tensorflow/models · error

RecomputeContext is required when force_recomputation=True.

Error message

RecomputeContext is required when force_recomputation=True.

What it means

Error "RecomputeContext is required when force_recomputation=True." thrown in tensorflow/models.

Source

Thrown at official/projects/bigbird/recomputing_dropout.py:131

      training: Python boolean indicating whether the layer should behave in
        training mode (adding dropout) or in inference mode (doing nothing).

    Returns:
      `inputs` masked according to layer configuration.

    Raises:
      ValueError: If `force_recomputation` is `True` and called outside a
        a recompute context.
    """
    if training is None:
      training = tf_keras.backend.learning_phase()

    def dropped_inputs():
      """Randomly drops elements of `inputs` when `training=True`."""
      recompute_context = recompute_grad_lib.get_recompute_context()
      if recompute_context is None:
        if self.force_recomputation:
          raise ValueError(
              'RecomputeContext is required when force_recomputation=True.')
        return tf.nn.dropout(
            inputs,
            noise_shape=self._get_noise_shape(inputs),
            seed=self.seed,
            rate=self.rate)
      seed = tf.stack([recompute_context.seed, self._recompute_seed])
      return stateless_dropout_lib.stateless_dropout(
          inputs,
          rate=self.rate,
          seed=seed,
          noise_shape=self._get_noise_shape(inputs))

    output = smart_cond(training, dropped_inputs, lambda: tf.identity(inputs))
    return output

  def compute_output_shape(self, input_shape):
    return input_shape

View on GitHub (pinned to e006f5f0d5)

When it happens

Trigger: Thrown at official/projects/bigbird/recomputing_dropout.py:131 when the library encounters an invalid state.

Common situations: See trigger scenarios.

AI-assisted analysis of tensorflow/models@e006f5f0d5 (2026-08-24). Data as JSON: /api/errors/b9a7b152925ad624. Report an issue: GitHub.