load_data
functiontf_keras.datasets.reuters.load_data(
path="reuters.npz",
num_words=None,
skip_top=0,
maxlen=None,
test_split=0.2,
seed=113,
start_char=1,
oov_char=2,
index_from=3,
**kwargs
)
Loads the Reuters newswire classification dataset.
This is a dataset of 11,228 newswires from Reuters, labeled over 46 topics.
This was originally generated by parsing and preprocessing the classic Reuters-21578 dataset, but the preprocessing code is no longer packaged with TF-Keras. See this GitHub discussion for more info.
Each newswire is encoded as a list of word indexes (integers). For convenience, words are indexed by overall frequency in the dataset, so that for instance the integer "3" encodes the 3rd most frequent word in the data. This allows for quick filtering operations such as: "only consider the top 10,000 most common words, but eliminate the top 20 most common words".
As a convention, "0" does not stand for a specific word, but instead is used to encode any unknown word.
Arguments
~/.keras/dataset
).num_words
most frequent words are kept. Any less frequent word
will appear as oov_char
value in the sequence data. If None,
all words are kept. Defaults to None
.oov_char
value in the dataset. 0 means no words are
skipped. Defaults to 0
.None
.0.
and 1.
. Fraction of the dataset to be
used as test data. 0.2
means that 20% of the dataset is used as
test data. Defaults to 0.2
.1
.num_words
or
skip_top
limits will be replaced with this character.Returns
(x_train, y_train), (x_test, y_test)
.x_train, x_test: lists of sequences, which are lists of indexes
(integers). If the num_words argument was specific, the maximum
possible index value is num_words - 1
. If the maxlen
argument was
specified, the largest possible sequence length is maxlen
.
y_train, y_test: lists of integer labels (1 or 0).
Note: The 'out of vocabulary' character is only used for
words that were present in the training set but are not included
because they're not making the num_words
cut here.
Words that were not seen in the training set but are in the test set
have simply been skipped.
get_word_index
functiontf_keras.datasets.reuters.get_word_index(path="reuters_word_index.json")
Retrieves a dict mapping words to their index in the Reuters dataset.
Actual word indices starts from 3, with 3 indices reserved for: 0 (padding), 1 (start), 2 (oov).
E.g. word index of 'the' is 1, but the in the actual training data, the index of 'the' will be 1 + 3 = 4. Vice versa, to translate word indices in training data back to words using this mapping, indices need to substract 3.
Arguments
~/.keras/dataset
).Returns
The word index dictionary. Keys are word strings, values are their index.