Save the date! Google I/O returns May 18-20 Register now

tf.config.threading.get_intra_op_parallelism_threads

TensorFlow 2 version View source on GitHub

Get number of threads used within an individual op for parallelism.

Certain operations like matrix multiplication and reductions can utilize parallel threads for speed ups. A value of 0 means the system picks an appropriate number.

Number of parallel threads