- Sort Score
- Result 10 results
- Languages All
Results 1 - 1 of 1 for MarkForCompilationPass (0.41 sec)
-
tensorflow/compiler/jit/device_util.h
// all operations in the cluster that XLA:CPU can compile. // // We provide the `allow_mixing_unknown_and_cpu` knob so that we can do both of // the following things: // // - Let MarkForCompilationPass not inject CPU-placed operations into clusters // that will run on unknown devices (because the unknown XLA backend may not // support every operation supported by CPU).
Registered: Sun Jun 16 05:45:23 UTC 2024 - Last Modified: Wed May 15 17:18:31 UTC 2024 - 7.1K bytes - Viewed (0)