import torch
from classifiers.standard.bert_model import BertForSequenceClassification
from classifiers.standard.classifier import Classifier
from classifiers.util import deleteEncodingLayers
from torch import nn
from torch.nn import CrossEntropyLoss
from transformers import BertPreTrainedModel, BertModel
class BertRandomClassifier(Classifier):
def __init__(self, config):
self.config = config
self.model = BertForSequenceClassification.from_pretrained(self.config['pretrained_model'])
# We cheat the framework here - we make a new model base o
self.model = BertForSequenceClassification(self.model.config)
# here, we can do some layer removal if we want to
self.model = deleteEncodingLayers(self.model, config['num_layers'])
# setup the optimizer
self.optimizer = torch.optim.Adam(self.model.parameters(), self.config['learning_rate'])
self.epochs = 0
print(self.model)
class BertForSequenceClassification(BertPreTrainedModel):
def __init__(self, config):
super(BertForSequenceClassification, self).__init__(config)
self.num_labels = config.num_labels
self.bert = BertModel(config)
self.dropout = nn.Dropout(config.hidden_dropout_prob)
self.classifier = nn.Linear(config.hidden_size, self.config.num_labels)
self.head = nn.Softmax(dim=1)
self.init_weights()
def forward(self, input_ids=None, attention_mask=None, token_type_ids=None,
position_ids=None, head_mask=None, inputs_embeds=None, labels=None):
outputs = self.bert(input_ids,
attention_mask=attention_mask,
token_type_ids=token_type_ids,
position_ids=position_ids,
head_mask=head_mask,
inputs_embeds=inputs_embeds)
pooled_output = outputs[1]
pooled_output = self.dropout(pooled_output)
logits = self.classifier(pooled_output)
logits = self.head(logits)
outputs = (logits,) + outputs[2:] # add hidden states and attention if they are here
if labels is not None:
loss_fct = CrossEntropyLoss()
loss = loss_fct(logits.view(-1, self.num_labels), labels.view(-1))
outputs = (loss,) + outputs
return outputs # (loss), logits, (hidden_states), (attentions)
