Skip to content

  • 体验新版
    • 正在加载...
  • 登录
  • PaddlePaddle
  • Paddle
  • Issue
  • #16316

P
Paddle
  • 项目概览

PaddlePaddle / Paddle
大约 2 年 前同步成功

通知 2325
Star 20933
Fork 5424
  • 代码
    • 文件
    • 提交
    • 分支
    • Tags
    • 贡献者
    • 分支图
    • Diff
  • Issue 1423
    • 列表
    • 看板
    • 标记
    • 里程碑
  • 合并请求 543
  • Wiki 0
    • Wiki
  • 分析
    • 仓库
    • DevOps
  • 项目成员
  • Pages
P
Paddle
  • 项目概览
    • 项目概览
    • 详情
    • 发布
  • 仓库
    • 仓库
    • 文件
    • 提交
    • 分支
    • 标签
    • 贡献者
    • 分支图
    • 比较
  • Issue 1,423
    • Issue 1,423
    • 列表
    • 看板
    • 标记
    • 里程碑
  • 合并请求 543
    • 合并请求 543
  • Pages
  • 分析
    • 分析
    • 仓库分析
    • DevOps
  • Wiki 0
    • Wiki
  • 成员
    • 成员
  • 收起侧边栏
  • 动态
  • 分支图
  • 创建新Issue
  • 提交
  • Issue看板
已关闭
开放中
Opened 3月 20, 2019 by saxon_zh@saxon_zhGuest

test_analyzer_transformer random fails on compare

Created by: luotao1

http://ci.paddlepaddle.org/viewLog.html?buildId=72226&tab=buildLog&buildTypeId=Paddle_PrCiPython35Night&logTab=tree&filter=all&_focus=13745

[23:57:29][ RUN      ] Analyzer_Transformer.compare
[23:57:29]I0319 23:55:10.920542 17062 analyzer_transformer_tester.cc:176] The number of samples to be test: 8
[23:57:29]I0319 23:55:10.922397 17062 tester_helper.h:65] AnalysisConfig {
[23:57:29]  NativeConfig {
[23:57:29]    PaddlePredictor::Config {
[23:57:29]      model_dir: 
[23:57:29]    }
[23:57:29]    use_gpu: 0
[23:57:29]    device: 0
[23:57:29]    fraction_of_gpu_memory: 0.00219227
[23:57:29]    specify_input_name: 1
[23:57:29]    cpu_num_threads: 1
[23:57:29]  }
[23:57:29]  prog_file: /root/.cache/inference_demo/transformer/model/model
[23:57:29]  param_file: /root/.cache/inference_demo/transformer/model/params
[23:57:29]  enable_ir_optim: 1
[23:57:29]  enable_ir_optim: 1
[23:57:29]  use_feed_fetch_ops: 1
[23:57:29]  use_tensorrt: 0
[23:57:29]  use_mkldnn: 0
[23:57:29]}
[23:57:29]W0319 23:55:10.925616 17062 init.cc:153] AVX is available, Please re-compile on local machine
[23:57:29]I0319 23:55:11.322595 17062 tester_helper.h:278] Running thread 0, warm up run...
[23:57:29]I0319 23:55:44.447240 17062 helper.h:270] ====== batch_size: 8, repeat: 1, threads: 1, thread id: 0, latency: 33124.5ms, fps: 0.0301891 ======
[23:57:29]I0319 23:55:44.447353 17062 tester_helper.h:301] Thread 0 run 1 times...
[23:57:29]I0319 23:56:17.681612 17062 helper.h:270] ====== batch_size: 8, repeat: 1, threads: 1, thread id: 0, latency: 33234.2ms, fps: 0.0300895 ======
[23:57:29]W0319 23:56:17.688809 17062 init.cc:153] AVX is available, Please re-compile on local machine
[23:57:29]--- Running analysis [ir_graph_build_pass]
[23:57:29]--- Running analysis [ir_analysis_pass]
[23:57:29]--- Running IR pass [infer_clean_graph_pass]
[23:57:29]--- Running IR pass [attention_lstm_fuse_pass]
[23:57:29]--- Running IR pass [seqpool_concat_fuse_pass]
[23:57:29]--- Running IR pass [seqconv_eltadd_relu_fuse_pass]
[23:57:29]--- Running IR pass [fc_lstm_fuse_pass]
[23:57:29]--- Running IR pass [mul_lstm_fuse_pass]
[23:57:29]--- Running IR pass [fc_gru_fuse_pass]
[23:57:29]--- Running IR pass [mul_gru_fuse_pass]
[23:57:29]--- Running IR pass [seq_concat_fc_fuse_pass]
[23:57:29]--- Running IR pass [fc_fuse_pass]
[23:57:29]---  detected 12 subgraphs
[23:57:29]--- Running IR pass [repeated_fc_relu_fuse_pass]
[23:57:29]--- Running IR pass [squared_mat_sub_fuse_pass]
[23:57:29]--- Running IR pass [conv_bn_fuse_pass]
[23:57:29]--- Running IR pass [conv_eltwiseadd_bn_fuse_pass]
[23:57:29]--- Running IR pass [is_test_pass]
[23:57:29]--- Running IR pass [identity_scale_op_clean_pass]
[23:57:29]--- Running analysis [ir_params_sync_among_devices_pass]
[23:57:29]--- Running analysis [ir_graph_to_program_pass]
[23:57:29]I0319 23:56:24.455240 17062 analysis_predictor.cc:396] == optimize end ==
[23:57:29]I0319 23:56:24.456599 17062 tester_helper.h:278] Running thread 0, warm up run...
[23:57:29]W0319 23:56:24.456904 17062 naive_executor.cc:43] The NaiveExecutor can not work properly if the cmake flag ON_INFER is not set.
[23:57:29]W0319 23:56:24.456938 17062 naive_executor.cc:45] Unlike the training phase, all the scopes and variables will be reused to save the allocation overhead.
[23:57:29]W0319 23:56:24.456956 17062 naive_executor.cc:48] Please re-compile the inference library by setting the cmake flag ON_INFER=ON if you are running Paddle Inference
[23:57:29]I0319 23:56:56.635799 17062 helper.h:270] ====== batch_size: 8, repeat: 1, threads: 1, thread id: 0, latency: 32179.1ms, fps: 0.031076 ======
[23:57:29]I0319 23:56:56.635892 17062 tester_helper.h:301] Thread 0 run 1 times...
[23:57:29]W0319 23:56:56.636173 17062 naive_executor.cc:43] The NaiveExecutor can not work properly if the cmake flag ON_INFER is not set.
[23:57:29]W0319 23:56:56.636209 17062 naive_executor.cc:45] Unlike the training phase, all the scopes and variables will be reused to save the allocation overhead.
[23:57:29]W0319 23:56:56.636229 17062 naive_executor.cc:48] Please re-compile the inference library by setting the cmake flag ON_INFER=ON if you are running Paddle Inference
[23:57:29]I0319 23:57:28.508708 17062 helper.h:270] ====== batch_size: 8, repeat: 1, threads: 1, thread id: 0, latency: 31872.8ms, fps: 0.0313748 ======
[23:57:29]/paddle/paddle/fluid/inference/tests/api/tester_helper.h:82: Failure
[23:57:29]      Expected: size
[23:57:29]      Which is: 645
[23:57:29]To be equal to: ref_size
[23:57:29]      Which is: 666
[23:57:29]/paddle/paddle/fluid/inference/tests/api/tester_helper.h:89: Failure
[23:57:29]      Expected: pdata_ref[j]
[23:57:29]      Which is: 6444
[23:57:29]To be equal to: pdata[j]
[23:57:29]      Which is: 23118
[23:57:29]/paddle/paddle/fluid/inference/tests/api/tester_helper.h:89: Failure
[23:57:29]      Expected: pdata_ref[j]
[23:57:29]      Which is: 10
[23:57:29]To be equal to: pdata[j]
[23:57:29]      Which is: 855
指派人
分配到
无
里程碑
无
分配里程碑
工时统计
无
截止日期
无
标识: paddlepaddle/Paddle#16316
渝ICP备2023009037号

京公网安备11010502055752号

网络110报警服务 Powered by GitLab CE v13.7
开源知识
Git 入门 Pro Git 电子书 在线学 Git
Markdown 基础入门 IT 技术知识开源图谱
帮助
使用手册 反馈建议 博客
《GitCode 隐私声明》 《GitCode 服务条款》 关于GitCode
Powered by GitLab CE v13.7