tensorflow/models · error · NotImplementedError

_instantiate_sub_task_models() is not implemented.

Error message

_instantiate_sub_task_models() is not implemented.

What it means

Error "_instantiate_sub_task_models() is not implemented." thrown in tensorflow/models.

Source

Thrown at official/modeling/multitask/base_model.py:35

import tensorflow as tf, tf_keras


class MultiTaskBaseModel(tf.Module):
  """Base class that holds multi-task model computation."""

  def __init__(self, **kwargs):
    super().__init__(**kwargs)
    self._sub_tasks = self._instantiate_sub_tasks()

  def _instantiate_sub_tasks(self) -> Dict[Text, tf_keras.Model]:
    """Abstract function that sets up the computation for each sub-task.

    Returns:
      A map from task name (as string) to a tf_keras.Model object that
        represents the sub-task in the multi-task pool.
    """
    raise NotImplementedError(
        "_instantiate_sub_task_models() is not implemented.")

  @property
  def sub_tasks(self):
    """Fetch a map of task name (string) to task model (tf_keras.Model)."""
    return self._sub_tasks

  def initialize(self):
    """Optional function that loads a pre-train checkpoint."""
    return

  def build(self):
    """Builds the networks for tasks to make sure variables are created."""
    # Try to build all sub tasks.
    for task_model in self._sub_tasks.values():
      # Assumes all the tf.Module models are built because we don't have any
      # way to check them.
      if isinstance(task_model, tf_keras.Model) and not task_model.built:

View on GitHub (pinned to e006f5f0d5)

When it happens

Trigger: Thrown at official/modeling/multitask/base_model.py:35 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/5deaedbc18366501. Report an issue: GitHub.