keras-team/keras · error · ValueError

Vocabulary file {vocabulary} does not exist.

Error message

Vocabulary file {vocabulary} does not exist.

What it means

set_vocabulary treats a string argument as a filesystem path to a vocabulary file. Here `tf.io.gfile.exists` reports that path as missing, so the layer cannot load it.

Source

Thrown at keras/src/layers/preprocessing/index_lookup.py:458

                f"`'tf_idf'`. Received: output_mode={self.output_mode} "
                f"and idf_weights={idf_weights}"
            )

        if isinstance(vocabulary, str):
            if serialization_lib.in_safe_mode():
                raise ValueError(
                    "Requested the loading of a vocabulary file outside of the "
                    "model archive. This carries a potential risk of loading "
                    "arbitrary and sensitive files and thus it is disallowed "
                    "by default. If you trust the source of the artifact, you "
                    "can override this error by passing `safe_mode=False` to "
                    "the loading function, or calling "
                    "`keras.config.enable_unsafe_deserialization(). "
                    f"Vocabulary file: '{vocabulary}'"
                )

            if not tf.io.gfile.exists(vocabulary):
                raise ValueError(
                    f"Vocabulary file {vocabulary} does not exist."
                )
            if self.output_mode == "tf_idf":
                raise ValueError(
                    "output_mode `'tf_idf'` does not support loading a "
                    "vocabulary from file."
                )
            self.lookup_table = self._lookup_table_from_file(vocabulary)
            self._record_vocabulary_size()
            return

        if not tf.executing_eagerly() and (
            tf.is_tensor(vocabulary) or tf.is_tensor(idf_weights)
        ):
            raise RuntimeError(
                f"Cannot set a tensor vocabulary on layer {self.name} "
                "when not executing eagerly. "
                "Create this layer or call `set_vocabulary()` "

View on GitHub (pinned to 7a34a03db6)

Solutions

1. Check the path before calling: `tf.io.gfile.exists(path)`.
2. Use paths relative to the project root, or ship the vocabulary file with the deployment artifact.
3. If you meant to pass tokens, pass a list of strings, not a single string.

Example fix

# before
layer.set_vocabulary('vocab.txt')  # a str is treated as a file path

# after
tokens = [line.strip() for line in open('vocab.txt')]
layer.set_vocabulary(tokens)

import tensorflow as tf
if isinstance(vocab, str):
    if not tf.io.gfile.exists(vocab):
        raise FileNotFoundError(vocab)
layer.set_vocabulary(vocab)

def is_vocab_file(p) -> bool:
    import tensorflow as tf
    return isinstance(p, str) and tf.io.gfile.exists(p)

try:
    layer.set_vocabulary(path)
except ValueError as e:
    raise FileNotFoundError('vocabulary file missing: %s' % path) from e

When it happens

Trigger: `layer.set_vocabulary('/data/vocab.txt')` where the path is wrong, lives on another machine, or is not mounted in the container; also hit when a bare token string is passed by accident, since a str is always treated as a path, never as one token.

Common situations: Absolute paths baked into saved models that don't exist on the loading machine; vocabulary file not packaged with the deployment.

Understand the failure class

Background: "File not found" and ENOENT errors: why libraries can't find a file that should exist — this error's family across 50 libraries.

Related errors

• Cannot set an empty vocabulary. Received: vocabulary={vocabu
• The passed vocabulary has at least one repeated term. Please
• Found reserved mask token at unexpected location in `vocabul
• Found reserved OOV token at unexpected location in `vocabula
• Attempted to set a vocabulary larger than the maximum vocab

AI-assisted analysis of keras-team/keras@7a34a03db6 (2026-08-25). Data as JSON: /api/errors/f0ade3a5b93e52f8. Report an issue: GitHub.