a .Sic@sPdZddlZddlmZddlmZdZedgdGdd d eZd d Z dS) z1Trace allows the profiler to trace Python events.N)_pywrap_traceme) tf_exportFzprofiler.experimental.Trace)v1c@s0eZdZdZddZddZddZdd Zd S) TraceaContext manager that generates a trace event in the profiler. A trace event will start when entering the context, and stop and save the result to the profiler when exiting the context. Open TensorBoard Profile tab and choose trace viewer to view the trace event in the timeline. Trace events are created only when the profiler is enabled. More information on how to use the profiler can be found at https://tensorflow.org/guide/profiler Example usage: ```python tf.profiler.experimental.start('logdir') for step in range(num_steps): # Creates a trace event for each training step with the step number. with tf.profiler.experimental.Trace("Train", step_num=step, _r=1): train_fn() tf.profiler.experimental.stop() ``` cKs$trtj|fi||_nd|_dS)aCCreates a trace event in the profiler. Args: name: The name of the trace event. **kwargs: Keyword arguments added to the trace event. Both the key and value are of types that can be converted to strings, which will be interpreted by the profiler according to the traceme name. Example usage: ```python tf.profiler.experimental.start('logdir') for step in range(num_steps): # Creates a trace event for each training step with the # step number. with tf.profiler.experimental.Trace("Train", step_num=step): train_fn() tf.profiler.experimental.stop() ``` The example above uses the keyword argument "step_num" to specify the training step being traced. N)enabledrZTraceMe_traceme)selfnamekwargsr \/var/www/html/django/DPS/env/lib/python3.9/site-packages/tensorflow/python/profiler/trace.py__init__2szTrace.__init__cCs|SNr )rr r r __enter__SszTrace.__enter__cKs |jr|r|jjfi|dS)aSets metadata in this trace event. Args: **kwargs: metadata in key-value pairs. This method enables setting metadata in a trace event after it is created. Example usage: ```python def call(function): with tf.profiler.experimental.Trace("call", function_name=function.name) as tm: binary, in_cache = jit_compile(function) tm.set_metadata(in_cache=in_cache) execute(binary) ``` In this example, we want to trace how much time spent on calling a function, which includes compilation and execution. The compilation can be either getting a cached copy of the binary or actually generating the binary, which is indicated by the boolean "in_cache" returned by jit_compile(). We need to use set_metadata() to pass in_cache because we did not know the in_cache value when the trace was created (and we cannot create the trace after jit_compile(), because we want to measure the entire duration of call()). N)rZ SetMetadata)rr r r r set_metadataWs zTrace.set_metadatacCs|jr|jdSr)rZStop)rexc_typeexc_valexc_tbr r r __exit__yszTrace.__exit__N)__name__ __module__ __qualname____doc__r rrrr r r r rs !"rc sFtr4}t|dd}|s(t|dd}t||Sfdd}|S)a.Decorator alternative to `with Trace(): ...`. It's faster. Args: trace_name: The name of the trace event, or a callable to be traced, in which case the name is inferred from qualname or name of the callable. **trace_kwargs: Keyword arguments added to the trace event. Both the key and value are of types that can be converted to strings, which will be interpreted by the profiler according to the traceme name. Returns: A decorator that can wrap a function and apply `Trace` scope if needed, or a decorated function if used as a decorator directly. Example usage: ```python @trace_wrapper('trace_name') def func(x, y, z): pass # code to execute and apply `Trace` if needed. # Equivalent to # with Trace('trace_name'): # func(1, 2, 3) func(1, 2, 3) ``` or ```python @trace_wrapper def func(x, y, z): pass # code to execute and apply `Trace` if needed. # Equivalent to # with Trace(func.__qualname__): # func(1, 2, 3) func(1, 2, 3) ``` rNrzunknown functioncstfdd}|S)NcsPtrBtfi|i|WdS1s80Y|i|Sr)rr)argsr )func trace_kwargs trace_namer r wrappeds,z5trace_wrapper..inner_wrapper..wrapped) functoolswraps)rrrr)rr inner_wrappersz$trace_wrapper..inner_wrapper)callablegetattr trace_wrapper)rrrr r!r r r r$~s*    r$) rrZ#tensorflow.python.profiler.internalr tensorflow.python.util.tf_exportrrobjectrr$r r r r s   b