tf.compat.v1.profiler.MultiGraphNodeProto

A ProtocolMessage

Attributes
`accelerator_exec_micros`	`int64 accelerator_exec_micros`
`children`	`repeated MultiGraphNodeProto children`
`cpu_exec_micros`	`int64 cpu_exec_micros`
`exec_micros`	`int64 exec_micros`
`float_ops`	`int64 float_ops`
`graph_nodes`	`repeated GraphNodeProto graph_nodes`
`name`	`string name`
`output_bytes`	`int64 output_bytes`
`parameters`	`int64 parameters`
`peak_bytes`	`int64 peak_bytes`
`requested_bytes`	`int64 requested_bytes`
`residual_bytes`	`int64 residual_bytes`
`total_accelerator_exec_micros`	`int64 total_accelerator_exec_micros`
`total_cpu_exec_micros`	`int64 total_cpu_exec_micros`
`total_exec_micros`	`int64 total_exec_micros`
`total_float_ops`	`int64 total_float_ops`
`total_output_bytes`	`int64 total_output_bytes`
`total_parameters`	`int64 total_parameters`
`total_peak_bytes`	`int64 total_peak_bytes`
`total_requested_bytes`	`int64 total_requested_bytes`
`total_residual_bytes`	`int64 total_residual_bytes`

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 2024-04-26 UTC.

tf.compat.v1.profiler.MultiGraphNodeProtoStay organized with collections Save and categorize content based on your preferences.