Changes
1 changed files (+9/-5)
-
-
@@ -14,6 +14,11 @@ parser.add_argument('-i', '--input', help='training data file', default='data')parser.add_argument('-e', '--epochs', help='number of epochs to train for', default=100) parser.add_argument('-s', '--seq-size', help='sequence size', default=32) parser.add_argument('-b', '--batch-size', help='size of each training batch', default=256) parser.add_argument('-m', '--embedding-dim', help='size of the embedding', default=512) parser.add_argument('-l', '--lstm-size', help='size of the LSTM hidden state', default=512) parser.add_argument('-a', '--layers', help='number of LSTM layers', default=3) parser.add_argument('-r', '--dropout', help='how much dropout to apply', default=0.2) parser.add_argument('-n', '--max-norm', help='maximum norm for gradient clipping', default=5) args = parser.parse_args()
-
@@ -24,7 +29,7 @@ print(len(dataloader))# Prepare model model = Model(dataset, 512, 512, 3, 0.2).to(args.device) model = Model(dataset, args.embedding_dim, args.lstm_size, args.layers, args.dropout).to(args.device) print(model)
-
@@ -62,17 +67,16 @@ for t in range(args.epochs):state_c = state_c.detach() _ = torch.nn.utils.clip_grad_norm_( model.parameters(), flags.gradients_norm) model.parameters(), args.max_norm) optimizer.step() if iteration % 10 == 0: if iteration % 1 == 0: print('Epoch: {}/{}'.format(t, args.epochs), 'Iteration: {}'.format(iteration), 'Loss: {}'.format(loss_value)) if iteration % 1000 == 0: predict(args.device, net, flags.initial_words, n_vocab, vocab_to_int, int_to_vocab, top_k=3) predict(args.device, dataset, model, 100, 2) torch.save(net.state_dict(), 'checkpoint/model-{}.pth'.format(iteration))
-