tensorflow/models · error · RuntimeError

Weight and label tensors were not of the same rank. weights.

Error message

Weight and label tensors were not of the same rank. weights.shape was %s, and labels.shape was %s.

What it means

Error "Weight and label tensors were not of the same rank. weights.shape was %s, and labels.shape was %s." thrown in tensorflow/models.

Source

Thrown at official/nlp/modeling/losses/weighted_sparse_categorical_crossentropy.py:30

# See the License for the specific language governing permissions and
# limitations under the License.

"""Weighted sparse categorical cross-entropy losses."""

import tensorflow as tf, tf_keras


def _adjust_labels(labels, predictions):
  """Adjust the 'labels' tensor by squeezing it if needed."""
  labels = tf.cast(labels, tf.int32)
  if len(predictions.shape) == len(labels.shape):
    labels = tf.squeeze(labels, [-1])
  return labels, predictions


def _validate_rank(labels, predictions, weights):
  if weights is not None and len(weights.shape) != len(labels.shape):
    raise RuntimeError(
        ("Weight and label tensors were not of the same rank. weights.shape "
         "was %s, and labels.shape was %s.") %
        (predictions.shape, labels.shape))
  if (len(predictions.shape) - 1) != len(labels.shape):
    raise RuntimeError(
        ("Weighted sparse categorical crossentropy expects `labels` to have a "
         "rank of one less than `predictions`. labels.shape was %s, and "
         "predictions.shape was %s.") % (labels.shape, predictions.shape))


def loss(labels, predictions, weights=None, from_logits=False):
  """Calculate a per-batch sparse categorical crossentropy loss.

  This loss function assumes that the predictions are post-softmax.
  Args:
    labels: The labels to evaluate against. Should be a set of integer indices
      ranging from 0 to (vocab_size-1).
    predictions: The network predictions. Should have softmax already applied.

View on GitHub (pinned to e006f5f0d5)

When it happens

Trigger: Thrown at official/nlp/modeling/losses/weighted_sparse_categorical_crossentropy.py:30 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/1f43b73737ff8370. Report an issue: GitHub.