It seems that tf.scatter_nd only can run on CPU?

[amortazi]

NOTE: Only file GitHub issues for bugs and feature requests. All other topics will be closed.

For general support from the community, see StackOverflow.
To make bugs and feature requests more easy to find and organize, we close issues that are deemed
out of scope for GitHub Issues and point people to StackOverflow.

For bugs or installation issues, please provide the following information.
The more information you provide, the more easily we will be able to offer
help and advice.

What related GitHub issues or StackOverflow threads have you found by searching the web for your problem?

Environment info

Operating System:

Installed version of CUDA and cuDNN:
(please attach the output of ls -l /path/to/cuda/lib/libcud*):

If installed from binary pip package, provide:

1. A link to the pip package you installed:
2. The output from python -c "import tensorflow; print(tensorflow.__version__)".

If installed from source, provide

1. The commit hash (git rev-parse HEAD)
2. The output of bazel version

If possible, provide a minimal reproducible example (We usually don't have time to read hundreds of lines of your code)

What other attempted solutions have you tried?

Logs or other output that would be helpful

(If logs are large, please upload as attachment or provide link).

[amortazi]

Is there any way that I can run tf.scatter_nd on GPU? or is there an equivalent command?

[drpngx]

Some operations should work. Which operations are you looking for?
https://github.com/tensorflow/tensorflow/blob/master/tensorflow/core/kernels/scatter_nd_op.cc#L348

[amortazi]

I am looking for tf.scatter_nd. when I am looking at log_device_placemnt it is saying that it is running in CPU. I am using tf.device to run it on GPU. The GPU version of this:
https://github.com/tensorflow/tensorflow/blob/master/tensorflow/core/kernels/scatter_nd_op.cc#L339

[drpngx]

Could you try registering the kernel for GPUs and see what happens? I'm not super familiar with that code and I'm not sure that it would work.

[drpngx]

@rmlarsen might have some input.

[amortazi]

I am not sure how i should register it!

[amortazi]

Even I couldn't find scatter_nd_op.cc file in my tensorflow directory. Can you give me some hint where is should be located?

[drpngx]

It should just be in core/kernels/scatter_nd_op.cc. You should run the same TF_CALL macro for all types as in the CPU version, but instead register the GPU ones, if that makes sense. If you have repro instructions on how to test I could give it a shot on my end.

[amortazi]

@drpngx Unfortunately, my sensorflow directory is different from github files. I installed it by using pip. it is version 0.12 for python 2.7 in ubuntu 14 64bit. Can you guide me why it is different and I couldn't locate scatter_nd_op.cc file. :( Thanks.

[drpngx]

For that, you have to install from source. You can't do it from a pip file unfortunately. You should follow this:
https://www.tensorflow.org/get_started/os_setup#installing_from_sources

[amortazi]

@drpngx I have installed tf from source. Did I need to installed it from source again if I make changes to scatter_nd_op.cc file?
Also, should I add sth like this to that file:
#define REGISTER_SCATTER_ND_GPU(type) REGISTER_SCATTER_ND(type, GPU);
TF_CALL_NUMBER_TYPES(REGISTER_SCATTER_ND_ADD_SUB_GPU);
// TODO(simister): Re-enable all types after binary size is under control.
TF_CALL_NUMBER_TYPES(REGISTER_SCATTER_ND_UPDATE_GPU);
TF_CALL_NUMBER_TYPES(REGISTER_SCATTER_ND_GPU);

Thanks,
Ali