Back to index

python3.2  3.2.2
Public Member Functions | Public Attributes | Static Public Attributes
Dict.DictCreation Class Reference
Inheritance diagram for Dict.DictCreation:
Inheritance graph
[legend]
Collaboration diagram for Dict.DictCreation:
Collaboration graph
[legend]

List of all members.

Public Member Functions

def test
def calibrate
def get_timer
def compatible
def calibrate_test
def run
def stat

Public Attributes

 last_timing
 warp
 times

Static Public Attributes

float version = 2.0
 Instance variables that each test should override.
int operations = 5
int rounds = 80000
int is_a_test = 1
 Internal variables.
tuple last_timing = (0.0, 0.0, 0.0)
int warp = 1
 calibration_runs = CALIBRATION_RUNS
 overhead_times = None
list times = []
 timer = TIMER_PLATFORM_DEFAULT

Detailed Description

Definition at line 3 of file Dict.py.


Member Function Documentation

Calibrate the test.

    This method should execute everything that is needed to
    setup and run the test - except for the actual operations
    that you intend to measure. pybench uses this method to
    measure the test implementation overhead.

Reimplemented from pybench.Test.

Definition at line 73 of file Dict.py.

00073 
00074     def calibrate(self):
00075 
00076         for i in range(self.rounds):
00077             pass

Here is the caller graph for this function:

def pybench.Test.calibrate_test (   self) [inherited]

Definition at line 262 of file pybench.py.

00262 
00263     def calibrate_test(self):
00264 
00265         if self.calibration_runs == 0:
00266             self.overhead_times = [0.0]
00267             return
00268 
00269         calibrate = self.calibrate
00270         timer = self.get_timer()
00271         calibration_loops = range(CALIBRATION_LOOPS)
00272 
00273         # Time the calibration loop overhead
00274         prep_times = []
00275         for i in range(self.calibration_runs):
00276             t = timer()
00277             for i in calibration_loops:
00278                 pass
00279             t = timer() - t
00280             prep_times.append(t / CALIBRATION_LOOPS)
00281         min_prep_time = min(prep_times)
00282         if _debug:
00283             print()
00284             print('Calib. prep time     = %.6fms' % (
00285                 min_prep_time * MILLI_SECONDS))
00286 
00287         # Time the calibration runs (doing CALIBRATION_LOOPS loops of
00288         # .calibrate() method calls each)
00289         for i in range(self.calibration_runs):
00290             t = timer()
00291             for i in calibration_loops:
00292                 calibrate()
00293             t = timer() - t
00294             self.overhead_times.append(t / CALIBRATION_LOOPS
00295                                        - min_prep_time)
00296 
00297         # Check the measured times
00298         min_overhead = min(self.overhead_times)
00299         max_overhead = max(self.overhead_times)
00300         if _debug:
00301             print('Calib. overhead time = %.6fms' % (
00302                 min_overhead * MILLI_SECONDS))
00303         if min_overhead < 0.0:
00304             raise ValueError('calibration setup did not work')
00305         if max_overhead - min_overhead > 0.1:
00306             raise ValueError(
00307                 'overhead calibration timing range too inaccurate: '
00308                 '%r - %r' % (min_overhead, max_overhead))

def pybench.Test.compatible (   self,
  other 
) [inherited]
Return 1/0 depending on whether the test is compatible
    with the other Test instance or not.

Definition at line 250 of file pybench.py.

00250 
00251     def compatible(self, other):
00252 
00253         """ Return 1/0 depending on whether the test is compatible
00254             with the other Test instance or not.
00255 
00256         """
00257         if self.version != other.version:
00258             return 0
00259         if self.rounds != other.rounds:
00260             return 0
00261         return 1

Here is the caller graph for this function:

def pybench.Test.get_timer (   self) [inherited]
Return the timer function to use for the test.

Definition at line 243 of file pybench.py.

00243 
00244     def get_timer(self):
00245 
00246         """ Return the timer function to use for the test.
00247 
00248         """
00249         return get_timer(self.timer)

Here is the caller graph for this function:

def pybench.Test.run (   self) [inherited]
Run the test in two phases: first calibrate, then
    do the actual test. Be careful to keep the calibration
    timing low w/r to the test timing.

Definition at line 309 of file pybench.py.

00309 
00310     def run(self):
00311 
00312         """ Run the test in two phases: first calibrate, then
00313             do the actual test. Be careful to keep the calibration
00314             timing low w/r to the test timing.
00315 
00316         """
00317         test = self.test
00318         timer = self.get_timer()
00319 
00320         # Get calibration
00321         min_overhead = min(self.overhead_times)
00322 
00323         # Test run
00324         t = timer()
00325         test()
00326         t = timer() - t
00327         if t < MIN_TEST_RUNTIME:
00328             raise ValueError('warp factor too high: '
00329                              'test times are < 10ms')
00330         eff_time = t - min_overhead
00331         if eff_time < 0:
00332             raise ValueError('wrong calibration')
00333         self.last_timing = (eff_time, t, min_overhead)
00334         self.times.append(eff_time)

def pybench.Test.stat (   self) [inherited]
Return test run statistics as tuple:

    (minimum run time,
     average run time,
     total run time,
     average time per operation,
     minimum overhead time)

Definition at line 357 of file pybench.py.

00357 
00358     def stat(self):
00359 
00360         """ Return test run statistics as tuple:
00361 
00362             (minimum run time,
00363              average run time,
00364              total run time,
00365              average time per operation,
00366              minimum overhead time)
00367 
00368         """
00369         runs = len(self.times)
00370         if runs == 0:
00371             return 0.0, 0.0, 0.0, 0.0
00372         min_time = min(self.times)
00373         total_time = sum(self.times)
00374         avg_time = total_time / float(runs)
00375         operation_avg = total_time / float(runs
00376                                            * self.rounds
00377                                            * self.operations)
00378         if self.overhead_times:
00379             min_overhead = min(self.overhead_times)
00380         else:
00381             min_overhead = self.last_timing[2]
00382         return min_time, avg_time, total_time, operation_avg, min_overhead

Here is the call graph for this function:

def Dict.DictCreation.test (   self)
Run the test.

    The test needs to run self.rounds executing
    self.operations number of operations each.

Reimplemented from pybench.Test.

Definition at line 9 of file Dict.py.

00009 
00010     def test(self):
00011 
00012         for i in range(self.rounds):
00013 
00014             d1 = {}
00015             d2 = {}
00016             d3 = {}
00017             d4 = {}
00018             d5 = {}
00019 
00020             d1 = {1:2,3:4,5:6}
00021             d2 = {2:3,4:5,6:7}
00022             d3 = {3:4,5:6,7:8}
00023             d4 = {4:5,6:7,8:9}
00024             d5 = {6:7,8:9,10:11}
00025 
00026             d1 = {}
00027             d2 = {}
00028             d3 = {}
00029             d4 = {}
00030             d5 = {}
00031 
00032             d1 = {1:2,3:4,5:6}
00033             d2 = {2:3,4:5,6:7}
00034             d3 = {3:4,5:6,7:8}
00035             d4 = {4:5,6:7,8:9}
00036             d5 = {6:7,8:9,10:11}
00037 
00038             d1 = {}
00039             d2 = {}
00040             d3 = {}
00041             d4 = {}
00042             d5 = {}
00043 
00044             d1 = {1:2,3:4,5:6}
00045             d2 = {2:3,4:5,6:7}
00046             d3 = {3:4,5:6,7:8}
00047             d4 = {4:5,6:7,8:9}
00048             d5 = {6:7,8:9,10:11}
00049 
00050             d1 = {}
00051             d2 = {}
00052             d3 = {}
00053             d4 = {}
00054             d5 = {}
00055 
00056             d1 = {1:2,3:4,5:6}
00057             d2 = {2:3,4:5,6:7}
00058             d3 = {3:4,5:6,7:8}
00059             d4 = {4:5,6:7,8:9}
00060             d5 = {6:7,8:9,10:11}
00061 
00062             d1 = {}
00063             d2 = {}
00064             d3 = {}
00065             d4 = {}
00066             d5 = {}
00067 
00068             d1 = {1:2,3:4,5:6}
00069             d2 = {2:3,4:5,6:7}
00070             d3 = {3:4,5:6,7:8}
00071             d4 = {4:5,6:7,8:9}
00072             d5 = {6:7,8:9,10:11}

Here is the caller graph for this function:


Member Data Documentation

pybench.Test.calibration_runs = CALIBRATION_RUNS [static, inherited]

Definition at line 206 of file pybench.py.

int pybench.Test.is_a_test = 1 [static, inherited]

Internal variables.

Definition at line 197 of file pybench.py.

tuple pybench.Test.last_timing = (0.0, 0.0, 0.0) [static, inherited]

Definition at line 200 of file pybench.py.

Definition at line 332 of file pybench.py.

Reimplemented from pybench.Test.

Definition at line 6 of file Dict.py.

pybench.Test.overhead_times = None [static, inherited]

Definition at line 209 of file pybench.py.

int Dict.DictCreation.rounds = 80000 [static]

Reimplemented from pybench.Test.

Definition at line 7 of file Dict.py.

pybench.Test.timer = TIMER_PLATFORM_DEFAULT [static, inherited]

Definition at line 215 of file pybench.py.

list pybench.Test.times = [] [static, inherited]

Definition at line 212 of file pybench.py.

pybench.Test.times [inherited]

Definition at line 234 of file pybench.py.

float Dict.DictCreation.version = 2.0 [static]

Instance variables that each test should override.

Reimplemented from pybench.Test.

Definition at line 5 of file Dict.py.

int pybench.Test.warp = 1 [static, inherited]

Definition at line 203 of file pybench.py.

pybench.Test.warp [inherited]

Definition at line 224 of file pybench.py.


The documentation for this class was generated from the following file: