Created by: luotao1

CI log

• test_slim_int8_googlenet

[21:44:36]2019-08-27 21:44:34,027-INFO: FP32: avg top1 accuracy: 0.7200
[21:44:36]2019-08-27 21:44:34,027-INFO: INT8: avg top1 accuracy: 0.7000
[21:44:36]F
[21:44:36]======================================================================
[21:44:36]FAIL: test_compression (__main__.TestMKLDNNPostTrainingQuantStrategy)
[21:44:36]----------------------------------------------------------------------
[21:44:36]Traceback (most recent call last):
[21:44:36]  File "/paddle/python/paddle/fluid/contrib/slim/tests/test_mkldnn_int8_quantization_strategy.py", line 263, in test_compression
[21:44:36]    accuracy_diff_threshold)
[21:44:36]  File "/paddle/python/paddle/fluid/contrib/slim/tests/test_mkldnn_int8_quantization_strategy.py", line 228, in _compare_accuracy
[21:44:36]    assert fp32_acc1 - int8_acc1 <= threshold
[21:44:36]AssertionError
[21:44:36]
[21:44:36]----------------------------------------------------------------------
[21:44:36]Ran 1 test in 33.041s

• test_transpose_int8_mkldnn_op

[21:43:04]C++ Call stacks: 
[21:43:04]op transpose2 does not have kernel for data_type[double]:data_layout[ANY_LAYOUT]:place[CPUPlace]:library_type[PLAIN] at [/paddle/paddle/fluid/framework/operator.cc:1003]
[21:43:04]PaddlePaddle Call Stacks: 
[21:43:04]0       0x7f7b5cf8870fp void paddle::platform::EnforceNotMet::Init<std::string>(std::string, char const*, int) + 255
[21:43:04]1       0x7f7b5cf88c5bp paddle::platform::EnforceNotMet::EnforceNotMet(std::string const&, char const*, int) + 139

• test_analyzer_seq_pool1

[21:39:11]I0827 21:39:09.997370 74239 analysis_predictor.cc:475] == optimize end ==
[21:39:11]I0827 21:39:09.998457 74239 tester_helper.h:332] Thread 0, number of threads 1, run 1 times...
[21:39:11]W0827 21:39:09.999147 74239 naive_executor.cc:43] The NaiveExecutor can not work properly if the cmake flag ON_INFER is not set.
[21:39:11]W0827 21:39:09.999157 74239 naive_executor.cc:45] Unlike the training phase, all the scopes and variables will be reused to save the allocation overhead.
[21:39:11]W0827 21:39:09.999162 74239 naive_executor.cc:48] Please re-compile the inference library by setting the cmake flag ON_INFER=ON if you are running Paddle Inference
[21:39:11]I0827 21:39:10.004389 74239 helper.h:322] ====== threads: 1, thread id: 0 ======
[21:39:11]I0827 21:39:10.004412 74239 helper.h:324] ====== batch size: 1, iterations: 1, repetitions: 1 ======
[21:39:11]I0827 21:39:10.004423 74239 helper.h:326] ====== batch latency: 5.90201ms, number of samples: 1, sample latency: 5.90201ms, fps: 169.434, data type: float ======
[21:39:11]F0827 21:39:10.012611 74239 tester_helper.h:119] Check failed: std::abs(pdata_ref[j] - pdata[j]) <= FLAGS_accuracy (0.03125 vs. 0.001) 
[21:39:11]*** Check failure stack trace: ***

• test_analyzer_int8_googlenet

[21:38:54]I0827 21:38:52.514907 74251 tester_helper.h:455] --- Accuracy summary --- 
[21:38:54]I0827 21:38:52.514912 74251 tester_helper.h:456] Accepted top1_accuracy drop threshold: 0.01. (condition: (FP32_top1_accuracy  - INT8_top1_accuracy ) <= threshold)
[21:38:54]I0827 21:38:52.514920 74251 tester_helper.h:460] FP32: avg top1_accuracy 0.7200
[21:38:54]I0827 21:38:52.514926 74251 tester_helper.h:462] INT8: avg top1_accuracy 0.7000
[21:38:54]F0827 21:38:52.514963 74251 tester_helper.h:527] Check failed: avg_acc_ref - avg_acc_quant <= FLAGS_quantized_accuracy (0.02 vs. 0.01) 

http://ci.paddlepaddle.org/viewLog.html?tab=buildLog&buildTypeId=Paddle_PrCiCoverage&buildId=149605

Error machine

processor	: 27
vendor_id	: GenuineIntel
cpu family	: 6
model		: 85
model name	: Intel(R) Xeon(R) Gold 5117 CPU @ 2.00GHz
stepping	: 4
microcode	: 0x2000043
cpu MHz		: 2000.000
cache size	: 19712 KB
physical id	: 1
siblings	: 14
core id		: 14
cpu cores	: 14
apicid		: 60
initial apicid	: 60
fpu		: yes
fpu_exception	: yes
cpuid level	: 22
wp		: yes
flags		: fpu vme de pse tsc msr pae mce cx8 apic sep mtrr pge mca cmov pat pse36 clflush dts acpi mmx fxsr sse sse2 ss ht tm pbe syscall nx pdpe1gb rdtscp lm constant_tsc arch_perfmon pebs bts rep_good nopl xtopology nonstop_tsc aperfmperf eagerfpu pni pclmulqdq dtes64 ds_cpl vmx smx est tm2 ssse3 fma cx16 xtpr pdcm pcid dca sse4_1 sse4_2 x2apic movbe popcnt tsc_deadline_timer aes xsave avx f16c rdrand lahf_lm abm 3dnowprefetch arat epb pln pts dtherm tpr_shadow vnmi flexpriority ept vpid fsgsbase tsc_adjust bmi1 hle avx2 smep bmi2 erms invpcid rtm cqm mpx avx512f avx512dq rdseed adx smap clflushopt avx512cd avx512bw avx512vl xsaveopt xsavec xgetbv1 xsaves cqm_llc cqm_occup_llc
bogomips	: 4004.54
clflush size	: 64
cache_alignment	: 64
address sizes	: 46 bits physical, 48 bits virtual
power management:

successful machine

processor	: 15
vendor_id	: GenuineIntel
cpu family	: 6
model		: 85
model name	: Intel(R) Xeon(R) Gold 5117 CPU @ 2.00GHz
stepping	: 4
microcode	: 0x1
cpu MHz		: 2000.071
cache size	: 4096 KB
physical id	: 15
siblings	: 1
core id		: 0
cpu cores	: 1
apicid		: 15
initial apicid	: 15
fpu		: yes
fpu_exception	: yes
cpuid level	: 13
wp		: yes
flags		: fpu vme de pse tsc msr pae mce cx8 apic sep mtrr pge mca cmov pat pse36 clflush mmx fxsr sse sse2 ss syscall nx pdpe1gb rdtscp lm constant_tsc arch_perfmon rep_good nopl eagerfpu pni pclmulqdq ssse3 fma cx16 pcid sse4_1 sse4_2 x2apic movbe popcnt tsc_deadline_timer aes xsave avx f16c rdrand hypervisor lahf_lm abm 3dnowprefetch arat fsgsbase tsc_adjust bmi1 hle avx2 smep bmi2 erms invpcid rtm mpx avx512f rdseed adx smap avx512cd xsaveopt xsavec xgetbv1
bogomips	: 4015.45
clflush size	: 64
cache_alignment	: 64
address sizes	: 40 bits physical, 48 bits virtual
power management: