`ivddlZddlZddlZddlZddlm Z GddZ didfddej ddd Z d Z dS) N)runtimeceZdZdZdZdefdZedej fdZ edej fdZ e dZ d d Zd Zd S) _PerfCaseResulta1 An obscure object encompassing timing results recorded by :func:`~cupyx.profiler.benchmark`. Simple statistics can be obtained by converting an instance of this class to a string. .. warning:: This API is currently experimental and subject to change in future releases. c|jdksJ|jdt|dzksJ|jddksJ||_||_||_dS)Nr)ndimshapelenname_ts_devices)selfr tsdevicess h/home/jaya/work/projects/VOICE-AGENT/VIET/agent-env/lib/python3.11/site-packages/cupyx/profiler/_time.py__init__z_PerfCaseResult.__init__s_w!||||x{c'llQ.....x{Q  returnc.|dS)z Returns a string representation of the object. Returns: str: A string representation of the object. Tshow_gputo_strrs r__repr__z_PerfCaseResult.__repr__s {{D{)))rc|jdS)zA :class:`numpy.ndarray` of shape ``(n_repeat,)``, holding times spent on CPU in seconds. These values are delta of the host-side performance counter (:func:`time.perf_counter`) between each repeat step. rr rs r cpu_timesz_PerfCaseResult.cpu_times%sx{rc |jddS)zA :class:`numpy.ndarray` of shape ``(len(devices), n_repeat)``, holding times spent on GPU in seconds. These values are measured using ``cudaEventElapsedTime`` with events recorded before/after each repeat step. rNrrs r gpu_timesz_PerfCaseResult.gpu_times/sx|rcJ|jdksJ|jdksJ|dz}d||}|jdkrP|d|||z }|S)Nrrg.Az {}: {:9.03f} usz2 +/- {:6.03f} (min: {:9.03f} / max: {:9.03f}) us)r sizeformatmeanstdminmax) device_namett_usss r_to_str_per_itemz _PerfCaseResult._to_str_per_item9sv{{{{vzzzz3w ! ( (diikk B B 6A:: ELL DHHJJ 44 4ArFc n|d|jdg}|rdt|jD]O\}}||d||jd|zPd|jd|S)NCPUrzGPU-{}rz {:<20s}:{} )r-r enumeraterappendr$r join)rrresultsids rrz_PerfCaseResult.to_strEs(( <<=  <!$-00 < <1))(//!*<*<*.(1q5/;;<<<<""49chhw.?.?@@@rc.|dS)NTrrrs r__str__z_PerfCaseResult.__str__Ns{{D{)))rN)F)__name__ __module__ __qualname____doc__rstrrproperty_numpyndarrayrr! staticmethodr-rr8rrrr s   *#****6>X6>X  \ AAAA*****rrrBi' )r n_warmup max_durationrc ||j}|tjf}t |st dt |tst dt |tst dt |tst dt |tst dt |tst dtj |st dt |tst d t||||||||S) aO Timing utility for measuring time spent by both CPU and GPU. This function is a very convenient helper for setting up a timing test. The GPU time is properly recorded by synchronizing internal streams. As a result, to time a multi-GPU function all participating devices must be passed as the ``devices`` argument so that this helper knows which devices to record. A simple example is given as follows: .. code-block:: py import cupy as cp from cupyx.profiler import benchmark def f(a, b): return 3 * cp.sin(-a) * b a = 0.5 - cp.random.random((100,)) b = cp.random.random((100,)) print(benchmark(f, (a, b), n_repeat=1000)) Args: func (callable): a callable object to be timed. args (tuple): positional arguments to be passed to the callable. kwargs (dict): keyword arguments to be passed to the callable. n_repeat (int): number of times the callable is called. Increasing this value would improve the collected statistics at the cost of longer test time. name (str): the function name to be reported. If not given, the callable's ``__name__`` attribute is used. n_warmup (int): number of times the callable is called. The warm-up runs are not timed. max_duration (float): the maximum time (in seconds) that the entire test can use. If the taken time is longer than this limit, the test is stopped and the statistics collected up to the breakpoint is reported. devices (tuple): a tuple of device IDs (int) that will be timed during the timing test. If not given, the current device is used. Returns: :class:`~cupyx.profiler._time._PerfCaseResult`: an object collecting all test results. Nz#`func` should be a callable object.z`args` should be of tuple type.z `kwargs` should be of dict type.z `n_repeat` should be an integer.z`name` should be a string.z `n_warmup` should be an integer.z)`max_duration` should be given in secondsz!`devices` should be of tuple type)r9_cupycuda get_device_idcallable ValueError isinstancetupledictintr=r?isreal_repeat)funcargskwargsn_repeatr rDrErs r benchmarkrVRsX` |}:++--/ D>>@>??? dE " "<:;;; fd # #=;<<< h $ $=;<<< dC 75666 h $ $=;<<< = & &FDEEE gu % %><===  dFHdHlG M MMrc:g}g} |D]} tj} tj| |tjj| tjjtj| #tj| wxYwt|D] } ||i| t||D]\} } tj} tj| | tj| n#tj| wxYw| g}d|D}d}t|D]M} t||D]n\} } tj} tj| | tj| V#tj| wxYwtj }||i|tj }||z }||t| |D]n\} } tj} tj| | tj| V#tj| wxYwt| |D]n\} } tj} tj| | tj| V#tj| wxYwtt|| D]F\} \}}tj||dz}|| |G|tj |z z }||krnOt!j|g|zt j}t'|||S)Ncg|]}gSrBrB).0r5s r z_repeat..s&&&&&&rrgMbP?)dtype)r)r getDevice setDevicer2rGrHstreamEventrangeziprecord synchronize_time perf_counterr1get_elapsed_timer?asarrayfloat64r)rRrSrTrUr rDrErevents_1events_2r5 prev_deviceeventdevicerr!durationt1t2cpu_timeev1ev2gpu_timers rrQrQsHH ++'))  +  a OOEJ-3355 6 6 6 OOEJ-3355 6 6 6  k * * * *G k * * * * 8__ dfXw// v'))  +  f % % % LLNNN  k * * * *G k * * * * I&&X&&&IH 8__%% 733 / /ME6!+--K /!&))) !+....!+....   ! ! df   ! !7""" 733 / /ME6!+--K /!&))) !+....!+.... 733 / /ME6!+--K /!&)))!!###!+....!+....&s8X'>'>?? * *MAzSz223<r|s******E*E*E*E*E*E*E*E*Rb5HM BUYHMHMHMHMHMVF6F6F6F6F6r