toad.utils.decorator module

class toad.utils.decorator.Decorator(*args, is_class=False, **kwargs)

Bases: object

base decorater class

__init__(*args, is_class=False, **kwargs)

class toad.utils.decorator.frame_exclude(*args, is_class=False, **kwargs)

Bases: Decorator

decorator for exclude columns

class toad.utils.decorator.select_dtypes(*args, is_class=False, **kwargs)

Bases: Decorator

decorator for select frame by dtypes

class toad.utils.decorator.save_to_json(*args, is_class=False, **kwargs)

Bases: Decorator

support save result to json file

class toad.utils.decorator.load_from_json(*args, is_class=False, **kwargs)

Bases: Decorator

support load data from json file

class toad.utils.decorator.support_dataframe(*args, is_class=False, **kwargs)

Bases: Decorator

decorator for supporting dataframe

class toad.utils.decorator.proxy_docstring(*args, is_class=False, **kwargs)

Bases: Decorator

class toad.utils.decorator.support_numpy(*args, is_class=False, **kwargs)

Bases: Decorator

decorator for supporting numpy array to use torch function

class toad.utils.decorator.xgb_loss(*args, is_class=False, **kwargs)

Bases: Decorator

decorator for converting function to xgb supported loss function

Parameters

• loss_func (callable) – loss function

• **kwargs – other arguments for loss function except pred and label

Examples:

>>> @xgb_loss(**kwargs)
>>> def loss_func(pred, label, **kwargs):
>>>     ...
>>>     return loss
>>>
>>> # or use `xgb_loss` directly
>>> xgb_func = xgb_loss(**kwargs)(loss_func)
>>>
>>> # use in xgb
>>> model = xgb.XGBClassifier(objective = xgb_func)

class toad.utils.decorator.performance(*args, is_class=False, **kwargs)

Bases: Decorator

decorator for analysis code performance

Parameters

loop (int) – loop times, default 1

Examples: >>> @performance(loop = 100) >>> def func(): >>> … # code >>> return res >>> >>> func() >>> >>> # or use performance in with statement >>> with performance(): >>> … # code