tensorflow/models · error · ValueError

The number of output groups must match #kernels.

Error message

The number of output groups must match #kernels.

What it means

Error "The number of output groups must match #kernels." thrown in tensorflow/models.

Source

Thrown at official/projects/mosaic/modeling/mosaic_blocks.py:78

        0.99.
      batchnorm_epsilon: A float for the epsilon value in BatchNorm. Defaults to
        0.001.
      activation: A `str` for the activation fuction type. Defaults to 'relu'.
      kernel_initializer: Kernel initializer for conv layers. Defaults to
        `glorot_uniform`.
      kernel_regularizer: Kernel regularizer for conv layers. Defaults to None.
      use_depthwise_convolution: Allows spatial pooling to be separable
        depthwise convolusions.
      **kwargs: Other keyword arguments for the layer.
    """
    super(MultiKernelGroupConvBlock, self).__init__(**kwargs)

    if output_filter_depths is None:
      output_filter_depths = [64, 64]
    if kernel_sizes is None:
      kernel_sizes = [3, 5]
    if len(output_filter_depths) != len(kernel_sizes):
      raise ValueError('The number of output groups must match #kernels.')
    self._output_filter_depths = output_filter_depths
    self._kernel_sizes = kernel_sizes
    self._num_groups = len(self._kernel_sizes)
    self._use_sync_bn = use_sync_bn
    self._batchnorm_momentum = batchnorm_momentum
    self._batchnorm_epsilon = batchnorm_epsilon
    self._activation = activation
    self._kernel_initializer = kernel_initializer
    self._kernel_regularizer = kernel_regularizer
    self._use_depthwise_convolution = use_depthwise_convolution
    # To apply BN before activation. Putting BN between conv and activation also
    # helps quantization where conv+bn+activation are fused into a single op.
    self._activation_fn = tf_utils.get_activation(activation)
    if self._use_sync_bn:
      self._bn_op = tf_keras.layers.experimental.SyncBatchNormalization
    else:
      self._bn_op = tf_keras.layers.BatchNormalization

View on GitHub (pinned to e006f5f0d5)

When it happens

Trigger: Thrown at official/projects/mosaic/modeling/mosaic_blocks.py:78 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/4552e3e9a3ae1ba5. Report an issue: GitHub.