tensorflow/models · error · ValueError

External states should be used with causal mode.

Error message

External states should be used with causal mode.

What it means

Error "External states should be used with causal mode." thrown in tensorflow/models.

Source

Thrown at official/projects/movinet/modeling/movinet.py:399

    self._conv_type = conv_type
    self._se_type = se_type
    self._input_specs = input_specs
    self._use_sync_bn = use_sync_bn
    self._activation = activation
    self._gating_activation = gating_activation
    self._norm_momentum = norm_momentum
    self._norm_epsilon = norm_epsilon
    self._norm = tf_keras.layers.BatchNormalization
    self._kernel_initializer = kernel_initializer
    self._kernel_regularizer = kernel_regularizer
    self._bias_regularizer = bias_regularizer
    self._stochastic_depth_drop_rate = stochastic_depth_drop_rate
    self._use_external_states = use_external_states
    self._output_states = output_states
    self._average_pooling_type = average_pooling_type

    if self._use_external_states and not self._causal:
      raise ValueError('External states should be used with causal mode.')
    if not isinstance(block_specs[0], StemSpec):
      raise ValueError(
          'Expected first spec to be StemSpec, got {}'.format(block_specs[0]))
    if not isinstance(block_specs[-1], HeadSpec):
      raise ValueError(
          'Expected final spec to be HeadSpec, got {}'.format(block_specs[-1]))
    self._head_filters = block_specs[-1].head_filters

    state_specs = None
    if use_external_states:
      self._set_dtype_policy(input_specs.dtype)
      state_specs = self.initial_state_specs(input_specs.shape)

    inputs, outputs = self._build_network(input_specs, state_specs=state_specs)

    super(Movinet, self).__init__(inputs=inputs, outputs=outputs, **kwargs)

    self._state_specs = state_specs

View on GitHub (pinned to e006f5f0d5)

When it happens

Trigger: Thrown at official/projects/movinet/modeling/movinet.py:399 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/9bb0b8d40c41aa64. Report an issue: GitHub.