tensorflow/models · error · ValueError

Padding for kernel size {} not known.

Error message

Padding for kernel size {} not known.

What it means

Error "Padding for kernel size {} not known." thrown in tensorflow/models.

Source

Thrown at official/projects/deepmac_maskrcnn/modeling/heads/hourglass_network.py:55

BATCH_NORM_EPSILON = 1e-5
BATCH_NORM_MOMENTUM = 0.1
BATCH_NORM_FUSED = True


class IdentityLayer(tf_keras.layers.Layer):
  """A layer which passes through the input as it is."""

  def call(self, inputs):
    return inputs


def _get_padding_for_kernel_size(kernel_size):
  if kernel_size == 7:
    return (3, 3)
  elif kernel_size == 3:
    return (1, 1)
  else:
    raise ValueError('Padding for kernel size {} not known.'.format(
        kernel_size))


def batchnorm():
  try:
    return tf_keras.layers.experimental.SyncBatchNormalization(
        name='batchnorm', epsilon=1e-5, momentum=0.1)
  except AttributeError:
    return tf_keras.layers.BatchNormalization(
        name='batchnorm', epsilon=1e-5, momentum=0.1, fused=BATCH_NORM_FUSED)


class ConvolutionalBlock(tf_keras.layers.Layer):
  """Block that aggregates Convolution + Norm layer + ReLU."""

  def __init__(self, kernel_size, out_channels, stride=1, relu=True,
               padding='same'):
    """Initializes the Convolutional block.

View on GitHub (pinned to e006f5f0d5)

When it happens

Trigger: Thrown at official/projects/deepmac_maskrcnn/modeling/heads/hourglass_network.py:55 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/740228d5627135d2. Report an issue: GitHub.