
{'data': (1024, 1024), 'axis': 0, 'begin': 0, 'end': 1}
235759.8438
0.0373
0.0551


slice_axis
{'data': (10000, 1), 'axis': 0, 'begin': 0, 'end': 1}
235757.7969
0.0371
0.0475


slice_axis
{'data': (10000, 100), 'axis': 0, 'begin': 0, 'end': 1}
235758.1875
0.0363
0.0554


slice_like
{'data': (1024, 1024), 'shape_like': (100, 100), 'axes': [0, 1]}
235817.9844
0.0431
0.0632


slice_like
{'data': (10000, 1), 'shape_like': (10, 1), 'axes': [0, 1]}
235798.0312
0.0405
0.04


slice_like
{'data': (10000, 100), 'shape_like': (100, 10), 'axes': [0, 1]}
235802.0156
0.067
0.0623


smooth_l1
{'data': (1024, 1024)}
5398988.5
0.2996
0.2847


smooth_l1
{'data': (10000, 1)}
5396931.5
0.0717
0.0648


smooth_l1
{'data': (10000, 100)}
5400911.5
0.5835
0.2741


softmax
{'data': (1024, 1024), 'axis': 0, 'dtype': 'float16'}
1608822.25
2.0278
0.9154


softmax
{'data': (10000, 1), 'axis': 0, 'dtype': 'float32'}
1607813.625
0.4857
0.0424


softmax
{'data': (10000, 100), 'axis': 0, 'dtype': 'float64'}
1615793.625
5.3102
2.2144


softmax_cross_entropy
{'data': (1024, 1024), 'label': (1024,)}
5398911.5
0.8294
---


softmin
{'data': (1024, 1024), 'axis': 0, 'dtype': 'float16'}
1614939.375
3.4102
1.8065


softmin
{'data': (10000, 1), 'axis': 0, 'dtype': 'float32'}
1614979.375
0.4609
0.0443


softmin
{'data': (10000, 100), 'axis': 0, 'dtype': 'float64'}
1622959.375
4.6611
2.0522


softsign
{'data': (1024, 1024)}
4194.3042
0.0804
0.0684


softsign
{'data': (10000, 1)}
20.0
0.0359
0.0382


softsign
{'data': (10000, 100)}
2000.0
0.0681
0.0653


sort
{'data': (1024, 1024), 'axis': 0}
237902.6094
22.5737
---


sort
{'data': (10000, 1), 'axis': 0}
233768.3125
1.0186
---


sort
{'data': (10000, 100), 'axis': 0}
239728.3125
21.2701
---


space_to_depth
{'data': (1, 4, 2, 4), 'block_size': 2}
261161.6406
0.0546
---


space_to_depth
{'data': (10, 25, 10, 100), 'block_size': 5}
262161.5625
0.5257
---


split
{'data': (1024, 1024), 'num_outputs': 2}
379061.5625
---
---


split
{'data': (10000, 1), 'num_outputs': 1}
377004.4062
---
---


split
{'data': (10000, 100), 'num_outputs': 10}
380984.4062
---
---


sqrt
{'data': (1024, 1024)}
2097.1521
0.7633
0.0697


sqrt
{'data': (10000, 1)}
40.0
0.0369
0.032


sqrt
{'data': (10000, 100)}
4000.0
0.644
0.066


square
{'data': (1024, 1024)}
2097.1521
0.0591
0.063


square
{'data': (10000, 1)}
20.0
0.0268
0.0285


square
{'data': (10000, 100)}
2000.0
0.0578
0.0577


squeeze
{'data': (1, 1024, 1024), 'axis': 0}
5582385.0
0.447
0.6478


squeeze
{'data': (32, 1, 256, 256), 'axis': 1}
5588676.5
1.4974
1.2437


stack
{'args0': '<NDArray 100x100 @cpu(0)>', 'args1': '<NDArray 100x100 @cpu(0)>', 'args2': '<NDArray 100x100 @cpu(0)>'}
379164.4062
0.0672
---


stop_gradient
{'data': (1024, 1024)}
2097.1521
0.4951
---


stop_gradient
{'data': (10000, 1)}
40.0
0.0206
---


stop_gradient
{'data': (10000, 100)}
4000.0
0.5134
---


sum
{'data': (1024, 1024), 'axis': 0}
223288.6562
0.8658
0.6977


sum
{'data': (10000, 1), 'axis': 0}
223286.6094
0.1189
0.0306


sum
{'data': (10000, 100), 'axis': 0}
223287.0
0.6296
0.6669


sum_axis
{'data': (1024, 1024), 'axis': 0}
223292.9531
0.8614
---


sum_axis
{'data': (10000, 1), 'axis': 0}
223290.9062
0.148
---


sum_axis
{'data': (10000, 100), 'axis': 0}
223291.2969
0.5627
---


swapaxes
{'data': (1024, 1024), 'dim1': 0, 'dim2': 1}
267953.0312
1.442
---


swapaxes
{'data': (10000, 1), 'dim1': 0, 'dim2': 1}
265895.875
0.0415
---


swapaxes
{'data': (10000, 100), 'dim1': 0, 'dim2': 1}
269875.875
1.2189
---


take
{'a': (1024, 1024), 'indices': (1, 1), 'axis': 0}
235806.1562
0.038
0.0609


tan
{'data': (1024, 1024)}
2097.1521
1.1871
0.0752


tan
{'data': (10000, 1)}
20.0
0.0398
0.0308


tan
{'data': (10000, 100)}
2000.0
0.8581
0.0552


tanh
{'data': (1024, 1024)}
2097.1521
1.4491
0.0813


tanh
{'data': (10000, 1)}
20.0
0.0405
0.0272


tanh
{'data': (10000, 100)}
2000.0
1.379
0.0796


tile
{'data': (1024, 1024), 'reps': 2}
335518.5938
3.1205
4.5071


tile
{'data': (10000, 1), 'reps': 2}
331404.2812
0.0503
0.0451


tile
{'data': (10000, 100), 'reps': 2}
339364.2812
2.9696
4.3805


topk
{'data': (1024, 1024), 'axis': 0, 'k': 1, 'dtype': 'float32'}
235734.4531
17.0116
---


topk
{'data': (10000, 1), 'axis': 0, 'k': 1, 'dtype': 'int32'}
235732.4062
0.2139
---


topk
{'data': (10000, 100), 'axis': 0, 'k': 1, 'dtype': 'float32'}
235732.8125
15.8706
---


transpose
{'data': (1024, 1024), 'axes': [0, 1]}
274167.3438
0.0696
---


transpose
{'data': (10000, 1), 'axes': [0, 1]}
272110.1875
0.0499
---


transpose
{'data': (10000, 100), 'axes': [0, 1]}
276090.1875
0.0627
---


trunc
{'data': (1024, 1024)}
372727.2812
0.3051
---


trunc
{'data': (10000, 1)}
370670.125
0.0274
---


trunc
{'data': (10000, 100)}
374650.125
0.293
---


where
{'condition': (1024,), 'x': (1024, 1024), 'y': (1024, 1024)}
242095.5625
0.2336
0.4337


where
{'condition': (10000,), 'x': (10000, 1), 'y': (10000, 1)}
240038.4062
0.0409
0.0476


where
{'condition': (10000,), 'x': (10000, 100), 'y': (10000, 100)}
244018.4062
0.2264
0.4194


zeros_like
{'data': (1024, 1024)}
2097.1521
0.0485
---


zeros_like
{'data': (10000, 1)}
20.0
0.016
---


zeros_like
{'data': (10000, 100)}
4000.0
0.0672
---