aramakus

## table.csv

          
            f1 score
            f1 score
            precision
            precision
            recall
            recall

            
              (Real)
              (Fake)
              (Real)
              (Fake)
              (Real)
              (Fake)

            
              ROBERTAClassifier
              0.9906
              0.9904
              0.9844
              0.9968
              0.9968
              0.9840

            
              RobertaForSequenceClassification
              0.9821
              0.9813
              0.9678
              0.9967
              0.9968
              0.9665

            
              BERTClassifier
              0.9734
              0.9726
              0.9658
              0.9805
              0.9811
              0.9649

            
              BertForSequenceClassification
              0.9689
              0.9675
              0.9541
              0.9835
              0.9842
              0.9521

## simple_model_RoBERTa.py
# Model with classifier layers on top of RoBERTa
class ROBERTAClassifier(torch.nn.Module):
    def __init__(self, dropout_rate=0.3):
        super(ROBERTAClassifier, self).__init__()

        self.roberta = RobertaModel.from_pretrained('roberta-base')
        self.d1 = torch.nn.Dropout(dropout_rate)
        self.l1 = torch.nn.Linear(768, 64)
        self.bn1 = torch.nn.LayerNorm(64)
        self.d2 = torch.nn.Dropout(dropout_rate)

## mask.py
for (source, target), _ in train_iter:
    mask = (source != PAD_INDEX).type(torch.uint8)

    y_pred = model(input_ids=source,
                   attention_mask=mask)

## report.txt
Classification Report:
              precision    recall  f1-score   support

           1     0.9844    0.9968    0.9906       634
           0     0.9968    0.9840    0.9904       626

    accuracy                         0.9905      1260
   macro avg     0.9906    0.9904    0.9905      1260
weighted avg     0.9906    0.9905    0.9905      1260

## iterators.py
# Set tokenizer hyperparameters.
MAX_SEQ_LEN = 256
BATCH_SIZE = 16
PAD_INDEX = tokenizer.convert_tokens_to_ids(tokenizer.pad_token)
UNK_INDEX = tokenizer.convert_tokens_to_ids(tokenizer.unk_token)


# Define columns to read.
label_field = Field(sequential=False, use_vocab=False, batch_first=True)
text_field = Field(use_vocab=False,

## tokenizer.py
tokenizer = RobertaTokenizer.from_pretrained("roberta-base")
	f1 score	f1 score	precision	precision	recall	recall
	(Real)	(Fake)	(Real)	(Fake)	(Real)	(Fake)
ROBERTAClassifier	0.9906	0.9904	0.9844	0.9968	0.9968	0.9840
RobertaForSequenceClassification	0.9821	0.9813	0.9678	0.9967	0.9968	0.9665
BERTClassifier	0.9734	0.9726	0.9658	0.9805	0.9811	0.9649
BertForSequenceClassification	0.9689	0.9675	0.9541	0.9835	0.9842	0.9521
	# Model with classifier layers on top of RoBERTa
	class ROBERTAClassifier(torch.nn.Module):
	def __init__(self, dropout_rate=0.3):
	super(ROBERTAClassifier, self).__init__()

	self.roberta = RobertaModel.from_pretrained('roberta-base')
	self.d1 = torch.nn.Dropout(dropout_rate)
	self.l1 = torch.nn.Linear(768, 64)
	self.bn1 = torch.nn.LayerNorm(64)
	self.d2 = torch.nn.Dropout(dropout_rate)
	for (source, target), _ in train_iter:
	mask = (source != PAD_INDEX).type(torch.uint8)

	y_pred = model(input_ids=source,
	attention_mask=mask)
	Classification Report:
	precision recall f1-score support

	1 0.9844 0.9968 0.9906 634
	0 0.9968 0.9840 0.9904 626

	accuracy 0.9905 1260
	macro avg 0.9906 0.9904 0.9905 1260
	weighted avg 0.9906 0.9905 0.9905 1260
	# Set tokenizer hyperparameters.
	MAX_SEQ_LEN = 256
	BATCH_SIZE = 16
	PAD_INDEX = tokenizer.convert_tokens_to_ids(tokenizer.pad_token)
	UNK_INDEX = tokenizer.convert_tokens_to_ids(tokenizer.unk_token)


	# Define columns to read.
	label_field = Field(sequential=False, use_vocab=False, batch_first=True)
	text_field = Field(use_vocab=False,