I am currently working on a school project that consists on analyzing the performance of Machine Learning models in order to assess their scalability and spot potential bottlenecks in their framework.

I am looking for a performance analysis tool that works for Python programs written using Tensorflow's library. My system is running 4 CPU's AMD EPYC 7451 on "Buster" version of Debian. I am willing to visualize the execution trace in order to understand the program's performance.

  • Lock contention and lack of parallelism will be solved by various different tools. Same for the others. So, please ask a question for each, specifying the OS (choose one) and language (choose one) and ML libraries for each. Thanks! Also, please read meta.softwarerecs.stackexchange.com/questions/336/… – Nicolas Raoul Apr 18 '19 at 7:20
  • The question was edited @NicolasRaoul. Thanks for the review. – J.Youssef Apr 19 '19 at 11:21
  • "visualize the execution trace" is still extremely vague. – Nicolas Raoul Apr 19 '19 at 11:22

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service, privacy policy and cookie policy

Browse other questions tagged or ask your own question.