Originally published at: TensorFlow Performance Logging Plugin nvtx-plugins-tf Goes Public | NVIDIA Technical Blog
The new nvtx-plugins-tf library enables users to add performance logging nodes to TensorFlow graphs. (TensorFlow is an open source library widely used for training DNN—deep neural network—models). These nodes log performance data using the NVTX (NVIDIA’s Tools Extension) library. The logged performance data can then be viewed in tools such as NVIDIA Nsight Systems and NVIDIA Nsight…
1 Like
Is this still operational for Keras and TF2?
pip install nvtx-plugins failed for me
/usr/local/lib/python3.8/dist-packages/tensorflow/include/tensorflow/core/framework/tensor_types.h:177:22: error: ‘bool_vals’ was not declared in this scope
177 | for (bool b : {bool_vals…}) {
| ^~~~~~~~~
INFO: Unable to build TensorFlow plugin, will skip it.
Traceback (most recent call last):
File “/usr/lib/python3.8/distutils/unixccompiler.py”, line 117, in _compile
self.spawn(compiler_so + cc_args + [src, ‘-o’, obj] +
File “/usr/lib/python3.8/distutils/ccompiler.py”, line 910, in spawn
spawn(cmd, dry_run=self.dry_run)
File “/usr/lib/python3.8/distutils/spawn.py”, line 36, in spawn
_spawn_posix(cmd, search_path, dry_run=dry_run)
File “/usr/lib/python3.8/distutils/spawn.py”, line 157, in _spawn_posix
raise DistutilsExecError(
distutils.errors.DistutilsExecError: command ‘/usr/bin/gcc’ failed with exit status 1