raw text
"The cats
were
running."
tokenize
tokens
[The, cats,
were, running,
.]
stem
stems
[the, cat,
were, run,
.]
lemmatize
lemmas
[the, cat,
be, run,
.]
ready for the
model
vector /
embedding /
feature row /
index lookup
three operations. each has a job. each has a failure mode.