chainer.function_hooks.TimerHook

class chainer.function_hooks.TimerHook

Function hook for measuring elapsed time of functions.

Example

Code example:

from chainer.function_hooks import TimerHook
hook = TimerHook()
with hook:
    trainer.run()
hook.print_report()

Output example:

       FunctionName  ElapsedTime  Occurrence
     LinearFunction      1.24sec        3900
               ReLU      0.59sec        2600
SoftmaxCrossEntropy      0.82sec        1300
           Accuracy      0.18sec         700

where FunctionName is the name of function that calls the hook, and ElapsedTime is the elapsed time the function consumed, and Occurrence is the number of calls.

Variables

~TimerHook.call_history – List of measurement results. It consists of pairs of the name of the function that calls this hook and the elapsed time the function consumes.

Methods

__enter__()

__exit__(*_)

added(function)

Callback function invoked when the function hook is registered

Parameters

function (FunctionNode) – Function object to which the function hook is added. None if the function hook is registered globally.

backward_postprocess(function, in_data, out_grad)

Callback function invoked after backward propagation.

Parameters

• function (FunctionNode) – Function object to which the function hook is registered.

• in_data (tuple of N-dimensional array) – Input of forward propagation.

• out_grad (tuple of N-dimensional array) – Gradient data of backward propagation.

backward_preprocess(function, in_data, out_grad)

Callback function invoked before backward propagation.

Parameters

• function (FunctionNode) – Function object to which the function hook is registered.

• in_data (tuple of N-dimensional array) – Input data of forward propagation.

• out_grad (tuple of N-dimensional array) – Gradient data of backward propagation.

deleted(function)

Callback function invoked when the function hook is unregistered

Parameters

function (FunctionNode) – Function object from which the function hook is deleted. None if the function hook was registered globally.

forward_postprocess(function, in_data)

Callback function invoked after forward propagation.

Parameters

• function (FunctionNode) – Function object to which the function hook is registered.

• in_data (tuple of N-dimensional array) – Input data of forward propagation.

forward_preprocess(function, in_data)

Callback function invoked before forward propagation.

Parameters

• function (FunctionNode) – Function object to which the function hook is registered.

• in_data (tuple of N-dimensional array) – Input data of forward propagation.

print_report(unit='auto', file=<_io.TextIOWrapper name='<stdout>' mode='w' encoding='UTF-8'>)

Prints a summary report of time profiling in functions.

Parameters

unit (str) – Supplementary units used for computational times. sec, ms, us, ns, auto`(default) and `auto_foreach are supported. If auto, units of times are aligned to the largest, and if auto_foreach, units of times are adjusted for each element.

summary()

Returns a summary of time profiling in functions.

Returns

A summarized dictionary whose keys are function names and values are dictionaries of elapsed_time and occurrence.

total_time()

Returns total elapsed time in seconds.

__eq__(value, /)

Return self==value.

__ne__(value, /)

Return self!=value.

__lt__(value, /)

Return self<value.

__le__(value, /)

Return self<=value.

__gt__(value, /)

Return self>value.

__ge__(value, /)

Return self>=value.

Attributes

name = 'TimerHook'

table = {'ms': 1000, 'ns': 1000000000, 'sec': 1, 'us': 1000000}