Grammer in readme.

This commit is contained in:
Joshua Eckroth 2013-08-06 12:16:03 -04:00
parent 914b65a5dc
commit 5d7faa2b22

View file

@ -344,7 +344,7 @@ Words appearing in the dataset will only be those appearing in the
documents (or a subset; by default, the most common 1000 words). This
presents a problem when new documents are loaded and used in a
classifier trained on other documents. The classifier will not know
how to handle word attributes that are not present in the training
how to handle word attributes that were not present in the training
set.
The `docs-to-dataset` function provides the ability to save the