tensorflow/models · error · ValueError

unrecognized pooler type: {pooler}

Error message

unrecognized pooler type: {pooler}

What it means

Error "unrecognized pooler type: {pooler}" thrown in tensorflow/models.

Source

Thrown at official/projects/videoglue/modeling/backbones/vit_3d.py:225

        num_heads=num_heads,
        dropout_rate=dropout_rate,
        attention_dropout_rate=attention_dropout_rate,
        kernel_regularizer=kernel_regularizer,
        kernel_initializer='glorot_uniform' if original_init else dict(
            class_name='TruncatedNormal', config=dict(stddev=.02)),
        init_stochastic_depth_rate=init_stochastic_depth_rate,
        pos_embed_origin_shape=pos_embed_shape,
        pos_embed_target_shape=None,
        add_pos_embed=add_pos_embed)(x)

    if pooler == 'token':
      x = x[:, 0]
    elif pooler == 'gap':
      x = tf.reduce_mean(x, axis=1)
    elif pooler == 'none':
      x = tf.reshape(x, [-1, nt, nh, nw, x.shape[-1]], name='encoded_tokens')
    else:
      raise ValueError(f'unrecognized pooler type: {pooler}')

    if representation_size:
      x = tf_keras.layers.Dense(
          representation_size,
          kernel_regularizer=kernel_regularizer,
          name='pre_logits',
          kernel_initializer='lecun_normal' if original_init else 'he_uniform')(
              x)
      x = tf.nn.tanh(x)
    else:
      x = tf.identity(x, name='pre_logits')

    if pooler == 'none':
      endpoints = {'encoded_tokens': x}
    else:
      endpoints = {
          'pre_logits':
              tf.reshape(x, [-1, 1, 1, 1, representation_size or hidden_size])

View on GitHub (pinned to e006f5f0d5)

When it happens

Trigger: Thrown at official/projects/videoglue/modeling/backbones/vit_3d.py:225 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/4591c4cedd0251be. Report an issue: GitHub.