Tensorflow: performing operations on a specific processor core

Question

Currently, you can specify which processor or graphics processor to use with the tf.device (...) function for certain operating systems, but is there where you can specify the processor core?

Answer 1

There is currently no API for binding operating systems to a specific kernel, although this will create a good function request . You can approach this function by creating several processor devices, each of which has a single-threaded file system, but this does not guarantee the preservation of the local pinning solution:

with tf.device("/cpu:4"): # ... with tf.device("/cpu:7"): # ... with tf.device("/cpu:0"): # ... config = tf.ConfigProto(device_count={"CPU": 8}, inter_op_parallelism_threads=1, intra_op_parallelism_threads=1) sess = tf.Session(config=config)