Deepesh Nathani comments

Results 8 comments of


                                            Deepesh Nathani

Add GRPC Interceptors to TF Serving to capture transport-level overheads

Hello, any progress on this request? Or if you can point to some code where we can add the custom client interceptors. That will be great

Memory usage of TensorFlow models

@singhniraj08 Any pointers on above?

Memory usage of TensorFlow models

Hey @singhniraj08 Can you point us to any workarounds in the meantime?

Memory usage of TensorFlow models

Thanks @singhniraj08. Without a good estimate this is causing regressions in our hosts where we are going OOM easily with no way to tell which model is the issue

Memory usage of TensorFlow models

@singhniraj08 Will the memory profile tool gets populated for profiling on CPUs? I havent see this getting populated on cpu profiles

Memory usage of TensorFlow models

@singhniraj08 This only gives us what tensorflow op is taking how much time right? Not the amount of memory it occupies on CPU.

Display full message in GRPC exception log

@singhniraj08 @bmzhao Is there an update to this issue?

TensorFlow profiler running into OOM issue on GPU

Hi Tensorflow team Can you help us with above? Is there a way to sample TensorFlow profiling on GPUs? This is blocking us from collecting any traces greater than 10s