pytorch_pfn_extras.training.extensions.ProfileReport#
- class pytorch_pfn_extras.training.extensions.ProfileReport(store_keys=None, report_keys=None, trigger=(1, 'epoch'), filename=None, append=False, format=None, **kwargs)#
Bases:
Extension
Writes the profile results to a file.
Times are reported by using the
pytorch_pfn_extras.profiler.TimeSummary.report()
context manager.- Parameters:
store_keys (iterable of strs) – Keys of values to write to the profiler report file.
report_keys (iterable of strs) – Keys of values that will be reported.
trigger (Optional[Union[Trigger, Callable[[ExtensionsManagerProtocol], bool], Tuple[float, str]]]) – Trigger that decides when to aggregate the result and output the values. This is distinct from the trigger of this extension itself. If it is a tuple in the form
<int>, 'epoch'
or<int>, 'iteration'
, it is passed toIntervalTrigger
.filename (str) – Name of the log file under the output directory. It can be a format string: the last result dictionary is passed for the formatting. For example, users can use ‘{iteration}’ to separate the log files for different iterations. If the log name is None, it does not output the log to any file.
append (bool, optionsl) – If the file is JSON Lines or YAML, contents will be appended instead of rewriting the file every call.
format (str, optional) – accepted values are ‘json’, ‘json-lines’ and ‘yaml’.
writer (writer object, optional) – must be callable. object to dump the log to. If specified, it needs to have a correct savefun defined. The writer can override the save location in the
pytorch_pfn_extras.training.ExtensionsManager
objectentries (list) – list of str
kwargs (Any) –
- Returns:
header string templates (str): template string for print values.
- Return type:
header (str)
Methods
__init__
([store_keys, report_keys, trigger, ...])finalize
(manager)Finalizes the extension.
initialize
(manager)Initializes up the manager state.
load_state_dict
(to_load)on_error
(manager, exc, tb)Handles the error raised during training before finalization.
Serializes the extension state.
Attributes
default_name
Default name of the extension.
is_async
name
needs_model_state
priority
trigger
- __call__(manager)#
Invokes the extension.
Implementations should override this operator. This method is called at iterations which the corresponding trigger accepts.
- Parameters:
manager (ExtensionsManager) – Manager object to call this operator.
- Return type:
None
- __init__(store_keys=None, report_keys=None, trigger=(1, 'epoch'), filename=None, append=False, format=None, **kwargs)#
- Parameters:
store_keys (Optional[Iterable[str]]) –
report_keys (Optional[Iterable[str]]) –
trigger (Optional[Union[Trigger, Callable[[ExtensionsManagerProtocol], bool], Tuple[float, str]]]) –
filename (Optional[str]) –
append (bool) –
format (Optional[str]) –
kwargs (Any) –
- finalize(manager)#
Finalizes the extension.
This method is called at the end of the training loop.
- Parameters:
manager (ExtensionsManagerProtocol) –
- Return type:
None
- load_state_dict(to_load)#
- Parameters:
to_load (Dict[str, Any]) –
- Return type:
None
- state_dict()#
Serializes the extension state.
It is called when a manager that owns this extension is serialized. It serializes nothing by default.
- Return type:
Dict[str, Any]