diff options
| author |  Mikalai Drabovich <nick.drabovich@amd.com> | 2018-01-26 11:27:28 -0800 |
|---|---|---|
| committer |  Rasmus Munk Larsen <rmlarsen@google.com> | 2018-01-26 11:27:28 -0800 |
| commit | a169db5d066efd696d80fd989091a54cc4fe5baf (patch) | |
| tree | 318284287e890d855bbfe98ed57f9e5a48a85467 | |
| parent | a12bb2e5719e3cd17916690cc8caef0ec7bcdf4d (diff) | |
Update README.md (#16437)
* Update README.md
-png, --sample_index options are not available in google-perftools (2.4-0ubuntu5.16.04.1).
Also, since Ubuntu 16.04 wrongly recommends to install pprof from 'tau' package
"
The program 'pprof' is currently not installed. You can install it by typing:
sudo apt install tau
"
the typical user command should probably be
google-pprof --pdf --nodecount=100 <filename>
* Add `google-perftools` installation Note.
| -rw-r--r-- | tensorflow/core/profiler/README.md | 5 |
1 files changed, 3 insertions, 2 deletions
diff --git a/tensorflow/core/profiler/README.md b/tensorflow/core/profiler/README.md index 9e628b1065..7997bdfa05 100644 --- a/tensorflow/core/profiler/README.md +++ b/tensorflow/core/profiler/README.md @@ -240,8 +240,9 @@ Open a Chrome browser, enter URL chrome://tracing and load the timeline file. # can also generate memory profile using `-select bytes` tfprof> code -select accelerator_micros -max_depth 100000 -output pprof:outfile=<filename> -trim_name_regexes .*apply_op.* -# Use pprof to visualize the generated file. -pprof -png --nodecount=100 --sample_index=1 <filename> +# Use google-pprof, from the google-perftools package to visualize the generated file. +# On Ubuntu you can install it with `apt-get install it google-perftools`. +google-pprof --pdf --nodecount=100 <filename> ```  |