A preprocessing layer which maps text features to integer sequences.

layer_text_vectorization(
  object,
  max_tokens = NULL,
  standardize = "lower_and_strip_punctuation",
  split = "whitespace",
  ngrams = NULL,
  output_mode = "int",
  output_sequence_length = NULL,
  pad_to_max_tokens = FALSE,
  vocabulary = NULL,
  ...
)

get_vocabulary(object, include_special_tokens = TRUE)

set_vocabulary(object, vocabulary, idf_weights = NULL, ...)

Arguments

object

What to compose the new Layer instance with. Typically a Sequential model or a Tensor (e.g., as returned by layer_input()). The return value depends on object. If object is:

  • missing or NULL, the Layer instance is returned.

  • a Sequential model, the model with an additional layer is returned.

  • a Tensor, the output tensor from layer_instance(object) is returned.

max_tokens

The maximum size of the vocabulary for this layer. If NULL, there is no cap on the size of the vocabulary. Note that this vocabulary contains 1 OOV token, so the effective number of tokens is (max_tokens - 1 - (1 if output_mode == "int" else 0)).

standardize

Optional specification for standardization to apply to the input text. Values can be NULL (no standardization), "lower_and_strip_punctuation" (lowercase and remove punctuation) or a Callable. Default is "lower_and_strip_punctuation".

split

Optional specification for splitting the input text. Values can be NULL (no splitting), "whitespace" (split on ASCII whitespace), or a Callable. The default is "whitespace".

ngrams

Optional specification for ngrams to create from the possibly-split input text. Values can be NULL, an integer or list of integers; passing an integer will create ngrams up to that integer, and passing a list of integers will create ngrams for the specified values in the list. Passing NULL means that no ngrams will be created.

output_mode

Optional specification for the output of the layer. Values can be "int", "multi_hot", "count" or "tf_idf", configuring the layer as follows:

  • "int": Outputs integer indices, one integer index per split string token. When output_mode == "int", 0 is reserved for masked locations; this reduces the vocab size to max_tokens - 2 instead of max_tokens - 1.

  • "multi_hot": Outputs a single int array per batch, of either vocab_size or max_tokens size, containing 1s in all elements where the token mapped to that index exists at least once in the batch item.

  • "count": Like "multi_hot", but the int array contains a count of the number of times the token at that index appeared in the batch item.

  • "tf_idf": Like "multi_hot", but the TF-IDF algorithm is applied to find the value in each token slot. For "int" output, any shape of input and output is supported. For all other output modes, currently only rank 1 inputs (and rank 2 outputs after splitting) are supported.

output_sequence_length

Only valid in INT mode. If set, the output will have its time dimension padded or truncated to exactly output_sequence_length values, resulting in a tensor of shape (batch_size, output_sequence_length) regardless of how many tokens resulted from the splitting step. Defaults to NULL.

pad_to_max_tokens

Only valid in "multi_hot", "count", and "tf_idf" modes. If TRUE, the output will have its feature axis padded to max_tokens even if the number of unique tokens in the vocabulary is less than max_tokens, resulting in a tensor of shape (batch_size, max_tokens) regardless of vocabulary size. Defaults to FALSE.

vocabulary

Optional for layer_text_vectorization(). Either an array of strings or a string path to a text file. If passing an array, can pass an R list or character vector, 1D numpy array, or 1D tensor containing the string vocabulary terms. If passing a file path, the file should contain one line per term in the vocabulary. If vocabulary is set (either by passing layer_text_vectorization(vocabulary = ...) or by calling set_vocabulary(layer, vocabulary = ...), there is no need to adapt() the layer.

...

standard layer arguments.

include_special_tokens

If True, the returned vocabulary will include the padding and OOV tokens, and a term's index in the vocabulary will equal the term's index when calling the layer. If False, the returned vocabulary will not include any padding or OOV tokens.

idf_weights

An R vector, 1D numpy array, or 1D tensor of inverse document frequency weights with equal length to vocabulary. Must be set if output_mode is "tf_idf". Should not be set otherwise.

Details

This layer has basic options for managing text in a Keras model. It transforms a batch of strings (one example = one string) into either a list of token indices (one example = 1D tensor of integer token indices) or a dense representation (one example = 1D tensor of float values representing data about the example's tokens).

The vocabulary for the layer must be either supplied on construction or learned via adapt(). When this layer is adapted, it will analyze the dataset, determine the frequency of individual string values, and create a vocabulary from them. This vocabulary can have unlimited size or be capped, depending on the configuration options for this layer; if there are more unique values in the input than the maximum vocabulary size, the most frequent terms will be used to create the vocabulary.

The processing of each example contains the following steps:

  1. Standardize each example (usually lowercasing + punctuation stripping)

  2. Split each example into substrings (usually words)

  3. Recombine substrings into tokens (usually ngrams)

  4. Index tokens (associate a unique int value with each token)

  5. Transform each example using this index, either into a vector of ints or a dense float vector.

Some notes on passing callables to customize splitting and normalization for this layer:

  1. Any callable can be passed to this Layer, but if you want to serialize this object you should only pass functions that are registered Keras serializables (see tf$keras$utils$register_keras_serializable for more details).

  2. When using a custom callable for standardize, the data received by the callable will be exactly as passed to this layer. The callable should return a tensor of the same shape as the input.

  3. When using a custom callable for split, the data received by the callable will have the 1st dimension squeezed out - instead of matrix(c("string to split", "another string to split")), the Callable will see c("string to split", "another string to split"). The callable should return a Tensor with the first dimension containing the split tokens - in this example, we should see something like list(c("string", "to", "split"), c("another", "string", "to", "split")). This makes the callable site natively compatible with tf$strings$split().

See also