diff options
Diffstat (limited to 'tensorflow/contrib/keras/python/keras/datasets/imdb.py')
| -rw-r--r-- | tensorflow/contrib/keras/python/keras/datasets/imdb.py | 2 |
1 files changed, 1 insertions, 1 deletions
diff --git a/tensorflow/contrib/keras/python/keras/datasets/imdb.py b/tensorflow/contrib/keras/python/keras/datasets/imdb.py index bafd92aca6..5c087fe63f 100644 --- a/tensorflow/contrib/keras/python/keras/datasets/imdb.py +++ b/tensorflow/contrib/keras/python/keras/datasets/imdb.py @@ -41,7 +41,7 @@ def load_data(path='imdb.npz', num_words: max number of words to include. Words are ranked by how often they occur (in the training set) and only the most frequent words are kept - skip_top: skip the top N most frequently occuring words + skip_top: skip the top N most frequently occurring words (which may not be informative). maxlen: truncate sequences after this length. seed: random seed for sample shuffling. |