- Sort Score
- Result 10 results
- Languages All
Results 1 - 2 of 2 for XlaDynamicSlice (0.24 sec)
-
tensorflow/compiler/jit/increase_dynamism_for_auto_jit_pass.h
// shown above jit/partially_decluster_pass extracts the actual_size(...) // computation to outside the XLA cluster, causing the cluster to be versioned // only on the actual size of the XlaDynamicSlice. This avoids recompilation // due to superficial changes that don't affect tensor shapes. // // Future Work TODO(b/111210515) // ----------------------------- //
Registered: Sun Jun 16 05:45:23 UTC 2024 - Last Modified: Fri Oct 26 21:01:34 UTC 2018 - 2.2K bytes - Viewed (0) -
tensorflow/compiler/jit/xla_ops_on_regular_devices.cc
XlaCompileOnDemandOp); \ REGISTER_KERNEL_BUILDER( \ Name("XlaDynamicSlice").HostMemory("size_indices").Device(DEVICE), \ XlaCompileOnDemandOp); \ REGISTER_KERNEL_BUILDER(Name("XlaDynamicUpdateSlice").Device(DEVICE), \
Registered: Sun Jun 16 05:45:23 UTC 2024 - Last Modified: Fri Aug 19 19:55:14 UTC 2022 - 8.8K bytes - Viewed (0)