Function train

13_2_rnn_classification.py:136–157

()

# Train cycle
def train():
    total_loss = 0

    for i, (names, countries) in enumerate(train_loader, 1):
        input, seq_lengths, target = make_variables(names, countries)
        output = classifier(input, seq_lengths)

        loss = criterion(output, target)
        total_loss += loss.data[0]

        classifier.zero_grad()
        loss.backward()
        optimizer.step()

        if i % 10 == 0:
            print('[{}] Train Epoch: {} [{}/{} ({:.0f}%)]\tLoss: {:.2f}'.format(
                time_since(start), epoch,  i *
                len(names), len(train_loader.dataset),
                100. * i * len(names) / len(train_loader.dataset),
                total_loss / i * len(names)))

    return total_loss


# Testing cycle