tiny_shakespeare

• Description:

40,000 lines of Shakespeare from a variety of Shakespeare's plays. Featured in Andrej Karpathy's blog post 'The Unreasonable Effectiveness of Recurrent Neural Networks': http://karpathy.github.io/2015/05/21/rnn-effectiveness/

To use for e.g. character modelling:

d = tfds.load(name='tiny_shakespeare')['train']
d = d.map(lambda x: tf.strings.unicode_split(x['text'], 'UTF-8'))
# train split includes vocabulary for other splits
vocabulary = sorted(set(next(iter(d)).numpy()))
d = d.map(lambda x: {'cur_char': x[:-1], 'next_char': x[1:]})
d = d.unbatch()
seq_len = 100
batch_size = 2
d = d.batch(seq_len)
d = d.batch(batch_size)

• Homepage: https://github.com/karpathy/char-rnn/blob/master/data/tinyshakespeare/input.txt

• Source code: tfds.datasets.tiny_shakespeare.Builder

• Versions:

  • 1.0.0 (default): No release notes.

• Download size: 1.06 MiB

• Dataset size: 1.06 MiB

• Auto-cached (documentation): Yes

• Splits:

Split	Examples
'test'	1
'train'	1
'validation'	1

• Feature structure:

FeaturesDict({
    'text': Text(shape=(), dtype=string),
})

• Feature documentation:

Feature	Class	Shape	Dtype	Description
	FeaturesDict
text	Text		string

• Supervised keys (See as_supervised doc): None

• Figure (tfds.show_examples): Not supported.

• Examples (tfds.as_dataframe):

• Citation:

@misc{
  author={Karpathy, Andrej},
  title={char-rnn},
  year={2015},
  howpublished={\url{https://github.com/karpathy/char-rnn} }
}