Word Count Vectorizer#
The Word Count Vectorizer builds a vocabulary from the training samples and transforms text blobs into fixed length sparse feature vectors. Each feature column represents a word or token from the vocabulary and the value denotes the number of times that word appears in a given document.
Interfaces: Transformer, Stateful, Persistable
Data Type Compatibility: Categorical
Parameters#
# | Name | Default | Type | Description |
---|---|---|---|---|
1 | maxVocabularySize | PHP_INT_MAX | int | The maximum number of unique tokens to embed into each document vector. |
2 | minDocumentFrequency | 0.0 | float | The minimum proportion of documents a word must appear in to be added to the vocabulary. |
3 | maxDocumentFrequency | 1.0 | float | The maximum proportion of documents a word can appear in to be added to the vocabulary. |
4 | tokenizer | Word | Tokenizer | The tokenizer used to extract features from blobs of text. |
Example#
use Rubix\ML\Transformers\WordCountVectorizer;
use Rubix\ML\Tokenizers\NGram;
$transformer = new WordCountVectorizer(10000, 0.01, 0.9, new NGram(1, 2));
Additional Methods#
Return an array of words that comprise each of the vocabularies:
public vocabularies() : array
Last update:
2021-04-02