tff.simulation.datasets.stackoverflow.load_word_counts

Loads the word counts for the Stack Overflow dataset.

tff.simulation.datasets.stackoverflow.load_word_counts(
    cache_dir=None, vocab_size: Optional[int] = None
)

Args
cache_dir	(Optional) directory to cache the downloaded file. If None, caches in Keras' default cache directory.
vocab_size	(Optional) when specified, only load the first vocab_size unique words in the vocab file (i.e. the most frequent vocab_size words).

Returns
A collections.OrderedDict where the keys are string tokens, and the values are the counts of unique users who have at least one example in the training set containing that token in the body text.

Raises
TypeError if vocab_size is not None or int. ValueError if vocab_size is not None but <= 0.