Type tf.profiler
Namespace tensorflow
Public static methods
object profile(object graph, object run_meta, object op_log, string cmd, IDictionary<string, object> options)
Profile model. Tutorials and examples can be found in:
https://github.com/tensorflow/tensorflow/tree/master/tensorflow/core/profiler/README.md
Parameters
-
objectgraph - tf.Graph. If None and eager execution is not enabled, use default graph.
-
objectrun_meta - optional tensorflow.RunMetadata proto. It is necessary to to support run time information profiling, such as time and memory.
-
objectop_log - tensorflow.tfprof.OpLogProto proto. User can assign "types" to graph nodes with op_log. "types" allow user to flexibly group and account profiles using options['accounted_type_regexes'].
-
stringcmd - string. Either 'op', 'scope', 'graph' or 'code'. 'op' view organizes profile using operation type. (e.g. MatMul) 'scope' view organizes profile using graph node name scope. 'graph' view organizes profile using graph node inputs/outputs. 'code' view organizes profile using Python call stack.
-
IDictionary<string, object>options - A dict of options. See core/profiler/g3doc/options.md.
Returns
-
object - If cmd is 'scope' or 'graph', returns GraphNodeProto proto. If cmd is 'op' or 'code', returns MultiGraphNodeProto proto. Side effect: stdout/file/timeline.json depending on options['output']
object profile(object graph, object run_meta, object op_log, string cmd, ImplicitContainer<T> options)
Profile model. Tutorials and examples can be found in:
https://github.com/tensorflow/tensorflow/tree/master/tensorflow/core/profiler/README.md
Parameters
-
objectgraph - tf.Graph. If None and eager execution is not enabled, use default graph.
-
objectrun_meta - optional tensorflow.RunMetadata proto. It is necessary to to support run time information profiling, such as time and memory.
-
objectop_log - tensorflow.tfprof.OpLogProto proto. User can assign "types" to graph nodes with op_log. "types" allow user to flexibly group and account profiles using options['accounted_type_regexes'].
-
stringcmd - string. Either 'op', 'scope', 'graph' or 'code'. 'op' view organizes profile using operation type. (e.g. MatMul) 'scope' view organizes profile using graph node name scope. 'graph' view organizes profile using graph node inputs/outputs. 'code' view organizes profile using Python call stack.
-
ImplicitContainer<T>options - A dict of options. See core/profiler/g3doc/options.md.
Returns
-
object - If cmd is 'scope' or 'graph', returns GraphNodeProto proto. If cmd is 'op' or 'code', returns MultiGraphNodeProto proto. Side effect: stdout/file/timeline.json depending on options['output']
object profile_dyn(object graph, object run_meta, object op_log, ImplicitContainer<T> cmd, ImplicitContainer<T> options)
Profile model. Tutorials and examples can be found in:
https://github.com/tensorflow/tensorflow/tree/master/tensorflow/core/profiler/README.md
Parameters
-
objectgraph - tf.Graph. If None and eager execution is not enabled, use default graph.
-
objectrun_meta - optional tensorflow.RunMetadata proto. It is necessary to to support run time information profiling, such as time and memory.
-
objectop_log - tensorflow.tfprof.OpLogProto proto. User can assign "types" to graph nodes with op_log. "types" allow user to flexibly group and account profiles using options['accounted_type_regexes'].
-
ImplicitContainer<T>cmd - string. Either 'op', 'scope', 'graph' or 'code'. 'op' view organizes profile using operation type. (e.g. MatMul) 'scope' view organizes profile using graph node name scope. 'graph' view organizes profile using graph node inputs/outputs. 'code' view organizes profile using Python call stack.
-
ImplicitContainer<T>options - A dict of options. See core/profiler/g3doc/options.md.
Returns
-
object - If cmd is 'scope' or 'graph', returns GraphNodeProto proto. If cmd is 'op' or 'code', returns MultiGraphNodeProto proto. Side effect: stdout/file/timeline.json depending on options['output']