def get_test_data(seed=6789, nDocTotal=100, **kwargs):
''' Create dataset of "heldout" docs, for testing purposes.
Uses different random seed than get_data, but otherwise similar.
'''
updateKwArgsWithDefaults(kwargs)
kwargs['seed'] = seed
kwargs['nDocTotal'] = nDocTotal
Data = WordsData.CreateToyDataFromMixModel(**kwargs)
Data.name = get_short_name()
Data.summary = get_data_info()
return Data
def get_data(**kwargs):
''' Create and return dataset.
Keyword Args
-------
nDocTotal
nWordsPerDoc
'''
updateKwArgsWithDefaults(kwargs)
Data = WordsData.CreateToyDataFromMixModel(**kwargs)
Data.name = get_short_name()
Data.summary = get_data_info()
return Data
def get_data_info():
s = 'Toy Bars Data with %d true topics. Each doc uses ONE topic.' % (K)
return s
def get_data(**kwargs):
''' Create and return dataset.
Keyword Args
-------
nDocTotal
nWordsPerDoc
'''
updateKwArgsWithDefaults(kwargs)
Data = WordsData.CreateToyDataFromMixModel(**kwargs)
Data.name = get_short_name()
Data.summary = get_data_info()
return Data
def updateKwArgsWithDefaults(kwargs):
for key in Defaults:
if key not in kwargs:
kwargs[key] = Defaults[key]
if __name__ == '__main__':
import bnpy.viz.BarsViz
WData = WordsData.CreateToyDataFromMixModel(**Defaults)
bnpy.viz.BarsViz.plotExampleBarsDocs(WData)