'Tensorflow profiler tutorial Metrics
I have tried for a fews days to use Tensorflow profiler to measure resource usage. But I can't understand the interface in order to get the data I need. I think it has something to do with "Metrics" option in the trace viewer, but it is not clear to me how to get information about CPU, Memory, Disk usage and etc from the interface in a certain interval. Everything appears to be in "ns". If you happen to know a tutorial that goes deep in the trace viewer, please. Let me know.
If this kind of profiling is not possible, also, let me know. This image shows the "Metrics" option that I refer to
Please help.
Sources
This article follows the attribution requirements of Stack Overflow and is licensed under CC BY-SA 3.0.
Source: Stack Overflow
| Solution | Source |
|---|