How can I run Tensorflow on one single core, single thread CPP?

Question

I am trying to restrict the number of threads that TensorFlow spawns. In python, I understand we need to use the following steps as pointed out Here. I was trying to do the same in CPP, but it doesn't seem that straight forward. Questions:

1. How to modify intra_op_parallelism_threads and inter_op_parallelism_threads correctly?
2. How to modify the device_count to control the core as well?

SessionOptions options;
ConfigProto* config = &options.config;
string key = "CPU";
//not sure if this is the correct way to do it.
(*config->mutable_device_count())[key] = 1; 
config->set_inter_op_parallelism_threads(1);
config->set_intra_op_parallelism_threads(1);

Answer 1

Answer to question 1:

tensorflow::SessionOptions options;
tensorflow::ConfigProto & config = options.config;
config.set_inter_op_parallelism_threads(1);
config.set_intra_op_parallelism_threads(1);
session->reset(tensorflow::NewSession(options));

This will reduce the total number of threads (but not to 1) generated by TensorFlow. The total number of threads generated by TensorFlow will still be multiple, depending on the number of cores in the CPU. In most of the cases, only one thread will be active while others will be in sleeping mode. I don't think it is possible to have a single-threaded TensorFlow.

The following github issues support my point of view. https://github.com/tensorflow/tensorflow/issues/33627 https://github.com/usnistgov/frvt/issues/30

Answer 2

Answer to 1 as Fisa pointed out is correct. With just minor adjustment as config is a pointer.

 SessionOptions options;
ConfigProto* config = &options.config;
//single thread control//
config->set_inter_op_parallelism_threads(1);
config->set_intra_op_parallelism_threads(1);    
fSession.reset(NewSession(options));