Hapi mnist accuracy drop with commit "Enable standalone executor for single-GPU training (#45913)"

[pmajchrzak]

问题描述 Please describe your issue

Commit https://github.com/PaddlePaddle/Paddle/commit/fb37628530823ad79baa0120511669205d6eba3b seems to be responsible of accuracy drop in hapi mnist model

application:

paddle_paddle/logs_paddle_paddle/build/python/paddle/tests/hapi_mnist_bf16_static.py

test run parameters:

FLAGS_use_mkldnn=true python /logs_paddle_paddle/build/python/paddle/tests/hapi_mnist_bf16_static.py --bf16=False But is also observed for training and for bf16

accuracy drop: 0,8678 -> 0,6504

---

Environment

            OS: Ubuntu18.04 with mkldnn 
 cmake version: 3.10.2
      Compiler: GCC 8.4.0

CMAKE parameters:

-DCMAKE_BUILD_TYPE=Release -DWITH_GPU=OFF -DWITH_MKLDNN=ON -DWITH_TESTING=ON -DWITH_PROFILER=OFF -DWITH_STYLE_CHECK=OFF -DON_INFER=ON -DWITH_INFERENCE_API_TEST=ON -DPY_VERSION=3.6

[paddle-bot[bot]]

您好，我们已经收到了您的问题，会安排技术人员尽快解答您的问题，请耐心等待。请您再次检查是否提供了清晰的问题描述、复现代码、环境&版本、报错信息等。同时，您也可以通过查看官网API文档、常见问题、历史Issue、AI社区来寻求解答。祝您生活愉快～

Hi! We've received your issue and please be patient to get responded. We will arrange technicians to answer your questions as soon as possible. Please make sure that you have posted enough message to demo your request. You may also check out the API，FAQ，Github Issue and AI community to get the answer.Have a nice day!

[paddle-bot[bot]]

Since you haven't replied for more than a year, we have closed this issue/pr. If the problem is not solved or there is a follow-up one, please reopen it at any time and we will continue to follow up. 由于您超过一年未回复，我们将关闭这个issue/pr。 若问题未解决或有后续问题，请随时重新打开，我们会继续跟进。