Fake_News_Detection/LiarLiar.py at main · kaustubh1955/Fake_News_Detection · GitHub

1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
61
62
63
64
65
66
67
68
69
70
71
72
73
74
75
76
77
import torch
import torch.nn as nn

from Attention import MultiHeadAttention, PositionFeedforward
from Encoder import Encoder

class arePantsonFire(nn.Module):

    def __init__(self, sentence_encoder: Encoder, explanation_encoder: Encoder, multihead_Attention: MultiHeadAttention,
                 position_Feedforward: PositionFeedforward, hidden_dim: int, max_length_sentence,
                 max_length_justification, input_dim, device='cpu'):
        """
        If you wish to shift on cpu pass device as 'cpu'
        """

        super(arePantsonFire, self).__init__()
        self.device = device

        self.sentence_pos_embedding = nn.Embedding(max_length_sentence, hidden_dim)
        self.justification_pos_embedding = nn.Embedding(max_length_justification, hidden_dim)

        self.sentence_encoder = sentence_encoder
        self.explanation_encoder = explanation_encoder
        self.attention = multihead_Attention
        self.position_feedforward = position_Feedforward

        self.upscale_conv, self.first_conv, self.flatten_conv = self.get_convolutions(input_dim=input_dim, hidden_dim=hidden_dim)
        self.linear1, self.linear2, self.bilinear, self.classifier = self.get_linears_layers(max_length_sentence=max_length_sentence)

    def forward(self, sentence, justification, credit_history): # 1 Marks

        sentence_pos = torch.arange(0, sentence.shape[2]).unsqueeze(0).repeat(sentence.shape[0],1).to(self.device).long()
        justification_pos = torch.arange(0, justification.shape[2]).unsqueeze(0).repeat(justification.shape[0], 1).to(self.device).long()

        sentence = self.upscale_conv(sentence)
        sentence = sentence + self.sentence_pos_embedding(sentence_pos).permute(0, 2, 1)

        justification = self.upscale_conv(justification)
        justification = justification + self.justification_pos_embedding(justification_pos).permute(0, 2, 1)

        encoded_sentence = self.sentence_encoder(sentence)
        encoded_justification = self.explanation_encoder(justification)

        attention_output = self.attention(encoded_sentence, encoded_justification,encoded_justification)
        attention_output = self.position_feedforward(attention_output)

        attention_output = self.first_conv(attention_output)
        rlu = nn.ReLU()
        attention_output = rlu(attention_output)

        attention_output = self.flatten_conv(attention_output)
        fltn = nn.Flatten()
        attention_output = fltn(attention_output)

        attention_output = self.linear1(attention_output)
        attention_output = self.linear2(attention_output)
        attention_output = self.bilinear(attention_output, credit_history)


        ans = self.classifier(attention_output)
        return ans
        # Your code goes here

    def get_convolutions(self, input_dim, hidden_dim): # 0.5 Marks
        upsacle_conv = nn.Conv1d(in_channels=input_dim,out_channels=hidden_dim,kernel_size=1,padding=0, stride=1)
        first_cov = nn.Conv1d(in_channels=hidden_dim,out_channels=hidden_dim//2,kernel_size=3,padding=1, stride=1)
        flatten_conv = nn.Conv1d(in_channels=hidden_dim//2,out_channels=1,kernel_size=5,padding=1, stride=1)
        return upsacle_conv,first_cov,flatten_conv;
    # Your code here

    def get_linears_layers(self, max_length_sentence): # 0.5 Marks
    # Your code goes here
        linear1 = nn.Linear(max_length_sentence,max_length_sentence//4)
        linear2 = nn.Linear(max_length_sentence//4, 6)
        bilinear =  nn.Bilinear(6, 5, 12, bias=True)
        classifier = nn.Linear(12, 6)
        return linear1,linear2, bilinear, classifier