tensorflow/models · error · ValueError

Unsupported act_type {}

Error message

Unsupported act_type {}

What it means

Error "Unsupported act_type {}" thrown in tensorflow/models.

Source

Thrown at official/projects/edgetpu/vision/modeling/heads/bifpn_head.py:40

import tensorflow as tf, tf_keras

from official.projects.edgetpu.vision.modeling import common_modules


def activation_fn(features: tf.Tensor, act_type: Text):
  """Customized non-linear activation type."""
  if act_type in ('silu', 'swish'):
    return tf.nn.swish(features)
  elif act_type == 'swish_native':
    return features * tf.sigmoid(features)
  elif act_type == 'hswish':
    return features * tf.nn.relu6(features + 3) / 6
  elif act_type == 'relu':
    return tf.nn.relu(features)
  elif act_type == 'relu6':
    return tf.nn.relu6(features)
  else:
    raise ValueError('Unsupported act_type {}'.format(act_type))


def build_batch_norm(is_training_bn: bool,
                     beta_initializer: Text = 'zeros',
                     gamma_initializer: Text = 'ones',
                     data_format: Text = 'channels_last',
                     momentum: float = 0.99,
                     epsilon: float = 1e-3,
                     strategy: Optional[Text] = None,
                     name: Text = 'tpu_batch_normalization'):
  """Builds a batch normalization layer.

  Args:
    is_training_bn: `bool` for whether the model is training.
    beta_initializer: `str`, beta initializer.
    gamma_initializer: `str`, gamma initializer.
    data_format: `str` either "channels_first" for `[batch, channels, height,
      width]` or "channels_last for `[batch, height, width, channels]`.

View on GitHub (pinned to e006f5f0d5)

When it happens

Trigger: Thrown at official/projects/edgetpu/vision/modeling/heads/bifpn_head.py:40 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/286f810fba3ee08c. Report an issue: GitHub.