|View source on GitHub|
Classifier model based on a BERT-style transformer-based encoder.
tfnlp.models.BertClassifier( network, num_classes, initializer='glorot_uniform', dropout_rate=0.1, use_encoder_pooler=True, **kwargs )
This is an implementation of the network structure surrounding a transformer encoder as described in "BERT: Pre-training of Deep Bidirectional Transformers for Language Understanding" (https://arxiv.org/abs/1810.04805).
The BertClassifier allows a user to pass in a transformer stack, and
instantiates a classification network based on the passed
num_classes is set to 1, a regression network is instantiated.
||A transformer network. This network should output a sequence output and a classification output. Furthermore, it should expose its embedding table via a "get_embedding_table" method.|
||Number of classes to predict from the classification network.|
||The initializer (if any) to use in the classification networks. Defaults to a Glorot uniform initializer.|
||The dropout probability of the cls head.|
||Whether to use the pooler layer pre-defined inside the encoder.|
call( inputs, training=None, mask=None )
Calls the model on new inputs.
In this case
call just reapplies
all ops in the graph to the new inputs
(e.g. build a new computational graph from the provided inputs).
||A tensor or list of tensors.|
Boolean or boolean scalar tensor, indicating whether to run
||A mask or list of masks. A mask can be either a tensor or None (no mask).|
|A tensor if there is a single output, or a list of tensors if there are more than one outputs.|