tf.compat.v1.GPUOptions.Experimental

A ProtocolMessage

Attributes
`collective_ring_order`	`string collective_ring_order`
`disallow_retry_on_allocation_failure`	`bool disallow_retry_on_allocation_failure`
`gpu_host_mem_disallow_growth`	`bool gpu_host_mem_disallow_growth`
`gpu_host_mem_limit_in_mb`	`float gpu_host_mem_limit_in_mb`
`internal_fragmentation_fraction`	`double internal_fragmentation_fraction`
`kernel_tracker_max_bytes`	`int32 kernel_tracker_max_bytes`
`kernel_tracker_max_interval`	`int32 kernel_tracker_max_interval`
`kernel_tracker_max_pending`	`int32 kernel_tracker_max_pending`
`num_dev_to_dev_copy_streams`	`int32 num_dev_to_dev_copy_streams`
`timestamped_allocator`	`bool timestamped_allocator`
`use_cuda_malloc_async`	`bool use_cuda_malloc_async`
`use_unified_memory`	`bool use_unified_memory`
`virtual_devices`	`repeated VirtualDevices virtual_devices`

Child Classes

Except as otherwise noted, the content of this page is licensed under the Creative Commons Attribution 4.0 License, and code samples are licensed under the Apache 2.0 License. For details, see the Google Developers Site Policies. Java is a registered trademark of Oracle and/or its affiliates. Some content is licensed under the numpy license.

Last updated 2023-10-06 UTC.