A sentiment classifier is built the only way this course has shown so far: a flat dense layer with a fixed-size input. To feed it text, each sentence is turned into a count vector — one slot per vocabulary word, holding how many times that word occurs in the sentence.
The vocabulary is five words, in this fixed slot order:
| slot | 1 | 2 | 3 | 4 | 5 |
|---|---|---|---|---|---|
| word | a | cat | chased | dog | the |
The sentence
the dog chased a cat
therefore becomes the count vector — each of the five words occurs exactly once.
Four new sentences arrive, all written with words from the same five-word vocabulary.
Which one does this model receive as a different input from the original?
Select all that apply.