connorgoggins/embedding_fix_all_ops_gpu_5_warmup_5_runs.md Secret

## embedding_fix_all_ops_gpu_5_warmup_5_runs.md

      
    Raw
  

              embedding_fix_all_ops_gpu_5_warmup_5_runs.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'}
357422.5312
0.0552
0.1204


Activation
{'data': (1024, 1024), 'act_type': 'sigmoid'}
359519.6875
0.1732
0.9284


Activation
{'data': (1024, 1024), 'act_type': 'softrelu'}
361616.8438
0.1674
0.8194


Activation
{'data': (1024, 1024), 'act_type': 'softsign'}
363714.0
0.0536
0.0596


Activation
{'data': (1024, 1024), 'act_type': 'tanh'}
365811.1562
0.1902
1.7102


BatchNorm
{'data': (32, 3, 256, 256), 'gamma': (3,), 'beta': (3,), 'moving_mean': (3,), 'moving_var': (3,)}
344514.0
2.6362
1.4606


BatchNorm
{'data': (32, 3, 10000, 10), 'gamma': (3,), 'beta': (3,), 'moving_mean': (3,), 'moving_var': (3,)}
370331.0938
6.5164
1.9314


BilinearSampler
{'data': (32, 2, 256, 256), 'grid': (32, 2, 256, 256)}
25165.8242
103.5358
117.9724


BlockGrad
{'data': (1024, 1024)}
2097.1521
0.4334
---


BlockGrad
{'data': (10000, 1)}
20.0
0.0216
---


BlockGrad
{'data': (10000, 100)}
2000.0
0.4104
---


CTCLoss
{'data': (1024, 100, 100), 'label': (100, 100)}
4147407.75
80.882
---


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'}
3172700.5
6.7754
4.3082


Custom
{'args': [(1024, 1024)], 'op_type': 'CustomAddOne'}
4270740.0
0.007
0.0088


Custom
{'args': [(10000, 1)], 'op_type': 'CustomAddOne'}
4270740.0
0.007
0.0072


Custom
{'args': [(10000, 10)], 'op_type': 'CustomAddOne'}
4270740.0
0.006
0.0068


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'}
4124289.25
25.1678
24.35


Dropout
{'data': (32, 3, 256, 256), 'p': 0.5, 'mode': 'always'}
318948.1875
3.5208
0.756


Dropout
{'data': (10000, 10), 'p': 0.5, 'mode': 'always'}
306965.2812
0.0868
0.0264


ElementWiseSum
{'args': (1024, 1024)}
2097.1521
0.0906
---


FullyConnected
{'data': (32, 3, 256, 256), 'num_hidden': 64, 'weight': (64, 196608), 'bias': (64,), 'flatten': True}
252891.9844
1.6274
4.4118


FullyConnected
{'data': (32, 3, 256, 256), 'num_hidden': 64, 'weight': (64, 256), 'bias': (64,), 'flatten': False}
259179.3438
1.3982
6.1992


GridGenerator
{'data': (32, 2, 256, 256), 'transform_type': 'warp', 'target_shape': (256, 6)}
8650.752
29.8418
21.215


GridGenerator
{'data': (256, 6), 'transform_type': 'affine', 'target_shape': (256, 6)}
1582.08
0.4098
0.578


LeakyReLU
{'data': (1024, 1024), 'act_type': 'leaky'}
370005.4688
0.0488
0.0556


LeakyReLU
{'data': (1024, 1024), 'act_type': 'elu'}
372102.625
0.1116
0.721


LeakyReLU
{'data': (1024, 1024), 'act_type': 'selu'}
374199.75
0.6926
0.2626


LeakyReLU
{'data': (1024, 1024), 'act_type': 'gelu'}
376296.9062
0.2298
2.0292


MakeLoss
{'data': (1024, 1024)}
4133219.0
0.071
0.03


MakeLoss
{'data': (10000, 1)}
4131161.75
0.0198
0.0106


MakeLoss
{'data': (10000, 100)}
4135141.75
4.8696
0.0288


Pooling
{'data': (32, 3, 64, 64), 'kernel': (3, 3), 'pool_type': 'sum', 'global_pool': 1, 'stride': (1, 1), 'pad': (0, 0)}
477931.0312
0.4686
0.2812


ROIPooling
{'data': (32, 3, 64, 64), 'rois': (32, 5), 'pooled_size': (2, 2), 'spatial_scale': 0.5}
477938.5
1.19
0.095


SequenceLast
{'data': (1024, 1024), 'axis': 0}
4270746.0
0.0308
---


SequenceLast
{'data': (10000, 1), 'axis': 0}
4270744.0
0.0186
---


SequenceLast
{'data': (10000, 100), 'axis': 0}
4270744.5
0.019
---


SequenceMask
{'data': (1024, 1024), 'axis': 0}
4277035.5
0.2584
0.2368


SequenceMask
{'data': (10000, 1), 'axis': 0}
4274978.5
0.021
0.0134


SequenceMask
{'data': (10000, 100), 'axis': 0}
4278958.5
0.2248
0.2208


SequenceReverse
{'data': (1024, 1024), 'axis': 0}
4283250.0
2.4684
2.607


SequenceReverse
{'data': (10000, 1), 'axis': 0}
4281192.5
0.0282
0.0242


SequenceReverse
{'data': (10000, 100), 'axis': 0}
4285172.5
1.9482
2.0684


Softmax
{'data': (1024, 1024), 'label': (1024, 1024)}
380491.2188
0.1542
---


Softmax
{'data': (10000, 1), 'label': (10000, 1)}
378434.0625
0.1078
---


Softmax
{'data': (10000, 100), 'label': (10000, 100)}
382414.0625
0.2178
---


SoftmaxActivation
{'data': (1024, 1024)}
386705.5312
0.7064
2.5148


SoftmaxActivation
{'data': (10000, 1)}
384648.375
0.0186
0.0558


SoftmaxActivation
{'data': (10000, 100)}
388628.375
0.6738
2.4004


UpSampling
{'args': (32, 3, 256, 256), 'scale': 2, 'sample_type': 'nearest'}
4290351.5
23.397
3.6678


UpSampling
{'args': (32, 3, 10000, 1), 'scale': 4, 'sample_type': 'nearest'}
4301460.0
16.8458
2.4004


abs
{'data': (1024, 1024)}
4194.3042
0.1058
0.6502


abs
{'data': (10000, 1)}
40.0
0.0628
0.0586


abs
{'data': (10000, 100)}
4000.0
0.1046
0.629


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}
3163247.0
0.4316
---


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}
3161189.75
0.4754
---


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}
3165169.75
2.5598
---


add_n
{'args': [(1024, 1024)]}
4139356.5
0.0234
---


add_n
{'args': [(10000, 1)]}
4139356.5
0.0166
---


add_n
{'args': [(10000, 10)]}
4139356.5
0.0162
---


all_finite
{'data': (1024, 1024)}
4283172.5
0.068
---


all_finite
{'data': (10000, 1)}
4283172.5
0.0128
---


all_finite
{'data': (10000, 100)}
4283172.5
0.0622
---


amp_cast
{'data': (1024, 1024), 'dtype': 'float32'}
4194.3042
0.3786
0.4678


amp_cast
{'data': (10000, 1), 'dtype': 'int32'}
40.0
0.0494
0.0424


amp_cast
{'data': (10000, 100), 'dtype': 'float32'}
4000.0
0.3944
0.4732


amp_multicast
{'args': [(1024, 1024)], 'num_outputs': 1}
0.004
0.1572
0.162


amp_multicast
{'args': [(10000, 1)], 'num_outputs': 1}
0.004
0.1478
0.134


arccos
{'data': (1024, 1024)}
4194.3042
1.1568
0.8204


arccos
{'data': (10000, 1)}
20.0
0.0188
0.0144


arccos
{'data': (10000, 100)}
2000.0
0.9992
0.7378


arccosh
{'data': (1024, 1024)}
2097.1521
0.7904
0.8444


arccosh
{'data': (10000, 1)}
20.0
0.0168
0.015


arccosh
{'data': (10000, 100)}
2000.0
0.6904
0.743


arcsin
{'data': (1024, 1024)}
2097.1521
0.9504
0.7368


arcsin
{'data': (10000, 1)}
20.0
0.0174
0.0148


arcsin
{'data': (10000, 100)}
2000.0
1.0078
0.7752


arcsinh
{'data': (1024, 1024)}
2097.1521
1.7784
0.4492


arcsinh
{'data': (10000, 1)}
20.0
0.0232
0.0134


arcsinh
{'data': (10000, 100)}
2000.0
1.6044
0.4174


arctan
{'data': (1024, 1024)}
2097.1521
0.8594
0.0626


arctan
{'data': (10000, 1)}
20.0
0.0178
0.0098


arctan
{'data': (10000, 100)}
2000.0
0.8828
0.0512


arctanh
{'data': (1024, 1024)}
4194.3042
1.2732
0.0628


arctanh
{'data': (10000, 1)}
40.0
0.038
0.0084


arctanh
{'data': (10000, 100)}
4000.0
1.2022
0.0754


argmax
{'data': (1024, 1024), 'axis': 0}
214515.375
27.4112
---


argmax
{'data': (10000, 1), 'axis': 0}
214513.3281
0.189
---


argmax
{'data': (10000, 100), 'axis': 0}
214513.7188
24.7242
---


argmax_channel
{'data': (1024, 1024)}
2.048
0.646
---


argmax_channel
{'data': (10000, 1)}
20.0
0.1812
---


argmax_channel
{'data': (10000, 100)}
20.0
0.7264
---


argmin
{'data': (1024, 1024), 'axis': 0}
214519.6719
29.0478
---


argmin
{'data': (10000, 1), 'axis': 0}
214517.625
0.21
---


argmin
{'data': (10000, 100), 'axis': 0}
214518.0156
15.3104
---


argsort
{'data': (1024, 1024), 'axis': 0, 'dtype': 'float32'}
220809.2812
63.8092
---


argsort
{'data': (10000, 1), 'axis': 0, 'dtype': 'int32'}
218752.125
1.3494
---


argsort
{'data': (10000, 100), 'axis': 0, 'dtype': 'float32'}
222732.125
76.6904
---


batch_dot
{'lhs': (32, 1024, 1024), 'rhs': (32, 1024, 1024)}
134217.7344
88.9665
---


batch_dot
{'lhs': (32, 1000, 10), 'rhs': (32, 1000, 10), 'transpose_b': True}
128000.0
130.6619
---


batch_dot
{'lhs': (32, 1000, 1), 'rhs': (32, 100, 1000), 'transpose_a': True, 'transpose_b': True}
12.8
0.5856
---


broadcast_add
{'lhs': (1024, 1024), 'rhs': (1024, 1024)}
2097.1521
0.058
0.1162


broadcast_add
{'lhs': (10000, 10), 'rhs': (10000, 10)}
200.0
0.0228
0.0196


broadcast_add
{'lhs': (10000, 1), 'rhs': (10000, 1)}
20.0
0.0184
0.0218


broadcast_div
{'lhs': (1024, 1024), 'rhs': (1024, 1024)}
2097.1521
0.0658
0.1528


broadcast_div
{'lhs': (10000, 10), 'rhs': (10000, 10)}
200.0
0.021
0.0342


broadcast_div
{'lhs': (10000, 1), 'rhs': (10000, 1)}
20.0
0.0188
0.0212


broadcast_equal
{'lhs': (1024, 1024), 'rhs': (1024, 1024)}
2097.1521
0.0514
---


broadcast_equal
{'lhs': (10000, 10), 'rhs': (10000, 10)}
200.0
0.0224
---


broadcast_equal
{'lhs': (10000, 1), 'rhs': (10000, 1)}
20.0
0.0176
---


broadcast_greater
{'lhs': (1024, 1024), 'rhs': (1024, 1024)}
2097.1521
0.0514
---


broadcast_greater
{'lhs': (10000, 10), 'rhs': (10000, 10)}
200.0
0.0184
---


broadcast_greater
{'lhs': (10000, 1), 'rhs': (10000, 1)}
20.0
0.0178
---


broadcast_greater_equal
{'lhs': (1024, 1024), 'rhs': (1024, 1024)}
2097.1521
0.0494
---


broadcast_greater_equal
{'lhs': (10000, 10), 'rhs': (10000, 10)}
200.0
0.0182
---


broadcast_greater_equal
{'lhs': (10000, 1), 'rhs': (10000, 1)}
20.0
0.0176
---


broadcast_hypot
{'lhs': (1024, 1024), 'rhs': (1024, 1024)}
2097.1521
0.4218
0.9334


broadcast_hypot
{'lhs': (10000, 10), 'rhs': (10000, 10)}
200.0
0.0612
0.1666


broadcast_hypot
{'lhs': (10000, 1), 'rhs': (10000, 1)}
20.0
0.021
0.0276


broadcast_lesser
{'lhs': (1024, 1024), 'rhs': (1024, 1024)}
2097.1521
0.0508
---


broadcast_lesser
{'lhs': (10000, 10), 'rhs': (10000, 10)}
200.0
0.0222
---


broadcast_lesser
{'lhs': (10000, 1), 'rhs': (10000, 1)}
20.0
0.0164
---


broadcast_lesser_equal
{'lhs': (1024, 1024), 'rhs': (1024, 1024)}
2097.1521
0.0508
---


broadcast_lesser_equal
{'lhs': (10000, 10), 'rhs': (10000, 10)}
200.0
0.019
---


broadcast_lesser_equal
{'lhs': (10000, 1), 'rhs': (10000, 1)}
20.0
0.0178
---


broadcast_logical_and
{'lhs': (1024, 1024), 'rhs': (1024, 1024)}
2097.1521
0.0902
---


broadcast_logical_and
{'lhs': (10000, 10), 'rhs': (10000, 10)}
200.0
0.0622
---


broadcast_logical_and
{'lhs': (10000, 1), 'rhs': (10000, 1)}
20.0
0.0108
---


broadcast_logical_or
{'lhs': (1024, 1024), 'rhs': (1024, 1024)}
2097.1521
0.109
---


broadcast_logical_or
{'lhs': (10000, 10), 'rhs': (10000, 10)}
200.0
0.0228
---


broadcast_logical_or
{'lhs': (10000, 1), 'rhs': (10000, 1)}
20.0
0.012
---


broadcast_logical_xor
{'lhs': (1024, 1024), 'rhs': (1024, 1024)}
2097.1521
0.1376
---


broadcast_logical_xor
{'lhs': (10000, 10), 'rhs': (10000, 10)}
200.0
0.0238
---


broadcast_logical_xor
{'lhs': (10000, 1), 'rhs': (10000, 1)}
20.0
0.0128
---


broadcast_maximum
{'lhs': (1024, 1024), 'rhs': (1024, 1024)}
2097.1521
0.0694
0.1108


broadcast_maximum
{'lhs': (10000, 10), 'rhs': (10000, 10)}
200.0
0.0184
0.022


broadcast_maximum
{'lhs': (10000, 1), 'rhs': (10000, 1)}
20.0
0.0134
0.0158


broadcast_minimum
{'lhs': (1024, 1024), 'rhs': (1024, 1024)}
2097.1521
0.0966
0.1106


broadcast_minimum
{'lhs': (10000, 10), 'rhs': (10000, 10)}
200.0
0.018
0.024


broadcast_minimum
{'lhs': (10000, 1), 'rhs': (10000, 1)}
20.0
0.013
0.0156


broadcast_minus
{'lhs': (1024, 1024), 'rhs': (1024, 1024)}
2097.1521
0.0478
---


broadcast_minus
{'lhs': (10000, 10), 'rhs': (10000, 10)}
200.0
0.0188
---


broadcast_minus
{'lhs': (10000, 1), 'rhs': (10000, 1)}
20.0
0.0126
---


broadcast_mod
{'lhs': (1024, 1024), 'rhs': (1024, 1024)}
2097.1521
2.011
0.169


broadcast_mod
{'lhs': (10000, 10), 'rhs': (10000, 10)}
400.0
0.2096
0.028


broadcast_mod
{'lhs': (10000, 1), 'rhs': (10000, 1)}
40.0
0.0344
0.021


broadcast_mul
{'lhs': (1024, 1024), 'rhs': (1024, 1024)}
2097.1521
0.0532
0.0872


broadcast_mul
{'lhs': (10000, 10), 'rhs': (10000, 10)}
200.0
0.0168
0.0196


broadcast_mul
{'lhs': (10000, 1), 'rhs': (10000, 1)}
20.0
0.0246
0.0156


broadcast_not_equal
{'lhs': (1024, 1024), 'rhs': (1024, 1024)}
2097.1521
0.0458
---


broadcast_not_equal
{'lhs': (10000, 10), 'rhs': (10000, 10)}
200.0
0.0216
---


broadcast_not_equal
{'lhs': (10000, 1), 'rhs': (10000, 1)}
20.0
0.0128
---


broadcast_plus
{'lhs': (1024, 1024), 'rhs': (1024, 1024)}
2097.1521
0.0464
---


broadcast_plus
{'lhs': (10000, 10), 'rhs': (10000, 10)}
200.0
0.0154
---


broadcast_plus
{'lhs': (10000, 1), 'rhs': (10000, 1)}
20.0
0.0122
---


broadcast_power
{'lhs': (1024, 1024), 'rhs': (1024, 1024)}
2097.1521
2.4362
6.0946


broadcast_power
{'lhs': (10000, 10), 'rhs': (10000, 10)}
200.0
0.381
0.8694


broadcast_power
{'lhs': (10000, 1), 'rhs': (10000, 1)}
20.0
0.0612
0.0946


broadcast_sub
{'lhs': (1024, 1024), 'rhs': (1024, 1024)}
2097.1521
0.0514
0.0648


broadcast_sub
{'lhs': (10000, 10), 'rhs': (10000, 10)}
200.0
0.0154
0.0162


broadcast_sub
{'lhs': (10000, 1), 'rhs': (10000, 1)}
20.0
0.0128
0.0134


cast
{'data': (1024, 1024), 'dtype': 'float32'}
2097.1521
---
0.568


cast
{'data': (10000, 1), 'dtype': 'int32'}
20.0
---
0.015


cast
{'data': (10000, 100), 'dtype': 'float32'}
2000.0
---
0.577


cast_storage
{'data': (1024, 1024), 'stype': 'default'}
4289464.0
0.3226
---


cast_storage
{'data': (10000, 1), 'stype': 'csr'}
4287507.0
0.1238
---


cast_storage
{'data': (10000, 100), 'stype': 'row_sparse'}
4289547.0
0.6306
---


cbrt
{'data': (1024, 1024)}
2097.1521
1.429
0.06


cbrt
{'data': (10000, 1)}
20.0
0.0248
0.0108


cbrt
{'data': (10000, 100)}
2000.0
1.287
0.06


ceil
{'data': (1024, 1024)}
2097.1521
0.1252
---


ceil
{'data': (10000, 1)}
20.0
0.0112
---


ceil
{'data': (10000, 100)}
2000.0
0.1316
---


choose_element_0index
{'data': (1024, 1024), 'index': (1, 1024), 'axis': 0}
2.048
0.013
---


choose_element_0index
{'data': (10000, 1), 'index': (1, 1), 'axis': 0}
0.002
0.011
---


choose_element_0index
{'data': (10000, 100), 'index': (1, 100), 'axis': 0}
0.2
0.0118
---


clip
{'data': (1024, 1024), 'a_min': 0.1, 'a_max': 0.9}
4295798.5
0.1544
0.2572


clip
{'data': (10000, 1), 'a_min': 0.1, 'a_max': 0.9}
4293741.5
0.0122
0.0096


clip
{'data': (10000, 100), 'a_min': 0.1, 'a_max': 0.9}
4297721.5
0.1334
0.2468


cos
{'data': (1024, 1024)}
4194.3042
0.476
0.515


cos
{'data': (10000, 1)}
40.0
0.0192
0.0134


cos
{'data': (10000, 100)}
4000.0
0.4694
0.5726


cosh
{'data': (1024, 1024)}
2097.1521
0.8858
1.5906


cosh
{'data': (10000, 1)}
20.0
0.0302
0.0242


cosh
{'data': (10000, 100)}
2000.0
0.8496
1.5106


ctc_loss
{'data': (1024, 100, 100), 'label': (100, 100)}
4174102.25
56.0514
---


cumsum
{'a': (1024, 1024), 'axis': 0, 'dtype': 'float32'}
4302013.0
1.1744
---


cumsum
{'a': (1024, 1024), 'axis': 0, 'dtype': 'int32'}
4304110.0
18.6308
---


cumsum
{'a': (1024, 1024), 'axis': 0, 'dtype': 'float32'}
4306207.0
1.1912
---


degrees
{'data': (1024, 1024)}
2097.1521
0.0408
0.037


degrees
{'data': (10000, 1)}
20.0
0.0118
0.01


degrees
{'data': (10000, 100)}
2000.0
0.111
0.0378


depth_to_space
{'data': (1, 4, 2, 4), 'block_size': 2}
226950.9219
0.0202
---


depth_to_space
{'data': (10, 25, 10, 100), 'block_size': 5}
227950.8594
1.873
---


dot
{'lhs': (1024, 1024), 'rhs': (1024, 1024)}
2097.1521
2.3134
5.078


dot
{'lhs': (1000, 10), 'rhs': (1000, 10), 'transpose_b': True}
2000.0
0.1722
0.2912


dot
{'lhs': (1000, 1), 'rhs': (100, 1000), 'transpose_a': True, 'transpose_b': True}
0.2
0.0896
0.0506


elemwise_add
{'lhs': (1024, 1024), 'rhs': (1024, 1024)}
2097.1521
0.1428
---


elemwise_add
{'lhs': (10000, 10), 'rhs': (10000, 10)}
200.0
0.0472
---


elemwise_add
{'lhs': (10000, 1), 'rhs': (10000, 1)}
20.0
0.0346
---


elemwise_div
{'lhs': (1024, 1024), 'rhs': (1024, 1024)}
2097.1521
0.2986
---


elemwise_div
{'lhs': (10000, 10), 'rhs': (10000, 10)}
200.0
0.0194
---


elemwise_div
{'lhs': (10000, 1), 'rhs': (10000, 1)}
20.0
0.0128
---


elemwise_mul
{'lhs': (1024, 1024), 'rhs': (1024, 1024)}
2097.1521
0.3588
---


elemwise_mul
{'lhs': (10000, 10), 'rhs': (10000, 10)}
200.0
0.0166
---


elemwise_mul
{'lhs': (10000, 1), 'rhs': (10000, 1)}
20.0
0.0128
---


elemwise_sub
{'lhs': (1024, 1024), 'rhs': (1024, 1024)}
2097.1521
0.0494
---


elemwise_sub
{'lhs': (10000, 10), 'rhs': (10000, 10)}
200.0
0.0178
---


elemwise_sub
{'lhs': (10000, 1), 'rhs': (10000, 1)}
20.0
0.0154
---


erf
{'data': (1024, 1024)}
4194.3042
0.929
0.7918


erf
{'data': (10000, 1)}
40.0
0.0236
0.0158


erf
{'data': (10000, 100)}
4000.0
0.859
0.7768


erfinv
{'data': (1024, 1024)}
2097.1521
3.3352
0.8054


erfinv
{'data': (10000, 1)}
20.0
0.0474
0.0186


erfinv
{'data': (10000, 100)}
2000.0
4.061
0.8742


exp
{'data': (1024, 1024)}
2097.1521
0.6932
---


exp
{'data': (10000, 1)}
20.0
0.0206
---


exp
{'data': (10000, 100)}
2000.0
0.6642
---


expm1
{'data': (1024, 1024)}
2097.1521
1.1448
0.6482


expm1
{'data': (10000, 1)}
20.0
0.0236
0.043


expm1
{'data': (10000, 100)}
2000.0
1.0888
0.603


fill_element_0index
{'lhs': (1024, 1024), 'mhs': (1024,), 'rhs': (1024,)}
4310401.5
0.0804
---


fill_element_0index
{'lhs': (10000, 1), 'mhs': (10000,), 'rhs': (10000,)}
4308344.0
0.0116
---


fill_element_0index
{'lhs': (10000, 100), 'mhs': (10000,), 'rhs': (10000,)}
4312324.0
0.0772
---


fix
{'data': (1024, 1024)}
2097.1521
0.3224
---


fix
{'data': (10000, 1)}
20.0
0.0166
---


fix
{'data': (10000, 100)}
2000.0
0.3128
---


flatten
{'data': (1024, 1024)}
2097.1521
0.0556
---


flatten
{'data': (10000, 1)}
20.0
0.0354
---


flatten
{'data': (10000, 100)}
2000.0
0.074
---


flip
{'data': (1024, 1024), 'axis': 0}
233742.3125
1.196
---


flip
{'data': (10000, 1), 'axis': 0}
231685.1562
0.0278
---


flip
{'data': (10000, 100), 'axis': 0}
235665.1562
1.1392
---


floor
{'data': (1024, 1024)}
2097.1521
0.1414
---


floor
{'data': (10000, 1)}
20.0
0.0154
---


floor
{'data': (10000, 100)}
2000.0
0.119
---


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}
3169461.25
1.565
---


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}
3167404.0
0.0352
---


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}
3171384.0
2.5164
---


ftrl_update
{'weight': (1024, 1024), 'grad': (1024, 1024), 'z': (1024, 1024), 'n': (1024, 1024), 'lr': 0.1, 'wd': 0.1, 'rescale_grad': 0.4}
3175675.5
2.263
---


ftrl_update
{'weight': (10000, 1), 'grad': (10000, 1), 'z': (10000, 1), 'n': (10000, 1), 'lr': 0.5, 'wd': 0.5, 'rescale_grad': 0.4}
3173618.5
0.0584
---


ftrl_update
{'weight': (10000, 100), 'grad': (10000, 100), 'z': (10000, 100), 'n': (10000, 100), 'lr': 0.9, 'wd': 0.9, 'rescale_grad': 0.4}
3177598.5
2.2906
---


gamma
{'data': (1024, 1024)}
4194.3042
3.8958
7.119


gamma
{'data': (10000, 1)}
40.0
0.0518
0.0768


gamma
{'data': (10000, 100)}
4000.0
3.696
6.9238


gammaln
{'data': (1024, 1024)}
4194.3042
23.4108
3.4744


gammaln
{'data': (10000, 1)}
20.0
0.2374
0.0414


gammaln
{'data': (10000, 100)}
2000.0
22.1348
3.2474


gather_nd
{'data': (1024, 1024), 'indices': (1, 1)}
246597.9844
0.0224
---


gather_nd
{'data': (10000, 1), 'indices': (1, 1)}
246597.9844
0.0208
---


gather_nd
{'data': (10000, 100), 'indices': (1, 1)}
246598.1875
0.0214
---


hard_sigmoid
{'data': (1024, 1024)}
392919.8125
0.0974
0.1074


hard_sigmoid
{'data': (10000, 1)}
390862.6875
0.0126
0.0102


hard_sigmoid
{'data': (10000, 100)}
394842.6875
0.097
0.1034


identity
{'data': (1024, 1024)}
2097.1521
0.337
---


identity
{'data': (10000, 1)}
20.0
0.0186
---


identity
{'data': (10000, 100)}
2000.0
0.4072
---


khatri_rao
{'args': [(32, 32), (32, 32)]}
0.008
0.017
---


khatri_rao
{'args': [(64, 64), (64, 64)]}
0.008
0.0158
---


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}
3181889.75
0.6372
---


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}
3179832.75
0.0384
---


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}
3183812.75
1.4582
---


lamb_update_phase2
{'weight': (1024, 1024), 'g': (1024, 1024), 'r1': (1, 1024), 'r2': (1, 1024), 'lr': 0.1}
3188104.25
0.2096
---


lamb_update_phase2
{'weight': (10000, 1), 'g': (10000, 1), 'r1': (1, 1), 'r2': (1, 1), 'lr': 0.5}
3186047.0
0.1938
---


lamb_update_phase2
{'weight': (10000, 100), 'g': (10000, 100), 'r1': (1, 100), 'r2': (1, 100), 'lr': 0.9}
3190027.0
0.2496
---


linalg_det
{'A': (1024, 1024)}
4327026.5
7.4042
13.9204


linalg_extractdiag
{'A': (1024, 1024)}
4324933.5
0.0152
0.044


linalg_extracttrian
{'A': (1024, 1024)}
4329131.5
0.5524
0.5548


linalg_gelqf
{'A': (1024, 1024)}
4340665.0
32.4998
---


linalg_gemm
{'A': (1024, 1024), 'B': (1024, 1024), 'C': (1024, 1024), 'axis': 0}
4342762.0
3.7104
6.553


linalg_gemm2
{'A': (1024, 1024), 'B': (1024, 1024), 'axis': 0}
4346956.5
2.1652
4.6222


linalg_inverse
{'A': (1024, 1024)}
4353248.0
12.8776
4.9036


linalg_makediag
{'A': (1024, 1024)}
10795699.0
393.1286
63.4552


linalg_maketrian
{'A': (1024, 1035)}
8664804.0
1.3042
1.2714


linalg_potrf
{'A': [[1, 0], [0, 1]]}
4324927.0
0.1056
---


linalg_potrf
{'A': [[2, -1, 0], [-1, 2, -1], [0, -1, 2]]}
4324927.0
0.0994
---


linalg_potri
{'A': (1024, 1024)}
8666948.0
9.8314
6.9932


linalg_slogdet
{'A': (1024, 1024)}
8666950.0
6.4578
---


linalg_sumlogdiag
{'A': (1024, 1024)}
8664851.0
0.0534
0.4248


linalg_syrk
{'A': (1024, 1024)}
8671143.0
2.8064
4.4366


linalg_trmm
{'A': (1024, 1024), 'B': (1024, 1024)}
8675337.0
3.321
6.7742


linalg_trsm
{'A': (1024, 1024), 'B': (1024, 1024)}
8679531.0
3.2282
6.8102


log
{'data': (1024, 1024)}
2097.1521
1.0424
0.0922


log
{'data': (10000, 1)}
20.0
0.0188
0.0106


log
{'data': (10000, 100)}
2000.0
0.9218
0.0502


log10
{'data': (1024, 1024)}
2097.1521
1.139
0.055


log10
{'data': (10000, 1)}
20.0
0.0198
0.0108


log10
{'data': (10000, 100)}
2000.0
0.9732
0.0496


log1p
{'data': (1024, 1024)}
2097.1521
1.0444
0.0556


log1p
{'data': (10000, 1)}
20.0
0.0264
0.015


log1p
{'data': (10000, 100)}
2000.0
0.9316
0.0512


log2
{'data': (1024, 1024)}
2097.1521
0.8226
0.0502


log2
{'data': (10000, 1)}
20.0
0.0242
0.0146


log2
{'data': (10000, 100)}
2000.0
0.8384
0.0494


log_softmax
{'data': (1024, 1024), 'axis': 0, 'dtype': 'float16'}
395988.4062
3.0658
1.8808


log_softmax
{'data': (10000, 1), 'axis': 0, 'dtype': 'float32'}
394979.8125
0.702
0.319


log_softmax
{'data': (10000, 100), 'axis': 0, 'dtype': 'float64'}
402959.8125
5.6274
4.5846


logical_not
{'data': (1024, 1024)}
2097.1521
0.1672
---


logical_not
{'data': (10000, 1)}
20.0
0.0164
---


logical_not
{'data': (10000, 100)}
2000.0
0.0398
---


make_loss
{'data': (1024, 1024)}
2097.1521
0.3752
---


make_loss
{'data': (10000, 1)}
20.0
0.0162
---


make_loss
{'data': (10000, 100)}
2000.0
0.432
---


max
{'data': (1024, 1024), 'axis': 0}
214467.8906
1.7804
1.47


max
{'data': (10000, 1), 'axis': 0}
214465.8438
0.3502
0.0246


max
{'data': (10000, 100), 'axis': 0}
214466.25
1.7192
1.424


max_axis
{'data': (1024, 1024), 'axis': 0}
214472.1875
1.8448
---


max_axis
{'data': (10000, 1), 'axis': 0}
214470.1406
0.432
---


max_axis
{'data': (10000, 100), 'axis': 0}
214470.5469
1.8058
---


mean
{'data': (1024, 1024), 'axis': 0}
214476.4844
1.86
2.1958


mean
{'data': (10000, 1), 'axis': 0}
214474.4375
0.4254
0.0362


mean
{'data': (10000, 100), 'axis': 0}
214474.8438
1.9616
2.2664


min
{'data': (1024, 1024), 'axis': 0}
214480.9844
1.7412
1.4844


min
{'data': (10000, 1), 'axis': 0}
214478.9375
0.3492
0.0248


min
{'data': (10000, 100), 'axis': 0}
214479.3438
1.7012
1.4112


min_axis
{'data': (1024, 1024), 'axis': 0}
214485.2812
1.7702
---


min_axis
{'data': (10000, 1), 'axis': 0}
214483.2344
0.3404
---


min_axis
{'data': (10000, 100), 'axis': 0}
214483.6406
1.669
---


moments
{'data': (1024, 1024), 'axes': [0, 1]}
8677434.0
50.9642
---


moments
{'data': (10000, 1), 'axes': [0, 1]}
8677434.0
0.5358
---


moments
{'data': (10000, 100), 'axes': [0, 1]}
8677434.0
49.0232
---


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}
3194318.5
0.299
---


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}
3192261.25
0.0306
---


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}
3196241.25
0.2694
---


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}
3200532.75
0.2008
---


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}
3198475.5
0.0392
---


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}
3202455.5
0.21
---


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}
3206747.0
0.168
---


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}
3204690.0
0.018
---


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}
3208670.0
0.1958
---


multi_all_finite
{'args': [(1024, 1024)], 'num_arrays': 1}
4139356.5
0.0152
---


multi_all_finite
{'args': [(10000, 1)], 'num_arrays': 1}
4139356.5
0.0636
---


multi_all_finite
{'args': [(10000, 10)], 'num_arrays': 1}
4139356.5
0.0208
---


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}
4316615.5
1.699
---


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}
4314558.5
0.02
---


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}
4318538.5
3.399
---


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)>'}
---
0.0146
---


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)>'}
---
0.0115
---


multi_sgd_update
{'args0': '<NDArray 5x5 @cpu(0)>', 'args1': '<NDArray 5x5 @cpu(0)>', 'lrs': 0.1, 'wds': 0.2, 'out': '<NDArray 5x5 @cpu(0)>'}
---
0.0118
---


multi_sum_sq
{'args': [(1024, 1024)], 'num_arrays': 1}
4139356.5
0.013
---


multi_sum_sq
{'args': [(10000, 1)], 'num_arrays': 1}
4139356.5
0.0134
---


multi_sum_sq
{'args': [(10000, 10)], 'num_arrays': 1}
4139356.5
0.0132
---


nag_mom_update
{'weight': (1024, 1024), 'grad': (1024, 1024), 'mom': (1024, 1024), 'lr': 0.1, 'wd': 0.1, 'rescale_grad': 0.4}
3212961.25
0.3648
---


nag_mom_update
{'weight': (10000, 1), 'grad': (10000, 1), 'mom': (10000, 1), 'lr': 0.5, 'wd': 0.5, 'rescale_grad': 0.4}
3210904.25
0.0214
---


nag_mom_update
{'weight': (10000, 100), 'grad': (10000, 100), 'mom': (10000, 100), 'lr': 0.9, 'wd': 0.9, 'rescale_grad': 0.4}
3214884.25
0.2836
---


nanprod
{'data': (1024, 1024), 'axis': 0}
214489.5781
1.901
1.492


nanprod
{'data': (10000, 1), 'axis': 0}
214487.5312
0.3438
0.0244


nanprod
{'data': (10000, 100), 'axis': 0}
214487.9375
1.6876
1.4132


nansum
{'data': (1024, 1024), 'axis': 0}
214493.875
1.91
1.7302


nansum
{'data': (10000, 1), 'axis': 0}
214491.8281
0.3524
0.0266


nansum
{'data': (10000, 100), 'axis': 0}
214492.2344
1.658
1.4848


negative
{'data': (1024, 1024)}
2097.1521
0.0514
---


negative
{'data': (10000, 1)}
20.0
0.0146
---


negative
{'data': (10000, 100)}
2000.0
0.0476
---


norm
{'data': (1024, 1024), 'axis': 0}
214498.1719
2.0084
1.44


norm
{'data': (10000, 1), 'axis': 0}
214496.125
0.378
0.025


norm
{'data': (10000, 100), 'axis': 0}
214496.5312
1.7686
1.386


one_hot
{'indices': (1, 1), 'depth': 0, 'dtype': 'float32'}
---
0.0048
---


one_hot
{'indices': (1, 1), 'depth': 0, 'dtype': 'int32'}
---
0.005
---


one_hot
{'indices': (1, 1), 'depth': 0, 'dtype': 'float32'}
---
0.0048
---


ones_like
{'data': (1024, 1024)}
2097.1521
0.0268
---


ones_like
{'data': (10000, 1)}
20.0
0.0734
---


ones_like
{'data': (10000, 100)}
2000.0
0.026
---


pick
{'data': (1024, 1024), 'index': (1, 1024), 'axis': 0}
246604.3438
0.016
0.2774


pick
{'data': (10000, 1), 'index': (1, 1), 'axis': 0}
246602.2969
0.0132
0.0158


pick
{'data': (10000, 100), 'index': (1, 100), 'axis': 0}
246602.6875
0.0126
0.3116


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)>'}
---
0.0119
---


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)>'}
---
0.0111
---


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)>'}
---
0.0114
---


prod
{'data': (1024, 1024), 'axis': 0}
214502.4688
1.8732
1.4706


prod
{'data': (10000, 1), 'axis': 0}
214500.4375
0.33
0.0254


prod
{'data': (10000, 100), 'axis': 0}
214500.8281
1.6508
1.3866


radians
{'data': (1024, 1024)}
4194.3042
0.05
0.0374


radians
{'data': (10000, 1)}
20.0
0.0162
0.0108


radians
{'data': (10000, 100)}
2000.0
0.047
0.0356


random_exponential
{'shape': (1024, 1024), 'dtype': 'float16'}
1048.576
2.7968
---


random_exponential
{'shape': (10000, 1), 'dtype': 'float32'}
20.0
0.0424
---


random_exponential
{'shape': (10000, 100), 'dtype': 'float64'}
4000.0
2.6212
---


random_gamma
{'shape': (1024, 1024), 'dtype': 'float16'}
1048.576
7.1436
---


random_gamma
{'shape': (10000, 1), 'dtype': 'float32'}
20.0
0.0842
---


random_gamma
{'shape': (10000, 100), 'dtype': 'float64'}
4000.0
8.557
---


random_generalized_negative_binomial
{'shape': (1024, 1024), 'dtype': 'float16'}
1048.576
9.6646
---


random_generalized_negative_binomial
{'shape': (10000, 1), 'dtype': 'float32'}
20.0
0.113
---


random_generalized_negative_binomial
{'shape': (10000, 100), 'dtype': 'float64'}
4000.0
9.2784
---


random_negative_binomial
{'k': 1, 'p': 1, 'shape': (1024, 1024), 'dtype': 'float16'}
1048.576
8.717
---


random_negative_binomial
{'k': 1, 'p': 1, 'shape': (10000, 1), 'dtype': 'float32'}
20.0
0.0988
---


random_negative_binomial
{'k': 1, 'p': 1, 'shape': (10000, 100), 'dtype': 'float64'}
4000.0
8.1354
---


random_normal
{'shape': (1024, 1024), 'dtype': 'float16'}
1048.576
4.206
---


random_normal
{'shape': (10000, 1), 'dtype': 'float32'}
20.0
0.0464
---


random_normal
{'shape': (10000, 100), 'dtype': 'float64'}
4000.0
3.9712
---


random_pdf_dirichlet
{'sample': (2,), 'alpha': [0.0, 2.5]}
0.002
0.0126
---


random_pdf_exponential
{'sample': (2,), 'lam': [1.0, 8.5]}
0.004
0.0118
---


random_pdf_gamma
{'sample': (2,), 'alpha': [0.0, 2.5], 'beta': [1.0, 0.7]}
0.004
0.0128
---


random_pdf_generalized_negative_binomial
{'sample': (2,), 'mu': [2.0, 2.5], 'alpha': [0.0, 2.5]}
0.004
0.0128
---


random_pdf_negative_binomial
{'sample': (2,), 'k': [20, 49], 'p': [0.4, 0.77]}
0.004
0.0128
---


random_pdf_normal
{'sample': (2,), 'mu': [2.0, 2.5], 'sigma': [1.0, 3.7]}
0.004
0.0126
---


random_pdf_poisson
{'sample': (2,), 'lam': [1.0, 8.5]}
0.004
0.0118
---


random_pdf_uniform
{'sample': (2,), 'low': [0.0, 2.5], 'high': [1.0, 3.7]}
0.004
0.0134
---


random_poisson
{'shape': (1024, 1024), 'dtype': 'float16'}
1048.576
2.4058
---


random_poisson
{'shape': (10000, 1), 'dtype': 'float32'}
20.0
0.0372
---


random_poisson
{'shape': (10000, 100), 'dtype': 'float64'}
4000.0
2.278
---


random_randint
{'low': 0, 'high': 5, 'shape': (1024, 1024), 'dtype': 'int32'}
2097.1521
1.7014
---


random_randint
{'low': 0, 'high': 5, 'shape': (10000, 1), 'dtype': 'int64'}
40.0
0.0526
---


random_randint
{'low': 0, 'high': 5, 'shape': (10000, 100), 'dtype': 'int32'}
2000.0
2.5374
---


random_uniform
{'low': 0, 'high': 5, 'shape': (1024, 1024), 'dtype': 'float16'}
1048.576
1.2788
---


random_uniform
{'low': 0, 'high': 5, 'shape': (10000, 1), 'dtype': 'float32'}
20.0
0.0228
---


random_uniform
{'low': 0, 'high': 5, 'shape': (10000, 100), 'dtype': 'float64'}
8000.0
1.1772
---


ravel_multi_index
{'data': (2, 1024), 'shape': (1024, 1024)}
246608.6406
0.0124
---


ravel_multi_index
{'data': (2, 1024), 'shape': (10000, 1)}
246610.6875
0.0122
---


ravel_multi_index
{'data': (2, 1024), 'shape': (10000, 100)}
246612.7344
0.0124
---


rcbrt
{'data': (1024, 1024)}
2097.1521
1.3436
1.3696


rcbrt
{'data': (10000, 1)}
20.0
0.0232
0.0222


rcbrt
{'data': (10000, 100)}
4000.0
1.3468
1.634


reciprocal
{'data': (1024, 1024)}
2097.1521
0.048
0.051


reciprocal
{'data': (10000, 1)}
20.0
0.0114
0.0098


reciprocal
{'data': (10000, 100)}
2000.0
0.0428
0.0498


relu
{'data': (1024, 1024)}
2097.1521
0.0882
0.0894


relu
{'data': (10000, 1)}
20.0
0.012
0.0108


relu
{'data': (10000, 100)}
2000.0
0.0684
0.0862


reset_arrays
{'args': [(1024, 1024)], 'num_arrays': 1}
---
0.0044
---


reset_arrays
{'args': [(10000, 1)], 'num_arrays': 1}
---
0.0034
---


reset_arrays
{'args': [(10000, 10)], 'num_arrays': 1}
---
0.0032
---


reshape_like
{'lhs': (1024, 1024), 'rhs': (1024, 1024)}
2097.1521
1.642
---


reshape_like
{'lhs': (10000, 10), 'rhs': (10000, 10)}
200.0
0.0638
---


reshape_like
{'lhs': (10000, 1), 'rhs': (10000, 1)}
20.0
0.0176
---


rint
{'data': (1024, 1024)}
2097.1521
0.2874
---


rint
{'data': (10000, 1)}
20.0
0.0164
---


rint
{'data': (10000, 100)}
2000.0
0.2702
---


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}
3219175.75
0.3986
---


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}
3217118.5
0.026
---


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}
3221098.5
0.8536
---


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}
3225390.0
0.8296
---


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}
3223332.75
0.379
---


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}
3227312.75
1.2188
---


round
{'data': (1024, 1024)}
2097.1521
0.3074
---


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


round
{'data': (10000, 100)}
2000.0
0.2922
---


rsqrt
{'data': (1024, 1024)}
2097.1521
0.6496
0.6744


rsqrt
{'data': (10000, 1)}
20.0
0.0182
0.0164


rsqrt
{'data': (10000, 100)}
2000.0
0.6212
0.6388


sample_exponential
{'lam': [1.0, 8.5], 'shape': (1024, 1024), 'dtype': 'float16'}
2097.1521
5.5818
---


sample_exponential
{'lam': [1.0, 8.5], 'shape': (10000, 1), 'dtype': 'float32'}
40.0
0.0626
---


sample_exponential
{'lam': [1.0, 8.5], 'shape': (10000, 100), 'dtype': 'float64'}
16000.0
5.1922
---


sample_gamma
{'alpha': [0.0, 2.5], 'shape': (1024, 1024), 'dtype': 'float16', 'beta': [1.0, 0.7]}
2097.1521
18.2952
---


sample_gamma
{'alpha': [0.0, 2.5], 'shape': (10000, 1), 'dtype': 'float32', 'beta': [1.0, 0.7]}
40.0
0.1946
---


sample_gamma
{'alpha': [0.0, 2.5], 'shape': (10000, 100), 'dtype': 'float64', 'beta': [1.0, 0.7]}
8000.0
21.4168
---


sample_generalized_negative_binomial
{'mu': [2.0, 2.5], 'shape': (1024, 1024), 'dtype': 'float16', 'alpha': [0.0, 2.5]}
4194.3042
31.4892
---


sample_generalized_negative_binomial
{'mu': [2.0, 2.5], 'shape': (10000, 1), 'dtype': 'float32', 'alpha': [0.0, 2.5]}
40.0
0.339
---


sample_generalized_negative_binomial
{'mu': [2.0, 2.5], 'shape': (10000, 100), 'dtype': 'float64', 'alpha': [0.0, 2.5]}
8000.0
29.4162
---


sample_multinomial
{'data': (32, 32), 'shape': (1024, 1024), 'dtype': 'float16'}
70438.9609
169.9096
---


sample_multinomial
{'data': (32, 32), 'shape': (10000, 1), 'dtype': 'float32'}
38164.5312
1.603
---


sample_multinomial
{'data': (32, 32), 'shape': (10000, 100), 'dtype': 'float64'}
293524.5312
165.1896
---


sample_negative_binomial
{'k': [20, 49], 'shape': (1024, 1024), 'dtype': 'float16', 'p': [0.4, 0.77]}
169718.8281
373.159
---


sample_negative_binomial
{'k': [20, 49], 'shape': (10000, 1), 'dtype': 'float32', 'p': [0.4, 0.77]}
169798.8281
3.4792
---


sample_negative_binomial
{'k': [20, 49], 'shape': (10000, 100), 'dtype': 'float64', 'p': [0.4, 0.77]}
185758.8281
353.4468
---


sample_normal
{'mu': [2.0, 2.5], 'shape': (1024, 1024), 'dtype': 'float16', 'sigma': [1.0, 3.7]}
184050.2969
8.488
---


sample_normal
{'mu': [2.0, 2.5], 'shape': (10000, 1), 'dtype': 'float32', 'sigma': [1.0, 3.7]}
182033.1406
0.0724
---


sample_normal
{'mu': [2.0, 2.5], 'shape': (10000, 100), 'dtype': 'float64', 'sigma': [1.0, 3.7]}
197993.1406
8.0432
---


sample_poisson
{'lam': [1.0, 8.5], 'shape': (1024, 1024), 'dtype': 'float16'}
196284.5938
10.3098
---


sample_poisson
{'lam': [1.0, 8.5], 'shape': (10000, 1), 'dtype': 'float32'}
194227.4375
0.132
---


sample_poisson
{'lam': [1.0, 8.5], 'shape': (10000, 100), 'dtype': 'float64'}
210227.4375
10.787
---


sample_uniform
{'low': [0.0, 2.5], 'shape': (1024, 1024), 'dtype': 'float16', 'high': [1.0, 3.7]}
208518.8906
2.666
---


sample_uniform
{'low': [0.0, 2.5], 'shape': (10000, 1), 'dtype': 'float32', 'high': [1.0, 3.7]}
206501.75
0.032
---


sample_uniform
{'low': [0.0, 2.5], 'shape': (10000, 100), 'dtype': 'float64', 'high': [1.0, 3.7]}
222461.75
2.4786
---


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}
3231604.25
0.1892
---


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}
3229547.0
0.0742
---


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}
3233527.0
0.1926
---


sgd_update
{'weight': (1024, 1024), 'grad': (1024, 1024), 'lr': 0.1, 'wd': 0.1, 'rescale_grad': 0.4, 'lazy_update': 0}
3237818.5
0.167
---


sgd_update
{'weight': (10000, 1), 'grad': (10000, 1), 'lr': 0.5, 'wd': 0.5, 'rescale_grad': 0.4, 'lazy_update': 0}
3235761.5
0.0148
---


sgd_update
{'weight': (10000, 100), 'grad': (10000, 100), 'lr': 0.9, 'wd': 0.9, 'rescale_grad': 0.4, 'lazy_update': 0}
3239741.5
0.1484
---


shape_array
{'data': (1024, 1024)}
0.008
0.006
---


shape_array
{'data': (10000, 1)}
0.008
0.0052
---


shape_array
{'data': (10000, 100)}
0.008
0.0054
---


shuffle
{'data': (1024, 1024)}
2097.1521
0.9654
---


shuffle
{'data': (10000, 1)}
20.0
0.5262
---


shuffle
{'data': (10000, 100)}
2000.0
1.6004
---


sigmoid
{'data': (1024, 1024)}
2097.1521
0.6882
0.0508


sigmoid
{'data': (10000, 1)}
20.0
0.018
0.0114


sigmoid
{'data': (10000, 100)}
2000.0
0.665
0.0492


sign
{'data': (1024, 1024)}
4194.3042
0.2434
0.04


sign
{'data': (10000, 1)}
40.0
0.0174
0.0098


sign
{'data': (10000, 100)}
2000.0
0.2352
0.043


signsgd_update
{'weight': (1024, 1024), 'grad': (1024, 1024), 'lr': 0.1, 'wd': 0.1, 'rescale_grad': 0.4}
3244033.0
0.1864
---


signsgd_update
{'weight': (10000, 1), 'grad': (10000, 1), 'lr': 0.5, 'wd': 0.5, 'rescale_grad': 0.4}
3241975.75
0.015
---


signsgd_update
{'weight': (10000, 100), 'grad': (10000, 100), 'lr': 0.9, 'wd': 0.9, 'rescale_grad': 0.4}
3245955.75
0.1788
---


signum_update
{'weight': (1024, 1024), 'grad': (1024, 1024), 'mom': (1024, 1024), 'lr': 0.1, 'wd': 0.1, 'rescale_grad': 0.4}
3250247.25
0.3112
---


signum_update
{'weight': (10000, 1), 'grad': (10000, 1), 'mom': (10000, 1), 'lr': 0.5, 'wd': 0.5, 'rescale_grad': 0.4}
3248190.0
0.2088
---


signum_update
{'weight': (10000, 100), 'grad': (10000, 100), 'mom': (10000, 100), 'lr': 0.9, 'wd': 0.9, 'rescale_grad': 0.4}
3252170.0
0.325
---


sin
{'data': (1024, 1024)}
2097.1521
0.4714
0.4946


sin
{'data': (10000, 1)}
20.0
0.015
0.0134


sin
{'data': (10000, 100)}
2000.0
0.4596
0.4754


sinh
{'data': (1024, 1024)}
2097.1521
1.5124
0.9122


sinh
{'data': (10000, 1)}
20.0
0.0284
0.0176


sinh
{'data': (10000, 100)}
2000.0
1.4062
0.8826


size_array
{'data': (1024, 1024)}
0.004
0.006
---


size_array
{'data': (10000, 1)}
0.004
0.0056
---


size_array
{'data': (10000, 100)}
0.004
0.006
---


slice
{'data': (1024, 1024), 'begin': 0, 'end': 1}
246616.8281
0.0232
0.026


slice
{'data': (10000, 1), 'begin': 0, 'end': 1}
246614.7812
0.0324
0.0126


slice
{'data': (10000, 100), 'begin': 0, 'end': 1}
246615.1875
0.0314
0.0268


slice_axis
{'data': (1024, 1024), 'axis': 0, 'begin': 0, 'end': 1}
246621.125
0.0126
0.0244


slice_axis
{'data': (10000, 1), 'axis': 0, 'begin': 0, 'end': 1}
246619.0781
0.0108
0.0136


slice_axis
{'data': (10000, 100), 'axis': 0, 'begin': 0, 'end': 1}
246619.4844
0.0112
0.0254


slice_like
{'data': (1024, 1024), 'shape_like': (100, 100), 'axes': [0, 1]}
246679.2812
0.0166
0.0324


slice_like
{'data': (10000, 1), 'shape_like': (10, 1), 'axes': [0, 1]}
246659.3125
0.0146
0.0142


slice_like
{'data': (10000, 100), 'shape_like': (100, 10), 'axes': [0, 1]}
246663.2969
0.015
0.0292


smooth_l1
{'data': (1024, 1024)}
4139433.5
0.2902
0.29


smooth_l1
{'data': (10000, 1)}
4137376.5
0.0294
0.011


smooth_l1
{'data': (10000, 100)}
4141356.5
0.3582
0.2798


softmax
{'data': (1024, 1024), 'axis': 0, 'dtype': 'float16'}
402105.5625
2.1964
0.8692


softmax
{'data': (10000, 1), 'axis': 0, 'dtype': 'float32'}
401096.9688
0.4202
0.033


softmax
{'data': (10000, 100), 'axis': 0, 'dtype': 'float64'}
409076.9688
5.5678
1.7112


softmax_cross_entropy
{'data': (1024, 1024), 'label': (1024,)}
4139356.5
0.8222
---


softmin
{'data': (1024, 1024), 'axis': 0, 'dtype': 'float16'}
408222.7188
2.0126
0.8584


softmin
{'data': (10000, 1), 'axis': 0, 'dtype': 'float32'}
407214.125
0.3604
0.0294


softmin
{'data': (10000, 100), 'axis': 0, 'dtype': 'float64'}
415194.125
5.5684
1.6654


softsign
{'data': (1024, 1024)}
2097.1521
0.0476
0.0474


softsign
{'data': (10000, 1)}
20.0
0.011
0.009


softsign
{'data': (10000, 100)}
2000.0
0.046
0.0442


sort
{'data': (1024, 1024), 'axis': 0}
231217.8906
67.7188
---


sort
{'data': (10000, 1), 'axis': 0}
224986.4219
1.657
---


sort
{'data': (10000, 100), 'axis': 0}
232946.4219
80.4938
---


space_to_depth
{'data': (1, 4, 2, 4), 'block_size': 2}
233665.3438
0.0206
---


space_to_depth
{'data': (10, 25, 10, 100), 'block_size': 5}
234665.2812
0.918
---


sqrt
{'data': (1024, 1024)}
2097.1521
0.8226
0.051


sqrt
{'data': (10000, 1)}
20.0
0.0164
0.0084


sqrt
{'data': (10000, 100)}
2000.0
0.7366
0.0484


square
{'data': (1024, 1024)}
2097.1521
0.0382
0.046


square
{'data': (10000, 1)}
20.0
0.0094
0.0108


square
{'data': (10000, 100)}
2000.0
0.0374
0.1336


squeeze
{'data': (1, 1024, 1024), 'axis': 0}
4322830.0
0.351
0.591


squeeze
{'data': (32, 1, 256, 256), 'axis': 1}
4329121.5
13.4974
4.5714


stop_gradient
{'data': (1024, 1024)}
2097.1521
0.3172
---


stop_gradient
{'data': (10000, 1)}
20.0
0.0196
---


stop_gradient
{'data': (10000, 100)}
2000.0
0.4142
---


sum
{'data': (1024, 1024), 'axis': 0}
214506.7656
1.898
1.421


sum
{'data': (10000, 1), 'axis': 0}
214504.7344
0.3772
0.0212


sum
{'data': (10000, 100), 'axis': 0}
214505.125
1.6958
1.358


sum_axis
{'data': (1024, 1024), 'axis': 0}
214511.0625
1.873
---


sum_axis
{'data': (10000, 1), 'axis': 0}
214509.0312
0.3752
---


sum_axis
{'data': (10000, 100), 'axis': 0}
214509.4219
1.7654
---


swapaxes
{'data': (1024, 1024), 'dim1': 0, 'dim2': 1}
240456.7344
3.0176
---


swapaxes
{'data': (10000, 1), 'dim1': 0, 'dim2': 1}
238399.5938
0.0414
---


swapaxes
{'data': (10000, 100), 'dim1': 0, 'dim2': 1}
242379.5938
2.544
---


take
{'a': (1024, 1024), 'indices': (1, 1), 'axis': 0}
246667.4375
0.0134
0.03


tan
{'data': (1024, 1024)}
2097.1521
1.1328
0.0546


tan
{'data': (10000, 1)}
20.0
0.0268
0.014


tan
{'data': (10000, 100)}
2000.0
1.0948
0.051


tanh
{'data': (1024, 1024)}
4194.3042
1.3814
0.059


tanh
{'data': (10000, 1)}
40.0
0.0276
0.0096


tanh
{'data': (10000, 100)}
4000.0
1.3408
0.052


topk
{'data': (1024, 1024), 'axis': 0, 'k': 1, 'dtype': 'float32'}
226952.5781
59.3784
---


topk
{'data': (10000, 1), 'axis': 0, 'k': 1, 'dtype': 'int32'}
226950.5312
0.5862
---


topk
{'data': (10000, 100), 'axis': 0, 'k': 1, 'dtype': 'float32'}
226950.9219
54.2758
---


transpose
{'data': (1024, 1024), 'axes': [0, 1]}
246671.0469
0.0488
---


transpose
{'data': (10000, 1), 'axes': [0, 1]}
244613.8906
0.0418
---


transpose
{'data': (10000, 100), 'axes': [0, 1]}
248593.8906
0.0572
---


trunc
{'data': (1024, 1024)}
2097.1521
0.2644
---


trunc
{'data': (10000, 1)}
20.0
0.0166
---


trunc
{'data': (10000, 100)}
2000.0
0.2662
---


where
{'condition': (1024,), 'x': (1024, 1024), 'y': (1024, 1024)}
252956.8438
0.5552
1.1214


where
{'condition': (10000,), 'x': (10000, 1), 'y': (10000, 1)}
250899.7031
0.0242
0.0222


where
{'condition': (10000,), 'x': (10000, 100), 'y': (10000, 100)}
254879.7031
0.5366
1.0564


zeros_like
{'data': (1024, 1024)}
2097.1521
0.0288
---


zeros_like
{'data': (10000, 1)}
20.0
0.0142
---


zeros_like
{'data': (10000, 100)}
2000.0
0.0784
---