keras-team/keras · error · TypeError

Unrecognized keyword arguments: {str(kwargs)}

Error message

Unrecognized keyword arguments: {str(kwargs)}

What it means

The legacy Tokenizer __init__ accepts only a fixed set of keyword arguments; after handling the deprecated nb_words alias and popping document_count, any remaining kwargs raise TypeError listing the offending names, so misspelled or removed arguments are never silently ignored.

Source

Thrown at keras/src/legacy/preprocessing/text.py:105

        num_words=None,
        filters='!"#$%&()*+,-./:;<=>?@[\\]^_`{|}~\t\n',
        lower=True,
        split=" ",
        char_level=False,
        oov_token=None,
        analyzer=None,
        **kwargs,
    ):
        # Legacy support
        if "nb_words" in kwargs:
            warnings.warn(
                "The `nb_words` argument in `Tokenizer` "
                "has been renamed `num_words`."
            )
            num_words = kwargs.pop("nb_words")
        document_count = kwargs.pop("document_count", 0)
        if kwargs:
            raise TypeError(f"Unrecognized keyword arguments: {str(kwargs)}")

        self.word_counts = collections.OrderedDict()
        self.word_docs = collections.defaultdict(int)
        self.filters = filters
        self.split = split
        self.lower = lower
        self.num_words = num_words
        self.document_count = document_count
        self.char_level = char_level
        self.oov_token = oov_token
        self.index_docs = collections.defaultdict(int)
        self.word_index = {}
        self.index_word = {}
        self.analyzer = analyzer

    def fit_on_texts(self, texts):
        for text in texts:
            self.document_count += 1

View on GitHub (pinned to 7a34a03db6)

Solutions

1. Fix each name printed in the message against the real signature (num_words, filters, lower, split, char_level, oov_token, document_count, analyzer)
2. Replace nb_words with num_words (only that alias is auto-handled)
3. Drop arguments removed in this Keras version

Example fix

# before
Tokenizer(num_word=5000, lower=True)
# after
Tokenizer(num_words=5000, lower=True)

allowed = {'num_words','filters','lower','split','char_level','oov_token','document_count','analyzer'}
bad = set(kwargs) - allowed
if bad:
    raise TypeError(f'typo(s): {bad}')

def valid_tokenizer_kwargs(kw):
    allowed = {'num_words','filters','lower','split','char_level','oov_token','document_count','analyzer'}
    return not (set(kw) - allowed)

When it happens

Trigger: Constructing Tokenizer with a misspelled or outdated kwarg such as num_word=100, or Keras 1.x-era names other than nb_words.

Common situations: Copy-pasting Tokenizer code from old tutorials; IDE autocompletion picking the wrong name; version migrations from Keras 1.x.

Understand the failure class

Background: "Must be a positive integer", "Invalid value", "Unsupported": the invalid-argument-value error family, when a library rejects the value you pass — this error's family across 35 libraries.

Related errors

• Specify a dimension (`num_words` argument), or fit on some t
• Fit the Tokenizer on some data before using tfidf mode.
• Unknown vectorization mode:
• If using `weights="imagenet"` as true, `classes` should be 1
• The number of repeats in `EfficientNet` must be > 0. Receive

AI-assisted analysis of keras-team/keras@7a34a03db6 (2026-08-25). Data as JSON: /api/errors/2a9b3c40aa09aaef. Report an issue: GitHub.