|---| | [ View source on GitHub](https://github.com/tensorflow/models/blob/v2.20.0/official/nlp/modeling/models/__init__.py) | Models are combinations of [`tf.keras`](https://www.tensorflow.org/api_docs/python/tf/keras) layers and models that can be trained. Several pre-built canned models are provided to train encoder networks. These models are intended as both convenience functions and canonical examples. ## Classes [`class BertClassifier`](https://www.tensorflow.org/api_docs/python/tfm/nlp/models/BertClassifier): Classifier model based on a BERT-style transformer-based encoder. [`class BertPretrainer`](https://www.tensorflow.org/api_docs/python/tfm/nlp/models/BertPretrainer): BERT pretraining model. [`class BertPretrainerV2`](https://www.tensorflow.org/api_docs/python/tfm/nlp/models/BertPretrainerV2): BERT pretraining model V2. [`class BertSpanLabeler`](https://www.tensorflow.org/api_docs/python/tfm/nlp/models/BertSpanLabeler): Span labeler model based on a BERT-style transformer-based encoder. [`class BertTokenClassifier`](https://www.tensorflow.org/api_docs/python/tfm/nlp/models/BertTokenClassifier): Token classifier model based on a BERT-style transformer-based encoder. [`class DualEncoder`](https://www.tensorflow.org/api_docs/python/tfm/nlp/models/DualEncoder): A dual encoder model based on a transformer-based encoder. [`class ElectraPretrainer`](https://www.tensorflow.org/api_docs/python/tfm/nlp/models/ElectraPretrainer): ELECTRA network training model. [`class Seq2SeqTransformer`](https://www.tensorflow.org/api_docs/python/tfm/nlp/models/Seq2SeqTransformer): Transformer model with Keras. [`class T5Transformer`](https://www.tensorflow.org/api_docs/python/tfm/nlp/models/T5Transformer): Transformer Encoder+Decoder for sequence to sequence. [`class T5TransformerParams`](https://www.tensorflow.org/api_docs/python/tfm/nlp/models/T5TransformerParams): Transformer parameters. [`class TransformerDecoder`](https://www.tensorflow.org/api_docs/python/tfm/nlp/models/TransformerDecoder): Transformer decoder. [`class TransformerEncoder`](https://www.tensorflow.org/api_docs/python/tfm/nlp/models/TransformerEncoder): Transformer encoder. [`class XLNetClassifier`](https://www.tensorflow.org/api_docs/python/tfm/nlp/models/XLNetClassifier): Classifier model based on XLNet. [`class XLNetPretrainer`](https://www.tensorflow.org/api_docs/python/tfm/nlp/models/XLNetPretrainer): XLNet-based pretrainer. [`class XLNetSpanLabeler`](https://www.tensorflow.org/api_docs/python/tfm/nlp/models/XLNetSpanLabeler): Span labeler model based on XLNet. ## Functions [`attention_initializer(...)`](https://www.tensorflow.org/api_docs/python/tfm/nlp/models/attention_initializer): Initializer for attention layers in Seq2SeqTransformer.


| ## Other Members || |---|---| | EOS_ID | `1` |