model.py

import torch
import torch.nn as nn
import torchvision.models as models


class EncoderCNN(nn.Module):
    def __init__(self, embed_size):
    	"""Load the pretrained ResNet152 and replace top fc layer."""

        super(EncoderCNN, self).__init__()
        resnet = models.resnet152(pretrained=True)
        for param in resnet.parameters():
            param.requires_grad_(False)
        
        modules = list(resnet.children())[:-1]
        self.resnet = nn.Sequential(*modules)
        self.embed = nn.Linear(resnet.fc.in_features, embed_size)
        self.init_weights()

    
    def forward(self, images):
    	"""Extract the image feature vectors."""

        features = self.resnet(images)
        features = features.view(features.size(0), -1)
        features = self.embed(features)
        return features

    def init_weights(self):
    	"""Initialize the weights."""
    	self.self.linear.weight.data.normal_(0.0, 0.02)
    	self.linear.bias.data.fill_(0)

class DecoderRNN(nn.Module):
    def __init__(self, embed_size, hidden_size, vocab_size, num_layers=1):
        """
        Set the hyper-parameters and build the layers.
        Parameters
        ----------
        - embed_size  : Dimensionality of image and word embeddings
        - hidden_size : number of features in hidden state of the RNN decoder
        - vocab_size  : The size of vocabulary or output size
        - num_layers  : Number of layers
        
        """
        super(DecoderRNN, self).__init__()
        self.hidden_size = hidden_size
        
        # embedding layer that turns words into a vector of a specified size
        self.word_embeddings = nn.Embedding(vocab_size, embed_size)
        
        # The LSTM takes embedded vectors as inputs
        # and outputs hidden states of hidden_size
        self.lstm = nn.LSTM(input_size = embed_size, 
                            hidden_size = hidden_size, 
                            num_layers = num_layers,
                            batch_first = True)
        
        # the linear layer that maps the hidden state output dimension 
        self.linear = nn.Linear(hidden_size, vocab_size)
        self.init_weights()
        
    
    def forward(self, features, captions):
    	"""Extract the image feature vectors."""

        captions = captions[:,:-1] 
        embeds = self.word_embeddings(captions)
        
        # Concatenating features to embedding
        # torch.cat 3D tensors
        inputs = torch.cat((features.unsqueeze(1), embeds), 1)
        
        lstm_out, hidden = self.lstm(inputs)
        outputs = self.linear(lstm_out)
        
        return outputs
    
    def init_weights(self):
        """Initialize weights."""

        self.embed.weight.data.uniform_(-0.1, 0.1)
        self.linear.weight.data.uniform_(-0.1, 0.1)
        self.linear.bias.data.fill_(0)
        
    def sample(self, inputs, states=None, max_len=20):
        """
		Greedy search:
        Samples captions for pre-processed image tensor (inputs) 
        and returns predicted sentence (list of tensor ids of length max_len)
        """
        
        predicted_sentence = []
        
        for i in range(max_len):
            
            lstm_out, states = self.lstm(inputs, states)

            lstm_out = lstm_out.squeeze(1)
            lstm_out = lstm_out.squeeze(1)
            outputs = self.linear(lstm_out)
            
            # Get maximum probabilities
            target = outputs.max(1)[1]
            
            # Append result into predicted_sentence list
            predicted_sentence.append(target.item())
            
            # Update the input for next iteration
            inputs = self.word_embeddings(target).unsqueeze(1)
            
        return predicted_sentence