tensorflow/models · error · ValueError
if `sequence_output` is not in encoder output, `latent_outpu
Error message
if `sequence_output` is not in encoder output, `latent_output` must be in encoder output anddecoder must exist.
What it means
Error "if `sequence_output` is not in encoder output, `latent_output` must be in encoder output anddecoder must exist." thrown in tensorflow/models.
Source
Thrown at official/projects/perceiver/modeling/models/classifier.py:126
self.num_classes = num_classes
self.head_name = head_name
self.initializer = initializer
self._decoder = decoder
self._network = network
inputs = self._network.inputs
outputs = self._network(inputs)
if 'sequence_output' not in outputs:
if 'latent_output' in outputs and self._decoder is not None:
decoder_inputs = {
'latent_output': outputs['latent_output'],
'input_mask': inputs['input_mask'],
}
decoder_outputs = self._decoder(decoder_inputs)
sequence_output = decoder_outputs['sequence_output']
else:
raise ValueError('if `sequence_output` is not in encoder output, '
'`latent_output` must be in encoder output and'
'decoder must exist.')
else:
sequence_output = outputs['sequence_output']
cls_inputs = sequence_output
if initializer is None:
stddev = 1. / np.sqrt(cls_inputs.shape[-1])
initializer = tf_keras.initializers.TruncatedNormal(stddev=stddev)
if cls_head:
classifier = cls_head
else:
classifier = layers.ClassificationHead(
inner_dim=cls_inputs.shape[-1],
num_classes=num_classes,
initializer=initializer,View on GitHub (pinned to e006f5f0d5)
When it happens
Trigger: Thrown at official/projects/perceiver/modeling/models/classifier.py:126 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/a12fe940b1fb185f.
Report an issue: GitHub.