Hey everyone, I have a Keras LSTM code, that I want to transfer in LSTM in PyTorch for Music Generation. This is my code for Keras LSTM -
def built_model(batch_size, seq_length, unique_chars): model = Sequential() model.add(Embedding(input_dim = unique_chars, output_dim = 512, batch_input_shape = (batch_size, seq_length), name = "embd_1")) model.add(LSTM(512, return_sequences = True, stateful = True, name = "lstm_first")) model.add(Dropout(0.4, name = "drp_1")) model.add(LSTM(512, return_sequences = True, stateful = True)) model.add(Dropout(0.4)) model.add(LSTM(512, return_sequences = True, stateful = True)) model.add(Dropout(0.2)) model.add(TimeDistributed(Dense(unique_chars))) model.add(Activation("softmax")) return model
And is the one that I am trying to do in PyTorch from my Sentiment Analysis model-
import torch.nn as nn class SentimentRNN(nn.Module): """ The RNN model that will be used to perform Sentiment analysis. """ def __init__(self, batch_size, seq_length, unique_chars, embedding_dim, output_size, hidden_dim, n_layers, drop_prob=0.5): """ Initialize the model by setting up the layers. """ super(SentimentRNN, self).__init__() self.output_size = output_size self.n_layers = n_layers self.hidden_dim = hidden_dim # embedding and LSTM layers self.embedding = nn.Embedding(unique_chars, embedding_dim) self.lstm = nn.LSTM(embedding_dim, hidden_dim, n_layers, dropout=drop_prob, batch_first=True) # dropout layer self.dropout = nn.Dropout(0.3) # linear and sigmoid layer self.fc = nn.Linear(hidden_dim, unique_chars) self.sig = nn.Softmax() def forward(self, x, hidden): """ Perform a forward pass of our model on some input and hidden state. """ batch_size = x.size(0) # embeddings and lstm_out embeds = self.embedding(x) lstm_out, hidden = self.lstm(embeds, hidden) # stack up lstm outputs lstm_out = lstm_out.contiguous().view(-1, self.hidden_dim) # dropout and fully connected layer out = self.dropout(lstm_out) out = self.fc(out) # sigmoid function sig_out = self.sig(out) # reshape to be batch_size first sig_out = sig_out.view(batch_size, -1) sig_out = sig_out[:, -1] # get last batch of labels # return last sigmoid output and hidden state return sig_out, hidden def init_hidden(self, batch_size): ''' Initializes hidden state ''' # Create two new tensors with sizes n_layers x batch_size x hidden_dim, # initialized to zero, for hidden state and cell state of LSTM weight = next(self.parameters()).data if(train_on_gpu): hidden = (weight.new(self.n_layers, batch_size, self.hidden_dim).zero_().cuda(), weight.new(self.n_layers, batch_size, self.hidden_dim).zero_().cuda()) else: hidden = (weight.new(self.n_layers, batch_size, self.hidden_dim).zero_(), weight.new(self.n_layers, batch_size, self.hidden_dim).zero_()) return hidden
I am trying to do this on my own. But I am clueless, where to start, can anyone please help to find one? Thanks.