Federated Sampling¶
From Fate v0.2 supports sample method. Sample module supports two sample modes: random sample mode and stratified sample mode.
- In random mode, "downsample" and "upsample" methods are provided. Users can set the sample parameter "fractions", which is the sample ratio within data.
- In stratified mode, "downsample" and "upsample" methods are also provided. Users can set the sample parameter "fractions" too, but it should be a list of tuples in the form (label_i, ratio). Tuples in the list each specify the sample ratio of corresponding label. e.g.
[(0, 1.5), (1, 2.5), (3, 3.5)]
Param¶
sample_param
¶
Classes¶
SampleParam (BaseParam)
¶
Define the sample method
Parameters¶
{'random', 'stratified'}'
specify sample to use, default: 'random'
{'downsample', 'upsample'}, default: 'downsample'
specify sample method
None or float or list
if mode equals to random, it should be a float number greater than 0, otherwise a list of elements of pairs like [label_i, sample_rate_i], e.g. [[0, 0.5], [1, 0.8], [2, 0.3]]. default: None
int, RandomState instance or None, default: None
random state
bool, default True
Indicate if this module needed to be run
Source code in federatedml/param/sample_param.py
class SampleParam(BaseParam):
"""
Define the sample method
Parameters
----------
mode: {'random', 'stratified'}'
specify sample to use, default: 'random'
method: {'downsample', 'upsample'}, default: 'downsample'
specify sample method
fractions: None or float or list
if mode equals to random, it should be a float number greater than 0,
otherwise a list of elements of pairs like [label_i, sample_rate_i], e.g. [[0, 0.5], [1, 0.8], [2, 0.3]]. default: None
random_state: int, RandomState instance or None, default: None
random state
need_run: bool, default True
Indicate if this module needed to be run
"""
def __init__(self, mode="random", method="downsample", fractions=None, random_state=None, task_type="hetero",
need_run=True):
self.mode = mode
self.method = method
self.fractions = fractions
self.random_state = random_state
self.task_type = task_type
self.need_run = need_run
def check(self):
descr = "sample param"
self.mode = self.check_and_change_lower(self.mode,
["random", "stratified"],
descr)
self.method = self.check_and_change_lower(self.method,
["upsample", "downsample"],
descr)
if self.mode == "stratified" and self.fractions is not None:
if not isinstance(self.fractions, list):
raise ValueError("fractions of sample param when using stratified should be list")
for ele in self.fractions:
if not isinstance(ele, collections.Container) or len(ele) != 2:
raise ValueError(
"element in fractions of sample param using stratified should be a pair like [label_i, rate_i]")
return True
__init__(self, mode='random', method='downsample', fractions=None, random_state=None, task_type='hetero', need_run=True)
special
¶Source code in federatedml/param/sample_param.py
def __init__(self, mode="random", method="downsample", fractions=None, random_state=None, task_type="hetero",
need_run=True):
self.mode = mode
self.method = method
self.fractions = fractions
self.random_state = random_state
self.task_type = task_type
self.need_run = need_run
check(self)
¶Source code in federatedml/param/sample_param.py
def check(self):
descr = "sample param"
self.mode = self.check_and_change_lower(self.mode,
["random", "stratified"],
descr)
self.method = self.check_and_change_lower(self.method,
["upsample", "downsample"],
descr)
if self.mode == "stratified" and self.fractions is not None:
if not isinstance(self.fractions, list):
raise ValueError("fractions of sample param when using stratified should be list")
for ele in self.fractions:
if not isinstance(ele, collections.Container) or len(ele) != 2:
raise ValueError(
"element in fractions of sample param using stratified should be a pair like [label_i, rate_i]")
return True